Galton, M. (1990) - Grouping and Group Work.

The Social Psychology of the
Primary School
This book offers an analysis of the ways in which the process and
structure of classroom life affect the interpersonal and academic
outcomes of schooling. Social psychology is seen to have a crucial role to
play in achieving effective teaching and meaningful learning, while
promoting other useful developments in the primary classroom.
The authors study classroom interaction and relationships and
consider how these might be structured for the best outcomes. With so
much attention being focused recently on the National Curriculum, this
book seeks to provide a balance for the current curricula-oriented view of
teaching by attempting to improve understanding of how curricula are
implemented in the classroom. There are contributions on the theory of
social psychology in the primary education process as well as its
practical application. Thus, motivation and the social development of
primary age children are covered as well as relationships and social
interaction in the classroom, gender, and special education needs.
The editors and contributors offer an extensive body of knowledge
and experience of educational psychology and research, initial teacher
training and INSET, and schools and schooling.
The Social Psychology of
the Primary School
Collaboratively edited by
Colin Rogers and
Peter Kutnick
London and New York

First published 1990
by Routledge
11 New Fetter Lane, London EC4P 4EE
This edition published in the Taylor & Francis e-Library, 2002.

Simultaneously published in the USA and Canada
by Routledge
29 West 35th Street, New York, NY 10001
© 1990 Colin Rogers and Peter Kutnick
All rights reserved. No part of this book may be reprinted or reproduced or utilized in any
form or by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.
British Library Cataloguing in Publication Data

The social psychology of the primary school.
1. Great Britain. Primary schools. Students. Socialisation
I. Rogers, Colin II. Kutnick Peter
372.181
ISBN 0-415-07197-6 (Print Edition)
Library of Congress Cataloging in Publication Data

The social psychology of the primary school/edited by Colin Rogers and Peter Kutnick.
p. cm.
Includes bibliographical references.
ISBN 0-415-07197-6 (Print Edition)
1. Education, Elementary—Social aspects. 2. Social psychology. 3. Interaction
analysis in education. 4. Classroom environment.
I. Rogers, Colin, 1950– . II. Kutnick, Peter.
LC192.3.S57 1990
372–dc20 90–32016
CIP
ISBN 0-203-19383-0 Master e-book ISBN

ISBN 0-203-19386-5 (Glassbook Format)
This volume is dedicated to our children
Lauren and Hazel
Sam and Laura
Contents
Figures and tables ix

Contributors x
Acknowledgements xi
1 Process and structure in the primary school
Peter Kutnick and Colin Rogers 1
2 Grouping and group work
Maurice Galton 11
3 Social interaction in the classroom
Jane French 31
4 Classroom discourse and classroom knowledge
Derek Edwards 49
5 Self-concept and school achievement
Dale H.Schunk 70
6 Motivation in the primary years
Colin Rogers 92
7 Social development of the child and the promotion of
autonomy in the classroom
Peter Kutnick 110
8 Towards an adequate conception of early moral
development
Derek Wright 131
9 The role of play in the nursery and
primary school curriculum
Peter K.Smith 144
10 The nature of friendship in the primary school
William Maxwell 169
vii
Contents
11 Sex roles in the primary classroom

Paul Croll and Diana Moses 190
12 Interaction with children with special educational needs
David Galloway 211
13 Co-operative learning
Robert E.Slavin 226
14 Individuals, groups and interventions
Colin Rogers and Peter Kutnick 247
Author index 254
Subject index 258
viii
Figures and tables
Figures
Figure 5.1 Self-efficacy model of cognitive skill learning 77
Figure 7.1 Re-interpretation of Piaget’s stages of moral
development as a stair-like matrix 120
Figure 9.1 A four-fold scheme for conceptualizing teacher
attitudes to play 150
Figure 9.2 Typical design of experimental studies of the effects
of play experience on problem solving 155
Figure 9.3 Typical design of a study investigating the effects of
fantasy play training 157
Figure 9.4 Plan of a study comparing the effects of play
tutoring and skills tutoring 158
Tables
Table 6.1 Brief definitions of categories used in the analysis of
children’s responses 104
Table 7.1 Development of social relationships with logical-
mathematical, moral and social perspective
(development) parallels 113
ix
Contributors
Paul Croll, Department of Education, Bristol

Polytechnic.
Derek Edwards, Department of Social Sciences,
Loughborough University.
Jane French, Department of Education, University of
York.
David Galloway, Department of Educational Research,
Cartmel College, University of Lancaster.
Maurice Galton, School of Education, University of
Leicester.
Peter Kutnick, Institute of Continuing and Professional
Education, University of Sussex.
William Maxwell, Grampian Regional Council Education
Department, Aberdeen, Scotland
Diana Moses, Department of Education, Bristol
Polytechnic.
Colin Rogers, Department of Educational Research,
Cartmel College, University of Lancaster.
Dale H.Schunk, School of Education, University of North
Carolina, USA.
Robert E.Slavin, Centre for Research on Elementary and
Middle Schools, The Johns Hopkins
University, Baltimore, USA.
Peter K.Smith, Department of Psychology, University of
Sheffield.
x
Acknowledgements
Irrespective of the order in which our names have finally appeared on the
cover of this volume, the editorial work represented here is the result of a
collaborative enterprise. We accept equal responsibility and (if any is
forthcoming) equal credit for this work.
We take this opportunity to thank all the numerous colleagues and
friends whose help has been invaluable during the production of this
book. We are naturally particularly indebted to the contributors of the
original work contained here and thankful for their promptness in
dealing with our numerous requests.
Peter Kutnick
Colin Rogers
xi
Chapter one
Process and structure in the primary

school
Peter Kutnick and Colin Rogers
Introduction
In this edited book we bring together a number of concerns and interests
which will be relevant to teachers and researchers of the primary school.
The book is a product of current research and practice in the primary
school, and many of the themes pursued here could not have been
undertaken until this point in time. The book acknowledges that
classroom practice has come a long way from the traditional teaching
strategies criticized in the Plowden Report (1967). We rarely find exces-
sively formal classrooms which have been linked to stratified results
(such as discrimination by social class, gender and stature of primary
school). Yet current studies show that the move to action-oriented and
child-centred classrooms has not (unambiguously) led to expected
successful results (of equal educational enhancement for all pupils). In
analysing and explaining this ambiguity, discussion has shifted to
classroom processes which may affect the quality of the educational
product.
In relating social psychology to educational practice in the primary
school we are mindful to take a ‘balanced view’. The balance draws upon
descriptions of educational practice, social psychological theory,
analytic tools and processes in social psychology, and the malleability of
the classroom (the ability to experiment and change practice). The timing
and presentation of this book will be apparent as it draws upon recent
studies of classroom practice and educational concerns in Britain. It
cannot present all practice and all theory, but focuses on up-to-date
concerns of grouping, social interaction, classroom talk, motivation and
development of the child. Social psychological issues are chosen to
represent classroom processes in a social context; even though the
product of schooling is usually presented in terms of the individual pupil,
that product cannot be divorced from the social processes of the
classroom.
The current educational climate dictated by the Educational Reform
Act (1988) establishes a national curriculum and diagnostic testing of
1
individual attainment of that curriculum. We fear that teachers may

become overwhelmed with concerns of integrating and establishing the
new curriculum. But the curriculum can only be applied through
processes of teaching and learning. While the Education Reform Act has
a distinct product in mind, teachers will still be reliant on grouping
strategies and patterns of teacher-pupil and peer interactions to achieve
the curricular products. Classroom processes are common to all of the
separate curricula being pursued in the Act.
In the past we would have referred to processes as blending the overt
and hidden curriculum of the classroom which affect intellectual and
social outcomes of the classroom. Now we refer more specifically to
types of task demands within the curriculum (roughly identified in
research by Bennett, et al. 1984 and others as cognitive enhancement,
practice, application and so on) and the range of classroom groupings
that may enhance or hinder self-concept and learning. We wish to
balance the effects of ‘curriculum-led’ schooling by understanding
effective processes to support curricular innovations. We do not wish to
see a repeat of classroom problems of management/control which have
hindered teachers’ matching of individual pupil’s educational need to
specific levels of developmental curricula, limited effectiveness of
groupings in class, the association of low motivation and self-esteem
with failure, and so on.
Each of the following chapters presents a general or specific focus
on processes in the primary classroom. As a totality, the book also
represents a reassessment of the role of social psychology in the
classroom. In recent years social psychology of education concerns
have been expressed in selective studies and chapters. We must look
back to the 1970s to find coherent volumes published in Britain
concerning social psychology and schooling (see Morrison and
McIntyre 1969, 1971, 1972). Since that time certain issues as the
relation of theory to practice have remained. New concerns have also
been introduced, concerns such as consideration of the developing
abilities of the child in the classroom (including intellectual
development and social sensitivity), academic attainments promoted or
hindered as related to self-concept, and the role of social context in
inducing type and quality of classroom understanding. There is now a
greater body of classroom descriptions (generally atheoretical) telling
us about activities related to outcomes in the classroom. Classroom
descriptions (found especially in studies undertaken by the ORACLE
team as presented by Galton— Chapter 2 in this volume) point out
problems and possibilities of social interaction, groupings of pupil and
teacher strategies. The descriptions provide useful backgrounds upon
which more theoretical analyses can be pursued and explained. In a
repeat of history, some of the concerns pursued here were first
2
Primary school process and structure
introduced in the 1920s and 1930s. At that time educators and

psychologists pursued studies into collective (and co-operative)
learning, social bases to cognitive development, social discourse
between child and teacher, and the relation of context and classroom
structure to learning. Expositions in this book draw upon some of those
preliminary studies, but bring current perspectives to bear.
With current classroom descriptions and advances/recapitulations of
theory, we become aware of the variety of developments and potential
problems that may occur in the classroom. Awareness and analyses of
classroom processes can occur at the broad macro level and the narrow
micro level; each involves separate methodologies for study and these
will be commented upon throughout the text From these analyses four
separate themes recur: a questioning of whether the real focus of
development in the classroom should be on the individual or the social
group; how the context of the classroom and the various types of
interactors (teacher, pupil, peers) affect development; that classroom
actions, whether planned or simply allowed to occur (through imposition
of school ethos, and so on) in the classroom context, are always
‘structured’ and this structure can be linked to specific products; and that
the teacher has a vital role to play in both structuring context and
extending her/his accountability to include process as well as product.
The four themes will be qualified by childrens’ general development in
the classroom. And development may be accounted for comparatively
over the years of the primary school as well as within individual and
group interactions in any particular year group.
Practice and theory

As social scientists intimately involved with education we are very
conscious that practice and theory do not always fit well together.
There have been a number of classic examples in which theory has
dominated practice, leading, at times, to suspect results. Theory may
work very well in small-scale experimental situations, but has limited
transferability to the real situation of the classroom. On the other hand,
there are many good descriptions of classroom activity which show
effectiveness (or lack of effectiveness) of practice. Without
theoretically induced tools of analysis or means of explanation, these
descriptions may only be left at the here-and-now level; neither
explaining the why’s of effective practice nor offering alternative
strategies to change practice. Some classroom descriptions actually
imply a theoretical critique; hence there are examples which: qualify
the (often noted) unequal power relationship between child and teacher
as displaying genuine negotiation; note that the actual social climate of
the classroom (whether competitive or co-operative, and so on) affects
3
quality of learning and control; show that the term ‘autonomy’ is used
variously by teachers, classroom observers and theoreticians. Thus
descriptions tell us of the power relationships between teacher and
pupils, the comparative role of peers for the pupil, and the laissez-faire
versus structured organization of classroom interaction by the teacher.
These descriptions also assert that the teacher is in the focal role of:
defining how current and innovatory curriculum is undertaken in
classroom process; planning for young childrens’ potential movement
from early dependence to establishing self-hood and autonomy; being
the reflective exemplar of classroom behaviour.
The perspective presented here does not deny the central position of
control into which teachers are often cast. Rather, we place greater
emphasis on potential problems inherent in today’s classrooms and use
both descriptive and theoretical evidence to expand our understanding
and provide future directions for classroom behaviour. We wish to see
that the movement away from traditional practices with discriminating
results remains a movement away, and not allow ourselves to be
persuaded that this ‘progressive’ movement has failed in its educational
products (as implied in recent announcements by the Secretary of State
for Education. Indeed, from comparative information based on HMI
reports (especially DES 1978) we should be aware that classrooms have
lost their traditional physical layout of desks in rows, incorporated
various methods of grouping pupils, maintained most teaching in the
‘core’ curriculum, and upheld academic standards). We do not wish to
see the national curriculum and testing obligations of schools and
teachers force a retreat in classroom process to pre-existing practice.
Given our value-laden approach, the relation of theory and practice in
this book is assuredly a relative one. In one sense, the theories which we
cite have been ‘grounded’ and developed through observation of social
practices (invariably of the pupil in the classroom). The theories drawn
upon are also useful tools for analysis of practice and suggest alternatives
to existing practice. Theory, except possibly in the chapter by Slavin
(Chapter 13), does not dominate our perspective on future directions for
practice. Even Slavin’s chapter is derived from a critique of ineffective
past classroom practices in the United States, and grounded experiences
of effective working groups, to establish a model of co-operative
learning. Thus our theoretical interests are bounded by practical
constraints of existing classrooms from which models of current and
projected practices can be analysed and recommended.
Accountability and the reflective teacher

Over the last decade teachers have been under an onslaught of cries for
accountability which have culminated in the imposition of the Education
4
Reform Act. Accountability is undoubtedly of benefit to education

generally and to classroom practice. Within the perspective of
accountability teachers are frequently requested to explain the
performance of their class and individual pupils to school administrators,
parents and government. While we fully support this perspective, we also
note that what is brought to account is generally the end-product of pupil
performance. Generally, pupil product in the classroom is a limited unit
of analysis. It does not allow for explanation in terms of a range of
performances in the classroom, and does not focus on classroom process
(which may be more or less effectively structured to enhance
performance for all pupils).
A genuine concern of this book is to promote movement away from
simple and end-product definitions by which teachers are held to
account. We do not wish to see the curriculum (and diagnostic testing)
‘tail wagging the process dog’. We know that a reliance on traditional
means of accounting causes pressure to be placed on the teacher to use
traditional means of classroom process. While traditional processes do
provide consistent end-products of educational attainment, we are aware
that the traditional product has been highly discriminatory and stratified
(as explained in social class and gender analyses of ‘11+’ results and
limited entry into grammar schools of children of the working class and
girls of high ability). To move away from traditional accountability
pressures, products in education have to be tied to the processes that
generate those results. The teacher must be in a position to know how to
relate to pupils, understand their growth and development, and structure
the experiences that are being offered in classroom. The teacher must
reflect on classroom process as it relates to product.
In taking on this reflective role (which numerous practitioners have
already adopted), there are a number of aspects for teachers to consider
which are explored here. The aspects include the need to structure pupil
groupings for particular classroom learning tasks, positive ways to
instil classroom motivation and autonomy in pupils, the role of play
and friendships in classroom planning, relationships between
classroom actors (teacher and pupils), and tools with which to analyse
the on-going processes of the classroom. All of these aspects are
premissed on a movement away from the individual as the unit of
teaching and learning and the assertion that the classroom is the social
context of meaning and development for the pupil. There are a number
of aspects of classroom practice which have been analysed previously,
and which we do not need to repeat here. We are already aware that:
traditional classroom process and products have been explored in
numerous other texts and that pupil groups are frequently used in the
classroom, although methods of grouping may range from laissez-faire
(unplanned) to friendships to ability (homogeneous or heterogeneous)
5
bases. Competitive versus co-operative learning processes will produce

various social and academic performance effects.
This book should enable the reflective teacher to draw upon a variety
of models and considerations of classroom process in planning for
classroom activity. The processes will relate to numerous curriculum
subjects. We hope to expand classroom accountability from product to
include process, and to legitimize that process is socially as opposed to
individually based in the classroom.
Structure of the book

The planning and arrangement of chapters in the book fall into two
sections. The sections are strongly linked to one another, and the links
will become clear as one progresses through the book. The sections
represent introductions and overviews of general issues initially, and
focused/applied consideration of issues in the second half.
The first half of the book (Chapters 2 to 6) begins with a description of
the group work and social interaction in the classroom. Consideration
then moves to quantity and quality of social interaction between children
and teachers with a special focus on discourse analysis. Development of
the self is explored through self-concept, motivation and autonomy—
each aspect drawing upon experimental, theoretical and practical studies
of the classroom. These chapters have links among themselves and with
the focused discussions in the second half of the book.
The second section is based on classroom realities; what has been done
and what can be done. Of prime concern are the relationships engendered
in the classroom. The quality of hierarchical and collaborative
relationships generated in Piaget’s consideration of moral development are
made relevant to the classroom. Activities of young children in school and
the role of teacher- induced structuring of activity are taken up in play and
friendships. Distinctions between the behaviour of boys and girls provide
insights into the potential of gender differentiation and disruption in the
classroom. The section culminates with an amalgamation of process and
curriculum concerns by citing an American model of co-operative learning
(which strongly substantiates that academic and social products can be
enhanced by consideration of specific grouping and learning processes
which teachers may structure) and a critique of the integration of children
with ‘special educational needs’ into the normal classroom.
General issues
The chapter on ‘Grouping and Group Work’, by Maurice Galton,
(Chapter 2) is reflective of the ORACLE observations of process and
product in junior schools. Groups are a convenient and much-used
6
method for organizing children in classrooms. Group work has been

strongly supported by HMI and government reports. But the effects of
using groups in the classroom are not always clear or advantageous.
Teachers need to plan/structure their groupings rather than simply
‘allowing them to happen’. Group size, composition and task demand are
all considerations that ought to be brought to bear on classroom
organization. This chapter sets a background for ensuing concerns in the
book, especially motivation (Chapter 6), autonomy (Chapter 7),
friendships (Chapter 10), and co-operative learning (Chapter 13).
With a background into the use of groups, a more focused look at the
intimacy of classroom interactions is explored in the chapter by French
(Chapter 3). This chapter is important in that it qualifies distinct and
meaningful actions in the classroom which may benefit or hold back
development of learning and express particular gender bias. A
discussion of social interaction notes that both teacher and children
have rights and meaningful positions in the classroom. The interactions
focused upon and explored will be limited by the perspective of the
social scientist and the methods chosen for study. The importance of
this chapter is immediately exemplified in discourse (Chapter 4),
provides tools of analysis for self-concept (Chapter 5), autonomy
(Chapter 7) and play (Chapter 9). Themes in this chapter are further
explored in the discussion of gender/sex roles in the classroom
(Chapter 11).
The study of classroom conversation and its relationship to the
development of knowledge reminds the reader that learning is the result
of a social process. This chapter critically assesses the role of the
individual in classroom learning, and explicitly shows that this learning
takes place in a social context (the classroom). The type of knowledge
generated in this context is dependent on the quality of social
relationships that are structured to take place between interactors.
Knowledge, in short, is a process of shared meanings in a climate of
specific power relationships. The discourse analysis used in the chapter is
an example of a social interactional methodology (from Chapter 3), and
has direct links to classroom activity in the development of classroom
autonomy (Chapter 7), the relationship between teacher and pupil
(Chapter 8), and peer relations in co-operation (Chapter 13).
More focused exploration of the development of the child’s self-
concept and its correlation to achievement is reviewed by Schunk in
Chapter 5. He reminds the reader of the child’s active role in the learning
process and how classroom performance is strongly linked to the
learning context itself. This chapter is premissed on past research
showing that the child’s self-identity is established through social
interactional and evaluational processes by which primary school
classroom life is made meaningful for the child. Discussion leads
7
directly to further consideration of classroom motivation (Chapter 6),

friendships generated through schooling (Chapter 10), and possibly in
themes identified in special needs (Chapter 12).
The pupil’s motivation to enhance school-based learning and/or
social relationships is introduced by Schunk but more fully explored by
Rogers in Chapter 6. This chapter draws upon studies of classroom
practice and social theories of identity. Pupils’ approach to classroom
work is strongly related to the social and interactional climate structured
by the teacher. The resultant quality of interaction structured in the
classroom will depend on the understanding of autonomy drawn upon by
the teacher (Chapter 7), specifically alludes to the relationship between
teacher and child (Chapter 8), and poses an alternative practice to
promote motivation amongst the whole class taken up in co-operative
learning (Chapter 13).
The general issues section is completed by providing some in-depth
consideration about the social development of the child. Social
development is discussed in terms of the type and quality of
relationships within which the child interacts. Archetypical
relationships with adults and peers are described as the roots of
autonomy, and are significantly affected by the structured relationships
and motivational concerns of the classroom. While theoretical in
nature, this chapter draws on the role of groups (Chapter 2),
relationships (Chapter 8), friendships (Chapter 10), and co-operation
(Chapter 13) in the classroom.
Focused issues
Derek Wright has been involved in the study and classroom implications
of moral education for many years. Do not mistake this chapter
(Chapter8) as a curriculum issue, as it is based upon practical concerns
and the importance of relationships between teachers and pupils. Themes
recurrent in this book are made relevant here, especially the qualities of
hierarchy and collaboration that the teacher may structure in classroom
life. The practicality of this chapter is complimented by theoretical and
structural concerns in the previous chapters on motivation (Chapter 6)
and autonomy (Chapter 7).
The role of play in the classroom again combines curricular and
practical considerations. Peter K.Smith has researched the complexity of
this subject over many years. Readers may be aware that issues presented
here (in Chapter 9) are immediately relevant to pre-schools and early
infant classes. Nevertheless, these issues describe the active roles played
by both teacher and pupil in their own development, and that teachers
must consider the careful structuring of play as essential to their
classroom process. Issues in this chapter are taken up in all of the
8
succeeding chapters, especially those on friendship (Chapter 10) and sex

roles (Chapter 11).
Maxwell’s consideration of friendship in Chapter 10 integrates
developmental and social concerns, and lays an important issue before
the classroom teacher. Much of children’s time will be spent developing,
extending and altering friendships. In one sense children display
‘natural’ tendencies towards certain types of friendships. On the other
hand we must be aware that friendships are often structured by the
culture and interactional experiences offered to children. This chapter
picks up themes of social interaction (Chapter 3) and play (Chapter 9),
and the importance of structuring relationships is further explored in co-
operative learning (Chapter 13).
Sex roles in the classroom is a hotly debated and debateable issue.
Here we need basic information, minimally to move away from common
prejudice. While it has often been assumed that girls perform more
poorly in schools than boys, this is not so in primary schools, and we
shall need to place strict qualifiers around the statement. There may be
distinct patterns of interaction that differentiate between boys and girls in
primary schools. Some of the interactional differences may not be
characteristic of boys or girls generally, but may be magnified by
particular teacher-based concerns (especially of disruption). Issues
raised in Chapter 11 are derived from social interaction (Chapter 3),
discourse analysis (Chapter 4), and motivation (Chapter 6). That some of
these sex differences may be the result of inadvertent teacher structuring
is discussed in Chapter 12 on special needs (including disruption). Some
of these behaviours may be changed in processes such as co-operative
learning (Chapter 13).
Consistent reports identify that up to 20 per cent of children in school
may present themselves as having special educational needs at any point
in their schooling careers. Galloway, who has researched the generation
and identification of classroom disorders, considers in Chapter 12 the
role of teacher and classroom structure in coping with the integration of a
large number of special needs children in the normal classroom. A
central dilemma for the teacher is how to structure ‘normal’ classroom
processes to include the child with special needs. Social dynamics of
pupil groups in the classroom may, if allowed to structure themselves,
result in discrimination and separation of pupils from one another. Other
classroom structures as co-operative heterogeneous groups have
produced significant movement towards integration and enhanced
motivation of all pupils (see Chapter 13).
The final issue chapter in this section is the one most consistently
referred to throughout the text. Robert Slavin, internationally known for
his practical research into co-operative learning, brings together a
number of issues. In acknowledging that traditional classroom structures
9
and laissez-faire groupings produce genuine learning amongst pupils

(but this is combined with discrimination), we ask how teachers may
structure their classrooms more effectively to produce good results for
all. The practice of co-operative learning, which has been explored for
some years in the United States, draws upon the realization that learning
is a social process. For learning to be effective at all, the teacher’s role in
structuring learning tasks, groups and assessment is essential. This
chapter (Chapter 13) will draw upon themes of co-operative learning and
report on its practical usage and results.
The final chapter of the book (Chapter 14) reviews a number of issues
discussed. In so doing, the intention is not to provide a summary. The
chapter will take the issues and focus upon models for classroom
structuring that the teacher may initiate. As stated in the opening of the
book, our intention is to draw upon current and up-to-date classroom
studies and social psychological concerns. The integration of these issues
provides a solid background for the assertion that processes of classroom
interaction should be studied and structured in teaching. Teachers and
schools should not be held to account solely for curricular-based results.
The reflective practitioner will plan for process concerns of the self,
motivation and autonomy in the relationships and grouping approaches
of the classroom, and these concerns will apply to all of the subjects in
the national curriculum.
References
Bennett, S.N., Desforges, C., Cockburn, A. and Wilkinson, B., (1984) The
Quality of Pupil-Learning Experiences, London: Lawrence Erlbaum
Associates.
Department of Education and Science (DES) (1978) Primary Education in
England, London: HMSO.
Education Reform Act (1988), London: HMSO.
Morrison, A. and McIntyre, D. (1969) Teachers and Teaching, London: Penguin.
Morrison, A. and McIntyre, D. (1971) Schools and Socialization, London:
Penguin.
Morrison, A. and McIntyre, D. (eds) (1972) The Social Psychology of Teaching,
London: Penguin.
Plowden Report (1967) Children and Their Primary Schools, 2 vols. Report of the
Central Advisory Council for Education in England, London: HMSO.
10
Chapter two
Grouping and group work
Maurice Galton
Editors’ introduction
Maurice Galton’s brief was to provide a review of research into group
work within the classroom. This he has done by drawing attention to a
number of central dilemmas that group work can invoke.
Central to these concerns are the issues of pupil autonomy and teacher
control. Galton demonstrates how the study of group work can lead to a
consideration of issues relating 10 pupil motivation and self-concept.
Pupils are required to negotiate their way through the often conflicting
demands made upon them both by their teachers and by their peers. Pupils
are seen to be striving to develop a sense of ‘ownership’ over their work
while at the same time seeking to avoid having to accept responsibility for
work that is judged to be less than adequate by their teacher or peers.
Group work can liberate the pupil from undue teacher control, enabling
ownership and encouraging the development of autonomy. At the same
time, the more closely involved with the group the individual pupil
becomes, the more public his or her efforts will be. Group work can thus
aid the pupil in the quest for ownership but also increase the threats implicit
in increased peer evaluation. The application of group work strategies
cannot therefore be considered unproblematic but must also be seen as
potentially rewarding.
The issues introduced here will be returned to on a number of
occasions throughout the book, but, in particular, common threads will
be found in the chapters on self-esteem (Chapter 5), pupil motivation
(Chapter 6) and pupil autonomy (Chapter 7). A related perspective on
group-based approaches to learning can be seen in the chapter on co-
operative study methods (Chapter 13).
Introduction
One of the most striking changes to have taken place in the post-war
decade at primary level has been the change in the pattern of organization
11
Maurice Galton
and the use of more informal seating arrangements. For example, writing
about the late 1950s, Donald Jones provides an account of a
Leicestershire classroom where, according to the teacher, ‘We used to
have classes of over fifty in such tight rows that nobody could leave the
room until somebody stood up’ (Jones 1987:35). Yet by the beginning of
the 1970s most junior school classrooms in the same authority used some
variation of group seating arrangements (Bealing 1972). The
organization of pupils into groups was one of the distinguishing
characteristics marking the move away from rigid streaming, following
the abolition of the 11+ and reorganization of secondary education along
comprehensive lines. By the time the ORACLE survey reported (Galton
et al. (1980), over 90 per cent of the classes visited were using some kind
of group arrangement to seat pupils with less than 1 per cent retaining the
traditional patterns of rows of desks.
Definitions of group work

When questioned by the ORACLE observers, however, many of these
teachers looked upon group work as little more than a form of seating
organization, particularly in mathematics, where children were seated in
homogeneous groups according to ability. Even where groups were
heterogeneous, to a large extent there was little emphasis on instituting
joint tasks and even less on allowing children to help one another with
their work. Beyond these seating arrangements group work then had a
range of meanings. At one extreme was joint group work where pupils
were engaged on specific tasks which then contributed to an overall
theme. For example, children might be mounting a wall display in the
classroom as part of an environmental study of the local surroundings.
Some children might be projecting a transparency of a map of the area on
to the classroom wall and colouring in various sections while others
would be engaged in mounting specimens of material collected during
the field trip and locating these at the various places where they were
found. For the most part decisions concerning who did what task would
have emerged during joint discussion with the teacher. Joint group work
is therefore characterized by the setting of individual tasks linked by a
common theme and contrasts with seated group work where the children
sit at the same table but engage in similar but unrelated tasks. For
example a group of children may be on the same mathematics worksheet
but they will be expected to complete the questions on their own.
At the other end of the scale is co-operative group work, where children
are expected to pool their ideas as part of a joint presentation. One
example, taken from the ORACLE study, concerned a group of children
who were engaged in a cloze procedure exercise. The children were
expected to pool their ideas in order to decide on the missing words. Co-
12
operative group work therefore has a joint outcome to which all members
of the group have, at one time or other, contributed.
The use of group work in the primary classroom

In the ORACLE study it was common to see examples of joint group work in
most classrooms involving common tasks but individual assignments. These
joint activities, however, were seldom used for teaching basic skills of
computation and English or for science, but were largely restricted to art and
craft or general studies where there was a practical element. Less than a
quarter of the teachers who made use of joint group work did so for more
than one curriculum area (Galton 1981). In a more recent investigation of
curriculum provision in small schools, the PRISMS study (Galton et al.
1987), similar patterns were observed in both infant and junior school
classrooms. Here there was greater teacher interaction with groups than in
the ORACLE study (12.6 per cent in infant classes, 15.8 per cent in junior
classes compared to 9.4 per cent in ORACLE classes). This was, in the main,
a consequence of the fact that in most classes in small schools the age range
of the children is wide and pupils tend to be seated and taught in groups of the
same age. In these groups, when the teacher was not present, pupil-pupil talk
was less than in ORACLE (13 per cent as against 19 per cent). However,
unlike ORACLE, in PRISMS 55 per cent of these pupils’ conversations were
task-related compared to only 20 per cent in the ORACLE study. In the
PRISMS analysis, however, a wider range of curriculum activities (including
games, dancing and drama) were included. Nevertheless, the overall patterns
of working showed that 79.1 per cent and 81.2 per cent of observations in
infant and junior classrooms involved a pupil working alone within the group
(seated group work). When children were involved in collaboration then 3.5
per cent and 4.2 per cent of all activity of infant and junior-aged children
respectively was in pairs compared to 4.8 per cent and 5 per cent in larger
groups. This suggests that, overall, the amount of collaborative work in small
schools was not much greater than in ORACLE. Further evidence for this
conclusion comes from recent studies conducted in the Inner London
Education Authority, at infant level (Tizard et al. 1988) and at junior level
(Mortimore et al. 1987). Here again, for the most part children sat in groups
but worked individually. According to Mortimore et al. (p. 82) ‘Not a great
deal of collaborative work was observed’ although:
there was some evidence that where pupils worked on the

same task as other pupils of roughly the same ability or when
pupils worked within the same curriculum area but on
different tasks at their own level, the effect on progress was
positive.
(Mortimore et al. 1987:230)
13
Maurice Galton
Differing prescriptions of good practice

In the last twenty years there have been continuous calls to include more
collaborative group work within primary pedagogy. Her Majesty’s
Inspectors’ survey (DES 1978) defined groups in terms of seating pattern
rather than teaching strategy and argued that the main advantage of such
grouping was to enable teachers to provide work of an appropriate level
of difficulty (para. 8.32). The inspectors called for more direct teaching
of groups, particularly in mathematics, arguing that this was a more
efficient way of introducing new aspects of the subject to children.
Groups were therefore seen by the inspectors as a way of increasing the
amount of contact between teachers and pupils, particularly if the work
involved thinking at higher levels of abstraction. They argued, for
example, that in teaching mathematics: ‘challenging questions and quick
recall of numbers and facts, including multiplication tables often
required a lively sustained contact between the teacher and a group of
children’ (DES 1978: para. 5.65).
The inspectors’ view of grouping, therefore, seems to give tacit
support to their finding that over 70 per cent of teachers place
children in ability groups for mathematics, since in such groups
‘children were doing more challenging work and the teacher was able
to inject more pace’.
This suggests that, for basic skills, groups should be set up where
pupils are able to undertake work of a similar level of difficulty—a view
supported by Mortimore et al. (1987).
This viewpoint, however, concerning the functioning of groups,
differs markedly from the prescription to be found in the Plowden (1967)
Report. The report placed great emphasis on the individualization of the
learning process but acknowledged that in a typical sized classroom this
meant that only limited amounts of teacher-pupil contact were possible.
Collaborative group work was a way of overcoming this problem and
was recommended, in particular, for science activities. According to the
Plowden prescription, teachers could economize on their time ‘by
teaching together a small group of children who are roughly at the same
stage’ (para. 754).
This was not a recommendation for streamlining within the classroom
since the Plowden Committee recommended that ‘the groups should be
based on interest or sometimes on achievement, but that they should
change in accordance with children’s needs’ (para. 824). The
organization of pupils into groups in this way, however, was advocated
by Plowden not only to allow more efficient use of teachers’ time but so
that the number of contacts between pupils and teacher could be
increased. Collaborative grouping also performed a pedagogic function
since the report claimed that within such groups children ‘make their
14
meanings clear to themselves by having to explain it to others and gain

some opportunity to teach as well as learn’. Group interaction of this
kind was thought to help the timid child who might be ‘less shy in risking
hypotheses in a group’, while apathetic children ‘may be infected by the
enthusiasm of the group while other children benefit from being caught
up in the thrust and counterthrust of conversation in a small group of
children similar to themselves’ (para. 757–8).
The Plowden view of group work was therefore in marked contrast to
that suggested by the inspectorate who appear to have seen its use as a
means of teachers’ directing work more efficiently and stimulating the
group to higher levels of thinking. Plowden, on the other hand, argued
that the teacher has ‘missed the whole point if he tells the children the
answers or indicates too readily or completely how the answers may be
found’ (para. 669). Class discussion should be introduced towards the
end when the individual pupils’ contributions are complete so that ‘the
pieces of the jigsaw can be fitted together…or seen not to fit’ (para. 760).
Thus the use of groups in Plowden’s sense involved co-operative
working between children which continued even when the teacher was
engaged elsewhere. The teacher’s role was to promote enquiry with the
twin objectives of stimulating pupils’ thinking and developing
communication skills.
Many of the same outcomes are endorsed by Kerry and Sands (1982).
They argued that the main advantage of collaborative group work is that
it helps pupils to work together co-operatively and allows them to learn
from each other, thus removing the stigma of failure for slow learners.
Pupils are given a chance to work at their own pace and become less
teacher dependent. Group work, according to Kerry and Sands, also
improves individual pupils’ self-image in that by working in groups the
children come to respect each other’s strengths and weaknesses. For the
teacher the advantages were said to be two-fold. First it provides an
opportunity for the teacher to circulate and correct individual work while
other children are collaborating actively within groups. Second, it
enables the teacher to tailor a range of tasks which are more appropriate
to the children’s needs and abilities. These aims are also endorsed by
Yeomans (1983).
Criticisms of group work

These views are not universally shared among those writing about
current primary practice. Robin Alexander, for example, takes to task
those local authority advisers whom he claims, following the ORACLE
findings, convey to teachers the message that ‘unless pupils are
interacting and collaborating within their groups, there is little point in
grouping them’ (Alexander 1988:178).
15
Maurice Galton
Alexander goes on to describe the work of a group of teachers who

place children into collaborative group situations, to engage in problem-
solving, while they themselves continue to instruct other pupils
individually although these latter pupils are seated in groups. This is
exactly the pattern recommended by Plowden. Alexander claims that the
teachers’ interpretation of the prescription for grouping requires everyone
in the class to be working in groups at the same time and, in so far as this is
attributed to ORACLE influence, it is incorrect. ORACLE, like most other
studies of primary classrooms, has persistently called for the use of mixed
organization strategies although, as has been argued elsewhere the use of
individual instruction coupled with children working on their own
collaboratively in groups is itself problematic (Galton 1989a).
There are others, however, who actually question the use of grouping as
part of a ‘mixed economy’ of primary practice. Wheldall et al. (1981) found
a significant increase in on-task behaviour when children were seated
individually in rows, and that levels of distraction increased when children
returned to seating in groups. Wheldall’s study has been replicated by
Bennett and Blundell (1983). They used the quality of the work as an
additional dependent variable as well as ‘time on task’. They found that the
quantity of work increased when children were moved from groups to rows
and that there was no decline in quality. In the United States, among those
making similar criticisms of group practices, there is a grudging acceptance
that some form of group participation may be required when the class
consists of vertically-aged cohorts. Writing in the third Handbook of
Research on Teaching, Brophy and Good (1986) argue that:
The small group approach requires well chosen assignments

that the students are willing to engage in and able to
complete successfully, as well as rules and procedures that
enable students to get help (if confused) or direction (about
what to do if finished) without disrupting the momentum of
the teacher’s lesson. Teachers…may find it takes too much
effort to be worth the trouble.
(Brophy and Good 1986:361)
Thus, in this country, despite the interest in group work which has
developed over the last twenty years at primary level, there is still little
agreement about the methods teachers should use when attempting to
develop collaboration and co-operation within their classrooms.
Research into group work and grouping in the United States

In the United States there has been much work on aspects of
collaborative working in groups, with somewhat different conclusions
16
emerging according to the different focuses of these studies. Much of the

American research has its origins in the investigation of earlier
researchers into the merits of co-operative, competitive and
individualistic goal structures. Thus the emphasis in most American
studies is on the effect of these different ‘goal structures’, on group
productivity and achievement Unlike the United Kingdom, there have
been fewer studies of a descriptive nature where the interactions between
pupils within such groups have been observed.
By far the largest body of research has emerged from the work of
Slavin and his colleagues (reported in chapter 13 of this volume).
Slavin’s model of group working in the classroom is based upon a team
games approach (Slavin 1983, 1986). Just as in the sporting analogy,
class teams are made up of high, average and low-performing students,
boys and girls, and, where appropriate, pupils with different racial or
ethnic backgrounds. As far as possible each team is representative of the
whole class structure. Team members have to accept responsibility for
the learning of all individuals within the team and not just their own
progress. While the pupils can help each other during team practices,
they have to take the test, on which the group score is based, on their own.
Individual scores of team members are then combined to form the team
score. There are also other versions of this procedure known as the
student teams achievement divisions (STAD), involving team games
tournaments (Devries and Slavin 1978) and team-assisted
individualization (Slavin et al. 1984).
In an analysis of some thirty-three studies in both elementary and
secondary classrooms it is claimed that twenty-two showed a significant
positive effect on student achievement, particularly when the study
involved student team-learning techniques. It is also claimed that such
arrangements improve intergroup relations particularly with students of
different ethnicities. Grouping was also said to improve self-esteem
(Slavin 1983), particularly when a team-learning approach was
combined with a jigsaw arrangement where each student was given the
opportunity to become an expert on some aspect of the work (Aronson
et al. 1978).
In this type of programme pupils who had to master the same area of
expertise were brought together in a group to pool ideas. They then
returned to their own teams to teach the other pupils what they had
learned in the expert group. As in the STAD programme, each pupil then
took individual quizzes and these were aggregated into team scores. The
essential characteristic of Slavin’s approach to group work, therefore, is
that it involves some group reward incentive structure. A summary of this
work is given by Bennett (1985).
A different strategy has been developed by Johnson and Johnson
(1976). The Johnsons’ model is based upon the definition of
17
Maurice Galton
collaborative group work used earlier. Unlike Slavin’s approach, where

individuals complete the assignment by themselves and the scores of
individuals are then aggregated, the Johnsons recommend that the
outcome of the groups’ activity should consist of a single assignment
sheet, as in the earlier example of cloze procedure. In addition, the
groups are judged not only on how well they do on the group task, but on
how well they have worked together in carrying out the assignment In an
analysis of a number of studies, Johnson and Johnson (1980) have
claimed that such co-operative structures promote higher achievement at
all age-levels and in all subject areas on tasks demanding a range of skills
such as concept attainment, predicting and evaluating. Johnson et al.
(1981) claim that the evidence demonstrates the clear superiority of co-
operation in promoting achievement and productivity although these
conclusions have been attacked by other researchers such as Cotton and
Cook (1982). These writers are particularly critical of the fact that many
of the studies cited by Johnson et al. were of very short duration (less
than two weeks) and were not carried out under normal classroom
conditions.
A number of studies have also been carried out in Israel with the aim
of improving the performance of pupils in science (Sharan 1989). These
researchers claim that working co-operatively in groups improved self-
esteem, the learning environment and student achievement Student
achievement was measured by using standardized tests and the learning
environment by means of an inventory developed by Walberg and
Anderson (1968). Self-esteem was assessed by means of a questionnaire,
based on an earlier version developed by Aronson et al. (1978). These
Israeli studies were, therefore, more concerned to test the hypothesis that
collaborative working in the classroom led to an improvement in student
outcome measures rather than attempting to explain the mechanism by
which working in groups facilitated such outcomes. This latter objective
can only be achieved by a study of pupil and teacher behaviour within the
classroom during group work supported by the teacher’s and the
children’s explanations for their observed behaviour, rather than the use
of rather crude questionnaires as in the majority of American and Israeli
studies. Explanations based upon this kind of analysis make up the
remaining part of this chapter. At present most of these conclusions
remain tentative and require further empirical testing. They do, however,
offer a framework for developing effective pedagogic strategies for
group work.
Effective group work in the primary classroom

The starting point for further discussion requires consideration of the
follow-up project to ORACLE, Effective Group Work in the Primary
18
Classroom. The main findings are described briefly in Galton (1987,

1989b). The project was conceived because of interest in a particular
teaching style identified during the ORACLE study. A small group of
teachers appeared to follow the HMI recommendation in DES (1978)
and engage children in joint group work. These teachers were called
group instructors. Associated with the use of this teaching strategy were a
group of pupils who worked together in groups. However, an
examination of their interaction patterns showed that, while they did
collaborate together, this collaboration generally involved sharing
materials and did not extend to conversation. This group of pupils were
therefore described as quiet collaborators. Given that the ORACLE study
had demonstrated that collaborative group work in the primary
classroom was a ‘neglected art’ (Galton 1981), teams of teachers were set
up charged with extending this ‘quiet collaboration’ into areas where
conversation about ideas, and so on, was required. These teams of
teachers then attempted to improve the practice of a group of colleagues
using a ‘team-training approach’ which is currently in vogue for the
development of new TGAT (Task Group on Assessment and Testing)
initiative (Black 1988). For a period of one term this second group of
teachers were observed when attempting to implement these new group
strategies. Their performance was compared with a control who had
followed a more conventional in-service course designed to improve
aspects of individualized and whole-class teaching.
Initially the results obtained from the first systematic observation of
the experimental and the control group were encouraging. The
proportions of lessons involving extended discussion in the experimental
group was 35.5 per cent compared to only 19.2 per cent in the control
group. In both cases the teacher was present with groups for
approximately the same amounts of time so that it appeared from this
evidence that the children in the experimental study were collaborating
together more frequently. Time on task, the degree of distraction and the
amount of time waiting for teacher were all lower in the experimental
group employing the collaborating group work strategy. This result
therefore was in accord with the findings of Johnson and Johnson (1976).
However, when the second set of observations were carried out some two
to three weeks later, these differences had all but disappeared. While
discussions in the control group remained around 20 per cent, in the
experimental group they dropped to around 24.6 per cent and distraction
rose in both groups to around 18 per cent. At the same time there was
increased teacher involvement with the experimental groups. Thus after
only a short time the patterns of interaction within both groups were
remarkably similar to those which had been found earlier in the original
ORACLE study and which have since been confirmed in other studies
such as PRISMS (Tizard et al. 1988 and Mortimer et al. 1988).
19
Maurice Galton
Further analysis of this data revealed an interesting pattern. An

examination of the original observations, where the experimental group
had increased the amount of collaborative activity between children in
groups to around 36 per cent, showed that there were large differences
according to the curriculum area involved. Most striking was the fact that
nearly 71 per cent of all collaborative interactions during this initial visit
to the classroom involved the use of materials within the experimental
group, compared with only 40 per cent in the control group. When
correlation co-efficients were calculated between pupil behaviour and
different curriculum activities during the first observation session the
reason for the high level of group work in the experimental group and its
decline by the second set of observations became apparent. In the
experimental group discussion was negatively correlated with reading (-
0.129), writing (-0.385) measuring (-0.351) and arithmetical
computation (+0.278) but, positively correlated with the use of materials
(+0.541). The main difference between the two groups initially,
therefore, was that the experimental group had made much greater use of
materials and hence obtained higher levels of collaboration. Once the
topic was under way and there was less use of materials and a need for
more writing, reading, measuring and discussion, the pattern of both
groups became strikingly similar, with a marked reluctance on the part of
the pupils to work together in the absence of the teacher.
Pupils’ perceptions of collaborative group work

The next stage of the project was to examine the pupils’ reasons for this
reluctance to work collaboratively. Rather than make use of
questionnaires a projective technique involving the use of cartoon
pictures was devised in the hope that pupils would then ‘say what they
felt’ rather than what they thought the interviewer required. The cartoon
pictures showed pupils in various classroom settings. Fuller details are
given in Galton (1987). Of special interest were two cartoons, one which
showed children talking to each other in a group without the teacher
present and the other portraying children talking with the teacher with a
tape recorder being used to monitor the discussion. The pupils were
asked to say which were their favourite cartoons and to describe the
conversation which was taking place. The children had to choose two
pupils from the group in the cartoon picture and describe what these two
children were either saying or thinking.
For the most part there was a strong rejection of collaborative group
work. In particular the pictures where no teacher was present accounted
for 30 per cent of all rejections. The least popular picture (17 per cent)
involved the pupils’ tape recording. Children gave various reasons for
their lack of enthusiasm for group work.
20
‘My friends make me silly’.

‘There’s no teacher. You’re worried if you get things wrong’.
Only in one class of 12-year-olds, among some fifty studied, was there a
strong positive response to the pictures where children were working on
their own without the presence of the teacher. This group of pupils had
very positive feelings about collaborating together.
— ‘Its good to work things out with the teacher. You can
have a laugh when you discuss’.
Some put it even more strongly:

‘I would like to work with my friends and discuss things on
our own because when you work with teachers they always
stop you’.
‘One can learn more from each other when there is no
teacher to nag’.
When the children in all the classes, including the one with a positive
attitude to co-operation, described what teachers were saying when they
came to a group there was an overwhelming perception that most
comments were likely to be negative, to do with the teacher’s attempt to
get them to work properly. Repeated remarks included such things as,
‘Look at the picture, get on!’

‘Now listen to me.”
‘You’ll not leave this room until you’ve finished’.
The contrast between the typical response presented by the pupils in the
sample and the children in the one class favouring group work led to an
extended case study of the latter, culminating with the author going to
teach in the school for a six-week period. Much of the analysis of this
experience is contained in Primary Teaching (Galton 1989b). Here,
however, I wish to concentrate on some of the dilemmas which emerged
as a result of the analysis of these case studies. These dilemmas centre
round three particular contradictory features of primary classrooms
which can be extended to other issues besides group work. They are
particularly relevant, however, to the latter activity because, as I shall
argue, group work provides pupils with very little opportunity to engage
in the standard ‘avoidance’ strategies which are commonly used to deal
with more typical classroom situations where children are taught
individually or as a class.
21
Maurice Galton
Ownership and risk during group work

The first dilemma centres around the conflict between the pupils’ desire
for public acknowledgement of their contribution to the group effort and
their concern that when the teacher and other pupils acknowledge their
contribution there is also the risk of a ‘public’ failure with corresponding
loss of self-esteem.
Fear of failure is an extremely strong element in the primary
classroom as documented by other writers such as Pollard (1985) and
Woods (1980). Pollard describes how pupils’ ‘interests at hand’ lead
them to develop what Measor and Woods (1984) call ‘knife-edging’
strategies of avoidance in order to develop a safe ‘working consensus in
the classroom’. In the ORACLE studies Galton et al. (1980) also
described similar strategies. Pupils interviewed as part of the follow-up
study to ORACLE, Effective Group Work in the Primary Classroom,
likened activities such as classroom discussion to ‘walking on a
tightrope’ (Galton 1987).
Thus the dilemma which faces pupils when working in groups
concerns the tension which exists between their desire to own their
ideas and the risk which such public acknowledgement carries of
appearing to fail in the eyes of their peers and of the teacher. Case
studies carried out, not only during the group work project but in a more
recent study of children’s writing, suggests that children want the work
they do to be acknowledged and to be seen as ‘theirs’. As Pollard (1987)
argues: ‘children learn best when they feel in control of their own
learning and interpretive sociologists reinforce this from a motivational
point of view’.
Pupils lose this sense of ‘control’ when teachers seek to impose some
order on the children’s thinking. Here, for example, is a sequence taken
from an interview with pupils during the project
Effective group work in the primary classroom

The group are explaining why it is they like working in groups
Pupil 1: When you are in groups you can discuss it, can’t
you, instead of working on your own. It’s better
working on your own than working with the class.
Pupil 2: I think it is best when the teacher comes because
they don’t want you to mess about Because when
you are on your own you are always talking about
other things but when you are with the teacher you
start working harder.
Pupil 3: Yes, the teacher helps you. She gives you different
ideas.
22
Pupil 1: I think the teacher wants to put her view into what
you are thinking which might make you change
your mind about something. You know, instead of
keeping to your own idea.
Interviewer: Is there anyone over here? Come on.
Pupil 4: Teachers stop you if you are right, Say you get, say
your answer’s right and they think it is wrong well
they will stop you and put what they think they
want you to do. They don’t like you to do your own
work but sometimes they do.
Pupil 2: When you have to do something, like we have had
that before, and you have got to do a certain number
of things, when the teacher comes up telling you
you’ve got a right good idea you go away and do it
and they will come back and alter it all and they will
make you do something else and (tell you) it’s got
to be like this.
Interviewer: Now, why do you think that is?
Pupil 1: Because they think it is best.
Interviewer: Because they think it’s best?
Pupil 1: Because they think it can be improved.
Interviewer: Does that stop you putting your ideas as well?
Chorus of pupils: No.
Interviewer: You can still put your ideas forward then?
Chorus of Pupils: Yes.
Interviewer: O.K. So what do you feel like if you think your idea
is a good idea and then it happens like you say and
the teacher comes and changes it?
Pupil 2: You feel a bit upset. You have put all that work into
it and then the teacher suddenly changes it
Pupil 3: You get a bit mad with her.
Pupil 1: You don’t feel it is your piece of work. You feel
as if it is the teacher’s. When you have done
everything to it and you think, that’s my piece of
work and no-one else has done owt to it. But
when the teacher has done something to it it
don’t feel as good.
Studies suggest that the idea of guidance which lies behind many of the
interventions by teachers which produce the above perceptions in
pupils come from what might be termed a ‘two-stage’ theory of
pedagogy. Initially the teacher attempts to support and extend the
pupil’s tentative ideas in the belief that this support will encourage the
pupil, increasing their confidence so that they will then continue
23
Maurice Galton
working independently. From the point of view of many pupils,

however, it would appear that this initial guidance is seen as a ‘take-
over bid’ where the pupils feel they are no longer in a position where
they think that the ideas are their own. As a result enthusiasm and
motivation decline and the work produced is what the teacher is
thought to want rather than what the pupil wished to do.
However, this desire for ownership has another dimension, since it
carries with it a public risk of failure. This is particularly so when
working in small groups where an individual’s contribution is open to
close scrutiny by both the teacher and one’s peers. As the beginning of
any new group activity, therefore, each participant’s self-esteem is at
considerable risk. Pupils with low ‘self-concepts’ will either be content
to accept the teacher’s intervention because it allows them to pass
ownership back to the teacher, or else they will attempt to persuade
another more competent member of the group to take over the teacher’s
role. During the initial stages of group work, therefore, these pupils
welcome the teacher’s intervention because they prefer to be dependent
on the teacher and thus escape the responsibility of ownership. These are
the pupils who disliked the cartoon picture when the teacher was not
present because their friends ‘made them silly’ or ‘they were worried
about getting things wrong’.
There are other pupils, like pupil no. 4 in the interview, with healthier
‘self-concepts’ who appear to resent the teacher’s intervention because it
denies them the opportunity to demonstrate ownership of their ideas. In
the initial stage, however, they are not sufficiently confident to wish to
express these ideas in public. The two-stage theory of guided discovery
seems to offer the worst solution to both kinds of pupils in that it
encourages dependence in one group and produces resentment in the
other as demonstrated by the transcript of the interview. An alternative
two-stage theory is, therefore, required whereby the teacher limits their
interventions initially during the ‘incubation’ period until the group
members are sufficiently confident in their own ideas to expose them to
the teacher. One reason for the success of the jigsaw arrangement
whereby the pupils begin their discussions in pairs is that it minimizes the
risk of failure in front of peers and the teacher. When such groups finally
come together it is very difficult for any member apart from their initial
partner, to know the extent of an individual’s responsibility for the ideas
which are presented.
Teacher interviews and evaluations

A second dilemma facing teachers relates to the framework which they
use to evaluate the success or failure of a group activity. We saw earlier
that the use of group work could be justified on a number of grounds,
24
including the improvement of understanding among slow learners, the

development of teaching skills among their peers and general
development of social cohesion and collaboration within the group. At
any one point in time during a lesson the teacher may have all of these
objectives in mind. However, when intervening within a group in order
to provide feedback, or some other form of evaluation, teachers tend,
naturally, to base this judgement on some specific criteria appropriate
to the particular situation. Thus some children may be praised because
they have shown social responsibility in that they reduced the amount
of disruptive behaviour while others might be criticized because they
did not produce a sufficiently coherent outcome. Teachers make these
judgements on the basis of their individual knowledge of children,
knowledge which other members of the class or the group may not
share.
Thus from the child’s point of view the situation may seem to be one
of confusion. For children to understand the teachers’ evaluation they
must be party to the criteria upon which the judgements were based. It
seemed strange therefore, in our studies, to find that although teachers
emphasized the processes of group work and were usually effective
when explaining what the children were required to do by way of tasks,
they rarely explained why the children were required to do the task in a
particular way. Indeed, many teachers told us that they thought that
children ‘of this age don’t need to know why they are doing things’
whereas the pupils, when interviewed, said that ‘if I knew why I was
doing it I wouldn’t mind, but I don’t know why I’m doing it and I can’t
see the point of it’. In the school where children rated co-operative group
work more highly teachers had carried out a careful analysis of the roles
of children within the group. In addition to leadership roles such as
identifying goals, allocating work and summarizing viewpoints, children
were identified as willing followers who sometimes carried out tedious
tasks such as tidying up and generally keeping things going by acting as
gatekeepers. Other pupils helped to resolve disputes, and attempted to
bring the discussion back to the point. These teachers defined effective
group work as the capacity for every child, at one time or another to act
out all these roles. Children were thus very clear on the criteria by which
they were being judged within the group, and much of the teacher’s
evaluations concerned the extent to which they had consciously taken up
these different roles.
Learning and behaving during group work

The third dilemma involved in group work concerns the ambiguity which
exists within many primary classrooms concerning learning and
behaviour. One of the main reasons for pupils continuing to reject the
25
Maurice Galton
opportunity to work co-operatively with their peers is contained in the

remark of the pupil quoted earlier ‘My friends make me silly’. Almost all
teachers interviewed during the effective group work project saw a clear
distinction in their role as ‘the teacher’ and as ‘the facilitator of
children’s learning’. Elsewhere I have written of the pervasive message
of the primary classroom which preaches to children the fact that when
it’s learning ‘you do as you think’, but when it’s behaviour ‘you do as I
say’ (Galton 1987). This s harp disjunction in the way in which learning
and behaviour is dealt with in the primary classroom has important
effects when children are asked to work together in groups. Teachers
appear to recognize this dilemma between the role of the teacher as
‘policeman’ and the times when they have to be ‘relaxed’ in order to
encourage informal learning (Nias 1988).
Unfortunately, while this distinction may be clear in teachers’
minds, it is not understood by most pupils. When working in groups
pupils are particularly vulnerable to the accusation (when the noise
level rises in the classroom) that not all this noise is productive. Even if
a teacher does not immediately reprimand the pupils for being noisy
they are likely to intervene in the group to check on the level of
purposeful activity, thus again creating in the children’s minds the
suggestion of a ‘take-over’.
Other researchers, such as Rowland (1987), argue strongly that the
tendency of pupils to ‘play safe’ out of ‘fear of doing things wrong’ is
also closely related to the evaluation of children’s work. Certainly in the
study, Effective Group Work in the Primary Classroom, it was noticeable
how frequently children, when responding to the cartoon pictures,
associated evaluative comments with disciplinary ones. Thus the teacher
would be described as saying, ‘Get it right or you will not go out to
play’, or ‘That’s wrong, stop fooling and concentrate’. Rowland’s way
out of this dilemma is to place greater emphasis, particularly in the early
stages of any activity, on the role of the teacher as critic where children
are encouraged to comment openly on the teacher’s ideas as well as on
each others.
Towards a model of effective co-operative group work

In some of our case studies there were teachers who appeared, intuitively,
to have a similar approach to Rowland. They reduced the risk to
individual children by, initially, never commenting directly on a
particular piece of work but by generalizing from a position of ‘neutral’
space within the classroom. In this context, ‘neutral’ space was defined
as taking up a position which was outside the territory of any one
particular group of pupils and, equally, was not seen as the teacher’s
space (that is, behind the desk). For example, one teacher (whose
26
practice was approved by the children later in the interview described

earlier) was observed, when visiting each group of children, to sit
alongside the groups listening to their initial reactions to a writing
stimulus. She appeared, deliberately, to convey to the children that she
was not interested in contributing to the discussion by sitting with her
elbow on the table facing slightly away from the group with one hand
clamped firmly across her mouth! This body language appeared to
convey to the children that she was there to listen and not talk. After
visiting all the groups this teacher then moved to the centre of the
carpeted area and said, ‘While listening to what you have been
discussing there have been a number of interesting ideas which I’d like to
share with all of you. I’d also like to put one or two of mine and get your
comments back’. Having done this she then revisited each of the tables
and on this second occasion said to the children, ‘Do you have any
comments on what I have just said?’
Most of the descriptions by Nias (1988) of the teachers’ attempts to
develop informal approaches within their classrooms seem to imply, yet
again, a ‘two-stage theory’ of teaching whereby one is firm at first and
only relaxes once the class is under control. Although in most primary
classrooms the teaching strategies used to obtain pupil co-operation are
generally milder versions of the ‘don’t smile until Christmas’ techniques
used in the secondary school, there were numerous examples in the
ORACLE study, particularly in the transfer to the age 9 to 13 schools, of
how this two-stage theory of control operated. At the beginning of the
school year teachers would talk about ‘My way of doing things’. They
would tell children that they were still ‘behaving like infants’ and would
therefore have to be treated as such! More usually they would seek to
capitalize on the advantage of the class-teacher relationship telling
children that their behaviour had made them (that is, the teacher) feel
‘sad’ and that as a result they did not feel like going on with this
particular exciting group activity. Instead children were required to do
worksheets individually.
Nias’s teachers seem to be searching for a model of a negotiated
classroom which allows responsibility for learning and behaviour to be
shared with the children. Writers such as Glasser (1969, 1986) have
long been concerned with the development of such approaches which
seek to confront children with responsibility for their teacher’s needs
as well as their own. For this to happen, however, children have to be
aware of the teacher’s needs; otherwise they cannot be in a position to
accept their share of responsibility in meeting them. Programmes such
as Thomas Gordon’s (1974) Teacher Effectiveness Training seek to
develop this approach to classroom behaviour as an alternative to the
current two-stage prescription in which the teacher has first to exercise
control either by power or persuasion. In this country we are only
27
Maurice Galton
beginning to develop Inservice Educating Teachers (INSET)

programmes which help teachers to negotiate, with their pupils, to
create the kind of classroom climate which furthers collaborative
working. Research into collaborative group work and other ‘high risk’
classroom activities, therefore, has wider implications for general
primary practice.
References
Alexander, R. (1988) ‘Garden or jungle: teachers’ development and informal
primary education’ in A.Blyth (ed.) Informal Primary Education Today:
Essays and Studies, Lewes: Falmer Press.
Aronson, E., Blaney, N., Stephen, C., Sikes, J. and Snapp, M. (1978) The
Jigsaw Classroom, Beverly Hills, Cal.: Sage Publications.
Bealing, D. (1972) ‘Organization of junior school classrooms’, Educational
Research 14: 231–5.
Bennett, S.N. (1985) ‘Interaction and achievement in classroom groups’, in
N. Bennett and C.Desforges (eds) Recent Advances in Classroom
Research, British Journal of Educational Psychology, Monograph series
no. 2.
Bennett, S.N. and Blundell, D. (1983) ‘Quantity and quality of work in rows
and classroom groups’, Educational Psychology 3(2): 93–105.
Black, P. (1988) ‘National curriculum, task group on assessment and
testing’, Three Supplementary Reports, London: Department of
Education and Science.
Brophy, J. and Good, T. (1986) Teacher behaviour and student achievement
’, in M.Wittrock (ed.) Handbook of Research on Teaching, 3rd edn. New
York: Macmillan.
Cotton, J. and Cook, M. (1982) ‘Meta-analyses and the effect of various
systems: some different conclusions from Johnson et al., Psychological
Bulletin 92: 176–83.
Department of Education and Science (DES) (1978) Primary Education in
England: a survey by HM Inspectors of Schools, London: HMSO.
Devries, D. and Slavin, R.E. (1978) ‘Teams-games-tournament: review of
ten classroom experiments.’, Journal of Research and Development in
Education 12: 28–38.
Galton, M. (1981) ‘Teaching groups in the junior school, a neglected art’,
Schools Organization 1 (2): 175–81.
Galton, M. (1987) ‘An ORACLE chronicle: a decade of classroom
research’, Teaching and Teacher Education 3 (4): 299–314.
Galton, M. (1989a, in press) ‘Primary teacher training: practice in search of
a pedagogy’, in A.McClelland and V.Varma (eds) Advances in Teacher
Education, London: Hodder & Stoughton.
Galton, M. (1989b) Primary Teaching, London: David Fulton.
Galton, M., Patrick, H., Appleyard, R., Hargreaves, L. and Bernbaum, G.
(1987) ‘Curriculum provision in small schools: the PRISMS project’.
Final Report, University of Leicester (mimeo).
Galton, M., Simon, B. and Croll P. (1980) Inside the Primary Classroom,
London: Routledge & Kegan Paul.
Glasser, W. (1969) Schools Without Failure, New York: Harper & Row.
28
Glasser, W. (1986) Control Theory in the Classroom, New York: Harper &
Row.
Gordon, T. (1974) T.E.T. Teacher Effectiveness Training, New York: Peter
Wyden.
Johnson, D. and Johnson, R. (1976) Learning Together and Alone,
Englewood Cliffs, NJ: Prentice-Hall.
Johnson, D. and Johnson, R. (1980) ‘The instructional use of cooperative
competitive and individualistic goal structures’, in H.Walberg (ed.),
Educational Environments and Effects, Berkeley, California:
McCutchan.
Johnson, D., Maruyama, G., Johnson, R., Nelson, D. and Shaw, L. (1981)
‘Effects of cooperative, competitive and individualistic goal structures in
achievement: a meta-analysis’. Psychological Bulletin 89: 47–62.
Jones, D. (1987) ‘Planning for progressivism: the changing primary school
in the Leicestershire authority during the Mason era 1947–71’, in
R.Lowe (ed.) The Changing Primary School, London: Falmer.
Kerry, T. and Sands, M. (1982) Handling Classroom Groups, University of
Nottingham, School of Education (mimeo).
Measor, L. and Woods, P. (1984) Changing Schools: Pupil Perspectives on
Transfer to a Comprehensive, Milton Keynes: Open University Press.
Mortimore, P., Sammons, P., Stoll, L., Lewis, D. and Ecob, R. (1987) The
Junior School Project, ILEA (Research and Statistics Branch), mimeo.
Mortimore, P., Sammons, P., Stoll, L., Lewis, D. and Ecob, R. (1988) School
Matters: The Junior Years, Wells: Open Books.
Nias, J. (1988) ‘Informal education in action: teachers’ accounts’, in
A.Blyth (ed.) Informal Primary Education Today: Essays and Studies,
Lewes: Falmer Press.
Plowden Report (1967) Children and their Primary Schools, 2 vols, Report
of the Central Advisory Council for Education in England, London:
HMSO.
Pollard, A. (1985) The Social World of the Primary School, London: Holt,
Rinehart & Winston.
Pollard, A. (ed.) (1987) Children and their Primary Schools, London:
Falmer Press.
Rowland, S. (1987) ‘An interpretative model of teaching and learning’, in
A. Pollard (ed.) Children and their Primary Schools, London: Falmer
Press.
Sharan, S. (1989 in press) (ed.) Co-operative Learning: Theory and
Research, New York: Praeger.
Slavin, R.E. (1983) Co-operative Learning, New York: Longman.
Slavin, R.E., Leavey, M. and Madden, N. (1984) ‘Combining cooperative
learning and individualised instruction: effects on student mathematics
achievements, attitudes and behaviours’, Elementary School Journal 84:
409–22.
Slavin, R.E. (1986) ‘Small group methods’, in M.Dunkin (ed.) The
International Encyclopaedia of Teaching and Teacher Education,
London: Pergamon.
Tizard, B., Blatchford, D., Burke, J., Farquhar, C. and Plewis, I. (1988)
Young Children at School in the Inner City, Hove and London: Lawrence
Erlbaum.
Walberg, H. and Anderson, G. (1968) ‘Classroom climate and individual
learning’. Journal of Educational Psychology 59: 414–9.
29
Maurice Galton
Wheldall, K., Morris, M., Vanehan, P. and Yin Yuk Ng (1981) ‘Rows v
tables: an example of the use of behavioural ecology in two classes of
eleven-year-old children’. Educational Psychology 1 (2): 171–83.
Woods. P. (ed.) (1980) Pupil Strategies, London: Croom Helm.
Yeomans, A. (1983) ‘Collaborative group work in primary schools, Britain
and the USA’, Durham and Newcastle Research Review X (51): 99–105.
30
Chapter three
Social interaction in the classroom
Jane French
Social interaction is clearly central to the concerns of this book. Any
developments that take place within the classroom context will do so as a
result of interaction of some sort. This makes a chapter on classroom
interaction necessary for any collection of readings concerned with the
social psychology of schooling.
Yet, while the pervasive nature of interactional processes makes this
chapter necessary, it also provides its author with a monumental
challenge. The range of issues that could be attended to is so vast that
selection is obviously required, yet the basis for that selection is never
likely to be readily apparent.
Jane French has concentrated on an analysis of verbal interaction, and
her intention has been to provide a practical introduction. The chapter
does not set out to provide a step by step guide of the ‘how to do it’
variety, but it does seek to raise basic issues that anyone would need to
attend to before starting to analyse some classroom interaction of their
own. In doing so it illustrates the complexities involved in making full
sense of even the most straightforward exchanges. Obviously it would be
possible to add into this a concern with many other aspects of interaction.
Non-verbal interaction (facial expression, gestures, posture, tone of
voice and so on) remain an important but much neglected area of
concern.
As will become apparent from a reading of the chapter itself, some
of Jane French’s own work involving the analysis of classroom
interaction has been concerned with an investigation of the nature of
gender effects. The reader will find some useful comparisons and
contrasts to be made with Chapter 11 by Croll and Moses, and is also
likely to find the discussion of ‘Discourse’ by Edwards (Chapter 4) a
useful compliment.
31
Jane French
Introduction
Over the past twenty years a great deal of research interest has focused
upon social interaction in classrooms. Perhaps not surprisingly, the area
has become something of a point of convergence for the social science
disciplines, with representation from sociology, psychology and
linguistics, as well as from the field of education more generally. This has
resulted in what now amounts to a considerable body of research, which,
at its best, provides us with a rich source of information as to what,
precisely, happens in classrooms.
In this chapter I shall draw, selectively, upon this material in seeking
to explore two central issues:
1. The nature of social interaction in classrooms.

2. The problems involved in conceptualizing and studying it.
I shall begin by setting the field within a historical context, taking into
account some of the social and disciplinary trends which led
researchers into classrooms in the first place. I shall then move on to
discuss selected aspects of classroom interaction: some formal
properties of teacher-pupil talk, the working out of social role identities
through interaction, and so on. Finally, I shall move on to consider
some of the difficulties involved in collecting, analysing and presenting
classroom interactional data.
The nature of social interaction in classrooms
Historical roots
Until the late 1960s very few researchers had mounted detailed studies of
the school classroom.1 In Britain there was a concentration on the
underachievement of pupils in the lower socio-economic groupings. In
America the overriding concern was the poor performance of black
pupils in relation to their white counterparts.
But on both sides of the Atlantic attention was focused not on the
school directly, but on what went into them, and what came out of them,
after x number of years’ compulsory schooling. As a number of
observers were later to note,2 the predominant approach adopted an
‘input-output’ model, with the school as a sort of unopened ‘black box’
between the two.
During the 1960s this emphasis began to change. As Delamont and
Hamilton (1976) note, student unrest during this period extended to a
profound dissatisfaction with the content and modes of working of
academic disciplines as well as with existing economic and social
32
arrangements. Within the social sciences this resulted in changes in

methodology as well as investigative focus.
In sociology, for example, the 1960s saw an emergence of interest in,
or, in some cases, a rediscovery of the works of figures such as Mead,
Schutz, Husserl, Simmel and, more latterly, Blumer and Garfinkel.3 The
end result of the renewed debate which followed appears to have been, at
the least, a heightened awareness of fundamental issues, and at the most,
a change of direction by significant numbers of practitioners.
One of the perspectives which aroused increased interest, and was
instrumental in leading sociologists into the classroom, was that of
symbolic interactionism.4 Its development over the years had already
tended toward distinct psychological and sociological adaptations (cf.
Rose 1962), and it has become even more fragmented since its
absorption into contemporary British sociology. But it is built upon
certain assumptions which were, and are, relevant to those involved in
the study of interaction in classrooms.
There is an emphasis, for example, on the need to study language and
interaction. Through interaction with, first, parents and family, then the
wider community, children learn the shared meanings of our culture, and,
over the course of time, become thinking, self-conscious human beings.
As Edwards and Westgate (1987:12) put it:
It is largely through talk that we develop our concepts of self,

as members of various social ‘worlds’ which can be brought
into focus and in which we can locate ourselves and
recognize the values, rights and obligations which permeate
them. As we listen and as we talk, we learn what it is
necessary to know, do and say in that area of social life or
that setting, and can display the competence necessary to be
accepted as a member.
Clearly the school classroom is a forum for this gradual process of

socialization, with children learning through interaction with both peers
and members of staff. But if the language of the classroom is to be
studied, the survey data, questionnaires and coding sheets traditionally
associated with the social sciences are clearly inadequate to the task: one
cannot mount a serious study of communication in the classroom merely
by asking teachers or pupils what they said or did, or by coding their
utterances into pre-set categories. Detailed observation and recording of
data, using modern audio and/or video recording equipment are required,
both to capture and to reproduce for analysis the complexities of
classroom communication. As Mehan warns, we must ‘[look] at the
window of language and not just through it’ (Mehan 1984:181, added
emphasis).
33
Jane French
Thus, because of both theoretical principles, and the practical

concerns of finding a research method adequate to the task in hand,
educationists from various social science disciplines began to use a range
of observational techniques which had hitherto been associated mainly
with anthropology. These ‘ethnographic’ methods included prolonged
observation and recording of the people under study, just as though they
were members of a foreign culture, immersion in their habitual practices
and customs, and, finally, the writing-up of research reports. Within
education Edwards and Furlong (1978) note that the term ‘ethnography’
was first adopted as a descriptive label for a (psychologically-oriented)
school-based study by Smith and Geoffrey in 1968. And Jackson’s
(1968) account of schooling in America, Hargreave’s (1967) and Lacey’s
(1970) detailed studies of British secondary schools, and Nash’s later
(1973) report, all stand as good examples of the ‘new’ approach, which is
summed up as follows by Edwards and Furlong (1978:53):
It was a wish to capture this view from the inside which drew
sociologists belatedly into classroom research. Using the
observational techniques of the anthropologist, they tried to
portray the realities of classroom life as they appeared to the
teachers and pupils involved.
Findings
So, having, albeit belatedly, gained entry to school classrooms, what
did researchers find there? The early, more broadly-focused, studies
provided an excellent foundation for later, more specific
investigations. Through their prolonged involvement and
painstaking observation, studies such as those of Hargreaves, and
Lacey indicated some of the areas which might fruitfully be exposed
to further inquiry. They also demonstrated, in a way which many of
the subsequent, more narrowly-focused studies have failed to do, the
enormous complexities of classroom life: the constellations of
factors which influence children’s relationships with one another,
with their teachers, and with education in a broad, overall sense. It is
with these complexities in mind that I shall try now to bring together
some of the findings from classroom research projects which pertain
specifically to the primary age child. Given the sheer volume of work
now available, my account will inevitably be highly selective. 5
For convenience of exposition, I shall consider the findings under the
subheadings of the Instructional context and the Social context It is an
artificial distinction, for in practice the two categories are mutually
constitutive and I must stress that they are separated here purely for ease
of expression.
34
The instructional context

Most people would agree that pupils attend school with a view to
acquiring specific forms of culturally approved knowledge. Thus it is
that lessons are structured in ways which facilitate the orderly
dissemination of that knowledge among comparatively large groups of
people. A number of recent research projects, representing various social
science disciplines, have considered both the transition of children from
the home to the school environment, and the forms of social interaction
which they can expect to find when they get there.
These studies have to be set against a background of concern
specifically for the linguistic and cognitive development of children from
the lower socio-economic groups, which permeated teacher-training from
the late 1960s onwards. At this time the work of Basil Bernstein became
highly influential in training colleges and departments and, subsequently,
in schools. As I, and others, have argued elsewhere (cf. Edwards 1986;
French 1987), the details of what was a contentious, ambitious and
complex theory were filtered down and presented in a simplified form in
the training institutions, with results which gave cause for concern.
It became apparent to many educationists that detailed information
was needed about the home environment, where primary socialization
occurs. In particular, there was a need to investigate the interaction which
actually occurs there, as opposed to the interaction which might be
supposed to occur there. By the same token, detailed material needed to
be collected from school classrooms.
Bearing the linguistic deficit theory in mind, it is interesting to note that
a number of studies remarked upon the lack of dissonance between the
home and the school, at least in terms of interactional structure. MacLure
and French (1981:237) note that from the point of view of children:
from whatever social background, there is little in the nature

of the interactional demands which will be made of them in
school that they will not already have become familiar with
at home, at the level of conversational structure.
Willes (1983) also remarks upon the ease with which children adapt to
the interactional requirements of being a ‘pupil’ rather than a ‘child’,
learning the norms of teacher-pupil communication within a matter of
weeks.
Yet these norms are different in terms of their distribution if not their
actual character from what most children have grown accustomed to in
the home. An obvious feature of classrooms as opposed to homes, for
example, is the fact that classrooms have many more occupants, and
potential conversational participants, than the average home!
35
Jane French
This affects the structure of the interaction in a number of ways. It

is clear, first, that teacher-pupil talk is informed by broadly two sets
of assumptions concerning what we might call the participants’
‘social identities’ and the ‘rights’ that they enjoy by virtue of these
identities.
Adult-child
On the one hand, the teacher is an adult, and the pupil is a child. And most
pupils (as well as their teachers!) will have learned and come to accept
long before they come to school that in conversation with adults, children
have restricted rights (cf, Sacks 1966; Speier 1976). Specifically, adults
control conversations in the sense that vested in their identities as adults
are the rights to, for example, select topics, evaluate children’s
contributions, and bring talk to a close without the agreement of the
other, child, party. Speier suggests that this common orientation by both
children and adults is predicated upon the ‘classical’ view of children as
pre- and in- competent adults. And, of course, this assumption is
repeatedly, though often implicity, relied upon within the school
classroom.
Given the comparative lack of autonomy of children in out-of-school
encounters with adults, it would not be surprising to find a yet more rigid
adherence to these practices in the school setting, since the social identity
‘teacher’ combines the subcategories ‘adult’ plus ‘instructor’, with all its
associated rights. Similarly, the identity ‘pupil’ combines ‘child’ and
‘instructee’.
Multi-party talk
Within a subdiscipline of sociology known as conversation analysis (cf.
Atkinson and Heritage 1984; Heritage 1984), attention has been paid to
the means by which orderly conversations are made possible because
participants share an unspoken adherence to basic and specifiable turn-
taking procedures (Sacks et al. 1974). Clearly, while conversations
between only two people present few problems as to who shall be the
next speaker, multi-party conversation, as in the classroom, can
potentially be problematic. That is, the presence of three or more
participants opens up the possibility of competition for turns, the
presence of four or more for the setting up of subconversations in parallel
with one another, and so on. In addition, participants in informal
conversation at least may choose whether and when they speak: that is,
they may ‘self-select’. They need not wait for another participant to
select or call upon them to take a turn, as frequently happens in the
classroom (cf. McHoul 1978).
36
Managing the class

The sets of assumptions concerning (1) adult-child identity, and (2) the
organization of multiparty conversation come together in the context of the
classroom in ways which bear upon the orderly structure of teacher-pupil
interaction. For example, many references are made in the literature to the
fact that, in classrooms, teachers do most of the talking and that they ask
children a great number of questions (cf. Flanders 1966). Indeed, as Tizard
and Hughes (1984:249–50) observe, this lies in direct contrast to most
homes, even those which educationists might categorize as the least
enabling: ‘Simply by being around their mothers, talking, arguing, and
endlessly asking questions … children … (are) provided with large
amounts of information relevant to growing up in our culture’.
In order successfully to ‘manage’ the larger-scale conversation of the
classroom, teachers find it useful to organize what is actually multiparty
interaction into interaction between two parties: the teacher, representing
one party, and the class collectively as the other (cf, Payne and Hustler
1980).The techniques involved in effecting this transformation involve
the teacher in maintaining a delicate balance between under- and over-
participation by pupils, which in turn involves ensuring that: (a) in the
main, one speaker speaks at a time, and (b) as many pupils as possible are
involved in the answering of questions.
One of the ways in which experienced teachers perform this balancing
act is through quick-fire question-answer-evaluation sequences which
serve to involve large numbers of pupils, and are paced so as to keep their
attention (cf. Hammersley 1974; Sinclair and Coulthard 1975; Mehan
1979, among others).
Example 1 below illustrates the way in which a series of such
sequences might work:
Example 1
1. Teacher: How many people here can tell me what a square is?
2. Pupils: (Raise hands. Teacher scans class and selects child
with hand up.)
3. Teacher: Alex?
4. Alex: It’s got four sides.
5. Teacher: Good boy. And what are those sides like?
6. Pupils: (Raise hands. Teacher again scans class and selects
child with hand up.)
7. Teacher: Becky?
8. Becky: They’re straight
9. Teacher: They’re straight. Yes, they’re not all wiggly and bendy
are they? They’re straight. And anything else? (scans
class) Yes, Katy.
37
Jane French
10. Katy: They’re the same length.

11. Teacher: Yes, good girl! They’re all the same length. That’s the
important thing, isn’t it? They’re all the same length.
In this straightforward example we can see clearly the way in which the
teacher is able to involve several children, who all have one short turn
each. The teacher first sets up a pool of possible respondents by asking
those who know the answer to raise their hands. This saves her from
asking a child who does not know or is unwilling to answer, which may
slow the whole lesson down, and involve her in repeating the question,
rephrasing it, and so on (cf. McHoul 1978). Her questions are also
designed so that pupils are able to answer with a short phrase, rather than
a long, possibly rambling, series of phrases. The lesson is thus able to
move along at a smart pace, involving several children, and, importantly,
eliciting from them the information the teacher wants.
Of course, questions are not always designed so as to facilitate short
answers. During ‘news’ sessions, for example, the teacher may select a
pupil or pupils to speak at greater length (cf. French and French 1984),
while realizing that the rest of the class may become bored, and taking
steps to prevent or remedy this. In addition, lessons may not always go
so smoothly. There may be problems, even in a short, ‘rapid fire’
session, so that the attention of one or more pupils may lapse, children
may start whispering to one another, or one child may start calling out
and quickly be followed by others. Experienced teachers sense
potential difficulties, and can be seen to deploy a variety of techniques
both to prevent their occurrence, and to repair them when they have
happened. Payne (1976), for example, notes that collective forms of
address are used, thereby, as he puts it, encouraging them to act ‘as a
unit, making their individual fates collectively interdependent’ (Payne
1976:54).
In addition, a story may be prefaced ‘Now listen to this story
because then I’m going to ask you some questions about it’ (cf. Payne
and Hustler 1981), or children may be asked to ‘look this way’, or to
display in one way or another that they are paying attention. If these
methods fail to obtain the desired result, teachers may move on to what
have been termed ‘problem statements’ (cf. French 1985; French and
French 1986; French and Peskett 1986): that is, they may state a
problem, couched at first in generally addressed terms (for example,
‘some people are still not listening’). Pupils are left to make the (albeit
often obvious) inference as to what is required, and to modify their
behaviour accordingly.
Alternatively, teachers may issue ‘directives’ (cf. French 1985)
which explicitly instruct pupils on how to behave, (for example, ‘Sit
down please’, ‘Don’t call out’, ‘Put your hands up’). Or they may use
38
question-related techniques to ‘trap’ suspected non-attenders by

deliberately addressing questions to them, or by recycling already
answered questions so that the non-attender is shown to be precisely
that (Teacher: What season is it John? John: Spring. Teacher: Right.
Mark, what season is it?).
The social context

But, of course, lessons are not simply strings of empty forms. Teachers
and pupils talk about things. During the course of lessons, knowledge is
transmitted and, one hopes, acquired, via some of the practices
described above, as well as through listening, reading, observing and
doing. In addition, messages may be passed, through the interaction
and largely unconsciously, about a range of factors known to influence
both the school performance and self-image of the individual pupil.
These may include the personal relationship of teacher to pupils,
attitudes (displayed by teacher and/or pupils) towards, say, different
races or to the sexes, or particular cultural, subcultural or religious
beliefs (again by teacher and/or pupils). These sorts of factors typically
underlie the interaction, informing it in subtle and sometimes non-
subtle ways, and moving in and out of focus as topics shift and
activities change.
Some teachers, for example, may find that their perceptions of pupils
are affected by some of the linguistically related issues discussed earlier.
Like other members of society, they may respond differently to people
who speak with a pronounced regional accent, or who use local dialect
forms. The increased provision of pre- and in-service courses in
language and education has helped in this connection, perhaps
particularly at the primary level. Most teachers are now aware that, say,
South Yorkshire English or Afro-Caribbean English are as rule-governed
and semantically complex as Home Counties or BBC English. By the
same token, a local accent is simply a local accent, and not an indication
per se of a deficient intellect.
However, and as the Cox Report (DES 1988) has recently noted,
judgements are still formed on the basis of accent and/or dialect Marked
local accents or use of dialect forms still convey the message, in class
conscious Britain, that a child is probably ‘working class’. And this
social identification tag may carry with it a number of expectations
concerning the child’s home life, the attitudes of his/her parents to
schooling, and, consequently, the child’s school performance. As
Edwards (1986) points out, the legacy of the linguistic deficit theory may
well have brought self-fulfilling prophecies back into play in the
classroom.
Dimensions of social identity other than socio-economic class
membership are also displayed in classroom interaction. Over recent
39
Jane French
years, for example, a great deal of attention has been paid to the ways in
which the sexes signal gender identity through language as well as
through the more obvious channels of mode of dress, hairstyle and so on
(cf. Thorne and Henley 1975; Fishman 1978).
We know that the classroom acts as a forum where children both
display and develop their competence in this capacity. It is well
established that during the pre-school years, parents and other significant
adults respond differently to boys and girls (cf. Block 1985; Hodgeon
1985; Grabrucker 1988). Thus, by the time they reach the age of
compulsory schooling, most children have a good idea of gender
identity, and the forms of behaviour which adults and other children find
appropriate.
In a study conducted at infant school level, French and French (1986)
found boys to be given and to take significantly more speaking turns than
girls in classroom discussions. They also found that boys were more
often the subject of teachers’ ‘control talk’ than girls. And interestingly,
in a subcategory of control talk known as ‘directives’, boys, but not girls,
were more likely to have appended some sort of explanation (for
example, ‘Sit down David, so that the people behind you can see’). I
discuss this finding at greater length in a later section of this chapter.
At the upper junior level, French and French (1984) analysed a single
lesson in some detail, showing the ways in which a particular, dominant
subgroup of boys was able to take up a disproportionately large chunk of
the time available. Some of these included calling out, making
extraordinary or ‘newsworthy’ claims so that the teacher invited them to
speak at greater than average length, or exploiting opportunities provided
by the teacher to contribute by being ‘different’ (for example, being the
only child who thinks that Maths is an unimportant subject!).
These sorts of findings have also obtained in studies elsewhere in the
world (cf. Good, et al. 1973; Schools Commission, Australia 1984;
Sadker and Sadker 1985; Morgan and Dunn 1988), and at all levels of the
educational system (cf. Stanworth 1983; Croll 1985; French and French
1986; Hodgeon 1985.
As Swan and Graddol (1988:63–4) points out, there seems to be a
‘consensus’ among girls, boys and teachers as to what constitutes normal
interactional behaviour for each sex, whereby:
an unequal distribution of talk is seen as normal. In

particular, girls seem to have learnt to expect a lower
participation level than boys, and boys seem to have learned
that their fair share is a larger one. These are expectations
that are brought to school by all participants, since such
inequalities in the distribution of talk are commonplace
amongst adults.
40
It is thus that the so-called ‘hidden curriculum’ of education is

transmitted: the signalling of status, the communication of implicit
approval or disapproval, the patterns of behaviour which we treat as
natural and, more often than not, do not even notice let alone
challenge. But whether or not we want to make changes, it is in
interaction that the transmission of social ‘messages’ is to be found.
For students or practising teachers who wish to consider any of these
issues, or to examine the group dynamics of a particular class or
group of pupils, the place to begin is in the practical world of
everyday interaction.
Problems in conceptualizing and studying classroom interaction
Collection of data
The widespread availability of audio and, in some cases, video recording
equipment nowadays has meant that many prospective and practising
teachers are able to try their hand at classroom interaction analysis.
While this can be an extremely valuable and satisfying exercise, directed
to a variety of practical purposes, it can also bring home to would-be
analysts some of the potential pitfalls and difficulties in both recording
interaction, and interpreting what one finds.
In the first place, and at a practical level, classrooms are notoriously
difficult places in which to record. They are usually comparatively
large rooms, with children moving about between activities, and a high
level of background noise. The difficulties of making good quality
recordings which capture the flavour of lessons, particularly where
children are spread out around the room engaged in group work, are
considerable. It is for these practical reasons that professional
researchers frequently focus narrowly on, say, teacher-class
discussions or on single groups of children working. This goes some
way to explaining why we know comparatively little about certain
aspects of classroom life, and why some areas and issues remain
unexplored and largely inaccessible.
Analysis and interpretation of data

The problems do not disappear, either, once data have been collected,
for the process of analysis and the subsequent presentation of materials
give rise to a number of issues concerning the status and reliability of
findings. Obviously the practising teacher who wants to use classroom
interaction analysis for essentially practical purposes cannot be
expected to explore these issues in the same depth and detail as, say, a
full-time post-graduate student. But if they are ignored altogether, then
the quality of the individual’s work will suffer, and its value will be
41
Jane French
questionable. I would also add that the capacity of that individual to be

critical in relation to the work of him/herself and others will not
properly have been developed.
Over-interpretation of data
One of the most common problems which arises, especially for the
novice analyst, is the tendency to ascribe more significance than is,
strictly speaking, justified, to impute unwarranted motives and
intentions to participants, and then to put one’s understandings
forward, implicitly or otherwise, as the only possible ways of
interpreting the data. It is important to balance evaluative or
prescriptive interests with analytic rigour and integrity, and to advise
caution in the interpretation of what is found. An example may help to
make this point more clearly.
As noted at an earlier point in the present chapter, French and French
(1986) found that in a subset of teachers’ talk known as ‘control talk’,
boys were more likely to have the rules and norms of classroom conduct
explained to them than girls. Why should this be? They suggested four
possible explanations (there may be more):
(i) First it may be connected with same-sex identification. That

is, that teachers assume girl pupils, as females like
themselves, know the rules and the reasons behind them.6
(ii) It may also be that girls are assumed to be more socially
adept and mature, and that rules need not be explained. Boys,
on the other hand, may be thought to need repetitive
restatement.
(iii) Girls may be seen as more compliant. On this basis teachers
may assume that they may safely dispense with
explanations: the girls will comply anyway. Boys, however,
may question the teacher’s authority and, therefore, in the
cause of expediency, explanations are appended.
(iv) The phenomenon may constitute evidence that teachers
prefer boys and consider them more important than girls.
Simply, boys are worthy of an explanation whereas girls
are not.
(French and French 1986:20–1)
As we go on to note, the possible interpretations listed above were all

suggested by teachers with whom the issue was discussed. But it is
important to recognize that although all of them are plausible
interpretations, retrospectively assigned, no one of them necessarily
provides a definitive account. Indeed, it is not always possible to
provide as definitive an account as one would perhaps wish: classroom
42
interaction analysis sometimes throws up intriguing findings which are

difficult to explain. Because of a variety of factors, such as our
common-sense knowledge of schools and teachers, our overall cultural
competence or our individual biographies, we may be able to come up
with explanations, but it is important to bear in mind that they are
generally, as in the case of the example quoted above, among several
plausible possibilities. They should not be given a privileged or
elevated status, or presented as ‘the truth’ about this or that
phenomenon. Rather, the limitations of this type of work should be
acknowledged.
Coding of interactional data

The coding of interactional data is a perennial problem for analysts. In
my view, the most detailed and elaborate scheme cannot be 100 per cent
accurate, and I regard the categorizations made during the course of a
research study as a useful shorthand form of representation rather than an
impressive ‘scientific’ system.
Again, an example may help to clarify the point. During the course of
interaction, in the classroom or elsewhere, participants might sometimes
interrupt one another. We all ‘know’ in one sense when we are
interrupting someone else, or when someone else is interrupting us, but
how as analysts do we decide what, precisely, constitutes an
interruption? A remark intended by a speaker as an ‘aside’ comment may
be interpreted by a hearer as an interruption. Alternatively, a speaker may
mistime an utterance so that his/her speech sounds like an interruption
when it was not intended as such.
It is therefore clear that simple definitions of phenomena such as
interruptions (that is, ‘interruption’ = overlapping talk) are
unsatisfactory. It seems, moreover, that in interpreting talk as
interruptive, conversational participants may use variable clusters of cues
including prosodic features of speech (for example, pitch, height,
loudness) (cf. French and Local 1983) and assessments as to the
supportive or non-supportive character of the talk of the incoming
speaker (cf. Bennett 1980). Students of interaction have long recognized
that one cannot transform such complexities into coding criteria, and that
the exercise of coding is always a matter, essentially, of common sense,
based on the analyst’s competence as a cultural member. As Sinclair and
Coulthard (1975) admit, there are always and always will be marginal
cases and arbitrary choices.
Some descriptive issues

Finally in this section I shall discuss some descriptive issues which arise
during the course of any piece of research, and which merit serious
consideration.
43
Jane French
A number of analysts have drawn attention to the processes of

selection which are inevitably engendered during the course of
producing any descriptive account (cf. Sacks 1963).7 In transforming
experience into a written version of events, one is enabled only ever to
record a partial version of events, which has undergone two basic forms
of selection. First, the words used have been selected from what is, in
theory, an indefinitely extendable list of possible descriptions. Second,
and more important from the point of view of the present paper, the
events to be described have been selected for recording at the cost of
excluding others.
This has traditionally been an acute problem for those engaged in
ethnographic work. Where data consist of observational field-notes,
inevitably much is lost in that one cannot capture and recreate all that
went on in the setting. The composition of field-notes constitutes a first
process of ‘filtering’ from raw data to eventual account, which sets up a
distance between the actual events giving rise to the account, and the
readers of the account And this in turn can mean that the only
interpretation of events available to readers is the analyst’s interpretation.
The reader is not able to challenge the analyst’s view because he/she
does not have access to the material necessary to do so. This then raises
questions as to the status of ethnographic accounts. Is the account purely
a subjective view, or does it have aspirations to a more detached and
‘objective’ status? If so, how can the inevitable withholding of original
data be justified?
The advent of widely-available audio and video recording
facilities represents an important advance. The capacity of
technology to reproduce data without significant loss means that
analyst and audience (whether on a professional level, or in the roles
of teacher-analyst and tutor, or lecturer and students) can have access
to the same materials, and may re-run and discuss particular aspects
of data as necessary.
However, the danger of selective presentation remains. There is
scarcely more merit in presenting chosen extracts from tape-recorded
materials than from field-notes. It is, in fact, in this connection that
writers such as Mehan (1979) have spoken disparagingly of researchers
presenting ‘tidbits’ of audio-visual materials or transcripts. In such cases,
researchers could seek to ‘prove’ whatever they liked, within reason,
about classroom interaction.
One solution to this problem is carefully to record data, and to
allow audiences the fullest possible access to materials. In the
context of seminars or lectures, this may involve the presentation of
audio and/or video recordings together with verbatim transcriptions.
In the case of written presentations, verbatim transcriptions have to
suffice.
44
Conclusion
In this paper, I have sought to explore a number of issues pertaining to
social interaction in classrooms. These have ranged from discussion of
the complex character of classroom communication to the problems
involved in collecting and analysing data. In my view they are important
issues, for they affect classroom practice in several respects.
First, knowledge of and about areas such as social interaction
informs and underlies one’s approach to a class. It sensitizes and
develops the critical faculties in much the same way that musical or
literary analysis sharpens the understanding and appreciation of works
of art And once the seeds of knowledge have been sown, the student is
enabled to ‘see’ teacher-pupil interaction in a different light: one
begins to notice the mechanisms which lie beneath the surface of
lessons, to appreciate their variety and the skills of the teachers who,
largely unconsciously, deploy them.
For the beginner in particular, this can be an eye-opening experience.
Although many training institutions now use video, it is often the case
that students are instructed simply to watch, without being told
specifically what to watch for. Yet many novice teachers crave practical
knowledge. Indeed, one of the chief criticisms directed at teacher-
trainers is that too much time is spent on theory and not enough on
practical skills. While I am emphatically not of the school of thought
which believes educational theory to be a useless irrelevance, it does
seem to me that there is much to be gained from observing experienced
practitioners. But in order to observe, in the strict sense of that term, one
must first be taught both where and how to look. The student who has
even these skills is at an immediate advantage.
Notes
1 Although see Waller (1932)
2 See particularly, Delamont and Hamilton (1976), Hammersley and Woods
(1976), and, for an American point of view, Mehan (1979)
3 See Garfinkel (1967), Blumer (1969) and, for discussions of the influence
of these and the other authors mentioned, see Rock (1979) and Leiter
(1980).
4 For a more detailed exposition of symbolic interactionist theory see Rose
(1962) and Rock (1979).
5 The selective focus of the paper has meant that, inevitably, much work of
value has been left out. I would draw readers’ attention in particular to the
considerable contribution of those working within linguistics and its
subdisciplines. Texts of especial relevance include Crystal (1976), Stubbs
(1983) (2nd edition) and Stubbs and Hillier (1983).
6 The teachers in our sample were all female.
7 For further discussion of these issues see Atkinson and Drew (1979) and
Heritage (1984).
45
Jane French
References
Atkinson, J.M. and Drew, P. (1979) Order in Court, London: Macmillan.
Atkinson, J.M. and Heritage, J.C. (1984) Structures of Social Action: Studies in
Conversation Analysis, Cambridge: Cambridge University Press.
Bennet, A. (1980) Interruptions and the Interpretation of Conversation, Tuscon:
University of Arizona, mimeo.
Block, J. (1985) Sex Role Identity and Ego Development, San Francisco: Jossey
Bass.
Blumer, H. (1969) Symbolic Interactionism: Perspective and Method, Englewood
Cliffs: Prentice-Hall.
Croll, P. (1985) ‘Teacher interaction with individual male and female pupils in
junior-age classrooms’. Educational Research, 27: 220–3.
Crystal, D. (1976) Child Language, Learning and Linguistics: an Overview
for the Teaching and Therapeutic Professions, London: Edward Arnold.
Delamont, S. and Hamilton, D. (1976) ‘Classroom research: a critique and a new
approach’, in M.Stubbs and S.Delamont (eds) Explorations in Classroom
Observations, New York: Wiley.
Department of Education and Science (DES) (1988) English for Ages 5–11:
Proposals of the Secretary of State for Education and Science and the
Secretary of State for Wales, (The Cox Report), London: DES.
Edwards, A.D. and Furlong, V.J. (1978) The Language of Teaching, London:
Heinemann.
Edwards, A.D. and Westgate, D.P.G. (1987) Investigating Classroom Talk,
London: Falmer.
Edwards, J. (1986) ‘Language and educational disadvantage: the persistence of
linguistic “deficit” theory’, in K.Durkin (ed.), Language Development in the
School Years, London: Croom Helm.
Fishman, P. (1978) ‘Interaction: the work women do’. Social Problems, 25 (4):
397–406.
Flanders, N.A. (1966) Interaction Analysis in the Classroom: a Manual for
Observers, Michigan, USA: School of Education, University of
Michigan.
French, J. (1987) ‘Language in the primary classroom’, in S.Delamont (ed), The
Primary School Teacher, London: Falmer.
French, J. and French, P. (1984) ‘Gender imbalances in the primary classroom: an
interactional account’. Educational Research, 26 (2): 127–36.
French, J. and French, P. (1986) Gender Imbalances in Infant School Classroom
Interaction: Final Report to the Equal Opportunities Commission, The
College of Ripon and York St John, Lord Mayor’s Walk, York.
French, P. (1985) ‘Problem statements and directives: some aspects of control
sequences in infant school classroom interaction’, in J.C.P.Auer and A.di
Luzio (eds) Ergebrisse und Methoden in Moderner Sprachen Wissenschaft,
Koln: Narr.
French, P. and Local, J. (1983) ‘Turn-Competitive incomings’, Journal of
Pragmatics, 7 (1): 17–38.
French, P. and Peskett, R. (1986) ‘Control instructions in the infant classroom’,
Educational Research 28 (3): 210–19.
Garfinkel, H. (1967) Studies in Ethnomethodology, Englewood Cliffs, NJ:
Prentice-Hall.
Good, T.L., Sykes, J.N. and Brophy, J.E. (1973) ‘Effects of teacher sex and
student sex on classroom interaction’. Journal of Educational Psychology, 65:
1–27.
46
Grabrucker, M. (1988) There’s a Good Girl: Gender Stereotyping in the First

Three Years of Life, a Diary, London: The Women’s Press.
Hammersley, M, (1974) ‘The organization of pupil participation’, Sociological
Review, 1: 355–67.
Hammersley, M. and Woods, P. (eds) (1976) The Process of Schooling, London:
Routledge & Kegan Paul with The Open University Press.
Hargreaves, D.H. (1967) Social Relations in a Secondary School, London:
Routledge & Kegan Paul.
Heritage, J. (1984) Garfinkel and Ethnomethodology, Cambridge: Polity Press.
Hodgeon, J. (1985) A Woman’s World: Report on a Project in Cleveland
Nurseries on Sex Differentiation in the Early Years, Cleveland Education
Committee..
Jackson, P.W. (1968) Life in Classrooms, New York: Holt, Rinehart & Winston.
Lacey, V.C. (1970) Hightown Grammar, Manchester: Manchester University
Press.
Leiter, K. (1980) A Primer of Ethnomethodology, London: Oxford University Press.
McHoul, A. (1978) ‘The organization of turns at formal talk in the classroom’,
Language in Society, 7: 183–213.
MacLure, M. and French, P. (1981) ‘A comparison of talk at home and at school’,
in G.Wells (ed.) Learning Through Interaction: the study of language
development, London: Cambridge University Press.
Mehan, H. (1979) Learning Lessons: Social Organization in the Classroom,
Cambridge, Mass.: Harvard University Press.
Mehan, H. (1984) ‘Language and Schooling’, Sociology of Education, 57:.
Morgan, V. and Dunn, S. (1988) ‘Chameleons in the classroom: visible and
invisible children in nursery and infant classrooms’, Educational Review, 40
(1): 3–12.
Nash, R. (1973) Classrooms Observed, London: Routledge & Kegan Paul.
Payne, G.C.F. (1976) ‘Making a lesson happen: an ethnomethodological
analysis’, in M.Hammersley and P.Woods (eds), The Process of Schooling,
London: Routledge & Kegan Paul with The Open University Press.
Payne, G.C.F. and Hustler, D. (1980) ‘Teaching the class: the practical
management of a cohort’, British Journal of the Sociology of Education, 1 (1):
49–66.
Payne, G.C.F. and Hustler, D. (1981) ‘Stories and storytime in an infant
classroom’, in P.French and M.MacLure (eds) Adult-Child Conversation,
London: Croom Helm.
Rock, P. (1979) The Making of Symbolic Interactionism, London: Macmillan.
Rose, A.M. (ed.) (1962) Human Behaviour and Social Processes, London:
Sacks, H. (1963) ‘Sociological description’, Berkeley Journal of Sociology, 8: 1–
16.
Sacks, H. (1966) Los Angeles: University of California, Lectures, spring
(mimeo).
Sacks, H., Schegloff, E, and Jefferson, G. (1974) ‘A simple systematics for the
turn-taking organization of conversation ’, Language, 50: 696–735.
Sadker, D. and Sadker, M. (1985) ‘Is the OK classroom OK?’ Phi Delta Kappan,
January, pp. 358–61.
Schools Commission, Australia (1984) Girls and Tomorrow: the Challenge for
Schools, Canberra: AGPS.
Sinclair, J.McH. and Coulthard, R.M. (1975) Towards an Analysis of Discourse:
the English used by Teachers and Pupils, London: Oxford University Press.
47
Jane French
Smith, L. and Geoffrey, W. (1968) The Complexities of an Urban Classroom,

New York: Holt, Rinehart & Winston.
Speier, M. (1976) ‘The child as conversationalist’, in M.Hammersley and P.
Woods (eds) The Process of Schooling, London: Routledge & Kegan Paul with
The Open University Press.
Stanworth, M, (1983) Gender and Schooling: A Study of Sexual Divisions in the
Classroom, London: Hutchinson, with The Explorations in Feminism
Collective.
Stubbs, M. (1983) Language, Schools and Classrooms, 2nd edn, London:
Methuen.
Stubbs, M. and Hillier, H. (eds) (1983) Readings on Language, Schools and
Classrooms, London: Methuen.
Swann, M. and Graddol, D. (1988) ‘Gender inequalities in classroom talk’,
English in Education, 22 (1): 48–65.
Thorne, B. and Henley, N. (eds) (1975) Language and Sex: Difference and
Dominance, Rowley, Mass.: Newbury House.
Tizard, B. and Hughes, M, (1984) Young Children Learning: Talking and
Thinking at Home and at School, London: Fontana.
Waller, W. (1932) The Sociology of Teaching, New York: John Wiley & Sons.
Willes, M.J. (1983) Children into Pupils: A Study of Language in Early
Schooling, London: Routledge & Kegan Paul.
48
Chapter four
Classroom discourse and classroom

knowledge
Derek Edwards
‘Classroom discourse and classroom knowledge’ carries on from the
previous chapters and adds new and critical focus to classroom process.
From the previous chapters, Edwards acknowledges that there is an
overwhelming tendency to teach the ‘individual’ in the classroom and
that there are a number of ‘hidden agendas’ within the classroom
concerning the structure of knowledge and power relationships between
teacher and pupil. Even though the classroom may be structured in
groups or draw upon groupings of children for economies of learning,
teachers and curricula tend to focus on the individual as the point for, and
point of, learning.
In his study Edwards draws upon a number of conversational extracts
where learning takes place and reinterprets learning as a ‘discourse’
between people. The process of generating knowledge is a shared
experience, whether between teacher and pupil or between pupils
themselves. Reinterpretation of learning criticizes individualized
developmental psychological theories that have informed the
construction of many of the curricula now present in primary school
classrooms, especially as the theories have characterized learning as an
individual creation. It is interesting that the theorist most often cited as
being responsible for individualization (Piaget) in the child-centred
classroom is also the person who provides the most explicit rationale for
the reinterpretation of learning and development as a shared process, and
we are also made aware of the importance of Vygotsky. Classroom
processes are seen as providing a social context for development as well
as being formative of that process. The reader is made vitally aware that
the relationships arranged within the classroom will direct the type and
quality of knowledge engendered.
One final aspect of this chapter is the role played by the curriculum.
Edwards has purposively chosen to research science and mathematics
curricula. In investigating the curricula in general, he makes the point
that scientific advances do not usually take place in a vacuum, but in a
49
Derek Edwards
social (and usually collective) context. Actual extracts that Edwards

provides show that at least two types of curricula-related learning can
take place in the classroom, each dependent on the dialogue and quality
of relationship in which the child participates: a socializational learning
in which pupils apprentice themselves to the teacher’s knowledge; and
argumentation of self-constructed knowledge which takes place with
peers. Socializational knowledge may be related to the imposition of the
National Curriculum in which the end-products of learning become more
important than the actual process. Argumentation is more clearly linked
to understanding and process. In either case the teacher is the person
responsible for structuring the classroom context so as to promote one or
both types of learning.
Introduction
Educationists sometimes, though not always optimistically, turn to
psychology for help in understanding the processes of teaching and
learning. In particular, Piaget’s classic studies of the development of
children’s thinking have helped to shape educational practice. The older,
‘traditional’, primary classrooms, all chalk and talk, discipline and rote
learning, have largely given way to child-centred activity. Children are
encouraged to develop their own understandings of things, to work
individually or in small groups, learning from their own practical
experience, rather than having ready-made conclusions thrust upon their
unwilling ears. The legacy of this influence is a notion of children’s
learning that is highly individualistic. It stresses the importance of
experience and action, playing down the significance of talk, of teaching,
and of the ready-made curriculum. While social psychology might be the
obvious place to look for insights into interpersonal relationships,
patterns of communication, group structure and leadership, roles and
friendships, and so on, it is to the psychology of individuals (learning and
motivation, IQ tests, cognitive development, and so on) that we turn for
insights into the processes of knowledge and learning, the development
of conceptual understanding itself. Social factors may be thought to
impinge upon such things social class, social context, social prejudice,
parental encouragement, teaching styles, and so on, but the acquisition of
knowledge is itself an individual process. Children learn from action and
experience, process information, develop ‘mental models’ or theories of
the world, to make sense of what they perceive, or of what happens when
they act upon the world. At least, this is now the established view, the one
that appears to prevail amongst teachers and the makers of educational
policy (see, for example, the recent policy document issued by the
Engineering Council and the Standing Conference on Schools’ Science
and Technology 1985).
50
Classroom discourse and classroom knowledge
This chapter will take an opposing view. It will draw upon some new
developments in the social psychology of ‘discourse’ and ‘rhetoric’,
and in the study of classroom education. It will look at the educational
process from a social perspective, taking the view that the development
of understanding is a communicative accomplishment, embodied in
classroom discourse. Education is presented here as the development
between teachers and pupils of shared understandings, shared
experiences and procedures, and a shared conceptual vocabulary.
Empirical studies depict classroom discourse as oriented mainly
towards the achievement of a teacher-dominated consensus of
understanding. We shall argue that this embodies an unnecessarily
limited sense of shared knowledge. A different emphasis needs to be
placed upon the importance of argument, disagreement, justification
and criticism—in other words, a ‘rhetorical’ approach to shared
knowledge. In view of the fact that scientific knowledge is often taken
to be the most solid, best established, least vulnerable knowledge that
can be taught, we shall concentrate our analysis upon science
education, with groups of 9-year olds.
Towards a social psychology of educational knowledge

In the world of education, the currently dominant psychologies of
conceptual development are individualistic. Piaget’s major legacy has
been the assumption that children are learners from experience,
actively constructing their own understandings of the world, via how
the world responds when it is pushed, pulled, grasped, poured, thrown
and generally manipulated. The teacher’s role is that of a supervisory
provider of suitable materials and learning opportunities. Since the late
1960s, following the recommendations of such influential publications
as the Plowden Report (1967), and the Nuffield Mathematics Project
(Nuffield Foundation 1967), the Piagetian notion of individualistic
learning-by-doing has been the major psychological underpinning for
the ‘progressive’ movement in education. The Nuffield
recommendation was embodied in the aphorism: ‘I hear and I forget. I
see and I remember. I do and I understand’. Teachers’ words are for
hearing and forgetting, unless they relate closely to the real source of
understanding —practical activity. Piaget himself had declared: ‘Each
time one prematurely teaches a child something he could have learned
for himself, the child is kept from inventing it and consequently from
understanding it completely’ (1970:715). Similarly, in the recent
educational policy document issued jointly by the Engineering Council
and the Standing Conference on Schools’ Science and Technology
(1985), we are urged to accept that problem-solving projects are
particularly well suited to ‘the ethos of the primary school’, because
51
Derek Edwards
‘the teaching is usually child centred and many learning experiences

guided by the teacher start from activities with which the child is
already familiar’ (p. 8).
However, in the world of academic psychology Piaget’s conception of
the child is no longer predominant. Critiques of the Piagetian orthodoxy
have appeared (for example, Donaldson 1978; Russell 1978;
Butterworth and Light 1982; Walkerdine 1982; Carey 1985; Edwards
and Mercer 1987). The view of the child as a ‘lone organism’ (Bruner
1986), constructing a succession of general models of the world as each
new stage is mastered, is giving way to two new approaches. One of these
is even more individualistic than Piaget’s. It derives from the new field of
cognitive science, and seeks the origins of mature conceptual thinking in
ever younger children, even to the point of denying that any major
conceptual changes take place at all, and that the main principles of
human conceptual thought are pre-established in early infancy, and
perhaps innately (for example, Fodor 1975; Carey 1985; Keil 1986). In
fact this approach is new in surface details only; it is the doctrine of
innate knowledge, at least as old as Plato, and essential to the rationalist
philosophical tradition of Descartes and Kant.
The second ‘new’ approach is also at least as old as Piaget’s, but is
only now becoming a major influence in British and American
developmental psychology. This is the work of the Soviet psychologist
Leo Vygotsky and his followers, which emphasizes the social-cultural,
linguistic origins of conceptual thinking, the importance of
communicative social interaction in learning, and the roles of education
and literacy in shaping what are often assumed to be natural, rational,
objectively scientific modes of adult thought. In Vygotsky’s perspective
(see Vygotsky 1987 and Wertsch 1985; Wood 1988) mind is socialized
and shaped through teaching and language. Education is formative of
conceptual thought, rather than merely a practical context in which
psychological theories and findings, derived from the experimental
laboratory, can be ‘applied’. Conceptual thought is derived from
dialogue, and so must follow the rules and categories of discourse, of
communicated symbols, and written text.
For example, in an extended treatment of this notion, David Olson
(1984) has argued (controversially) that it is the invention and
widespread use of written forms of language that has resulted in the
development of rationality itself, of logical reasoning, of formal
education, and of scientific thought. And it is the acquisition of literacy
that is the key to how individuals and societies, in the modern world, also
acquire those powerful modes of thought. The mere fact of having
written text available to us for rereading, and repeated scrutiny, not only
facilitates the creation of large cultures of formal knowledge and
procedure (science, literature and education), but also encourages an
52
analytic perspective on language and knowledge that differs

fundamentally from common sense. We learn to isolate, examine and test
statements for their truth value (science and philosophy), and their
interrelationships. The form of language dictates the form of thought,
such that ‘what we call “intelligence” in our culture is little more than a
mastery of the forms of literate uses of language’ (Olson 1984:238)—a
far cry indeed from Piaget and practical action.
Clearly, it is the latter, Vygotskian approach which offers the most
promise for a social psychology of educational knowledge. However,
rather than offering a straightforward Vygotskian account of education,
we shall enrich the analysis by drawing upon some recent developments
in social psychology itself. There has been in recent years a discernible
shift of interest in social psychology towards the study of language, and
especially towards conversational discourse and text (see Billig 1987;
Potter and Wetherell 1987; Billig et al. 1988). It is argued that much of
the traditional subject matter of social psychology may fruitfully be
studied in terms of the communicative work that people do in talk and
text. Discourse analysis, and a concern with rhetoric (argumentation), are
also at the heart of the new ‘sociology of scientific knowledge’ (for
example, Latour and Woolgar 1979; Gilbert and Mulkay 1984; Woolgar
1988). These studies examine how scientists work together in
laboratories, talk about their work, write about it in scientific journals,
argue with each other, and generally ‘construct’ scientific knowledge as
they do so. Scientific knowledge is studied as a socially constructed
system of shared (and disputed) assumptions, understandings and
procedures. For example, a contrast is often revealed between the
official, ‘scientific method’, version of science, with its impersonal
procedures and heroic, prize-winning individuals, and the social
practices and communications through which scientific knowledge is
created and established. Delamont (1987) has drawn attention to the lack
of cross-fertilization between these new insights into the nature and
workings of science, and the assumptions about science that prevail in
the policies and practices of science education.
Alongside these developments in the analysis of discourse, rhetoric
and scientific knowledge, there have been a series of studies of the
discourse of classroom education itself (for example, Mehan 1979;
Driver 1983; Edwards and Mercer 1986, 1987). Mehan’s work is
ethnomethodological (see Heritage 1984) in that its concern is with
showing how the practical business of classroom education is
accomplished— how the social order of the classroom education is
accomplished—how the social order of the classroom is created and
maintained over time, how teacher and pupils manage the moment-to-
moment taking of turns at talking. As linguists have also discovered
(Sinclair and Coulthard 1975), the pattern of classroom talk is highly
53
Derek Edwards
organized, and even in apparently child-centred, ‘progressive’ sorts of

classrooms, highly teacher-dominated. Teachers control the flow of
conversation, ask most of the questions, determine who speaks when,
and on what, and what the upshot shall be.
Rosalind Driver’s studies (Driver 1983; Driver and Oldham 1986;
Brook et al. 1988) are concerned more directly with knowledge than
with properties of talk, with how children think, how they spontaneously
explain things, and how they deal with the sorts of practical experiences
out of which, in our modern Piagetian classrooms, children are meant to
reinvent for themselves the findings and principles of science. She shows
the process to be highly problematical. In the absence of any prior
hypothesis or expectation, children simply do not observe what they are
supposed to. They also offer reasonable but erroneous explanations: the
time taken for a radio’s sound to fade away means that there must be a
very long piece of wire inside; the higher up something is, the greater the
gravitational force exerted upon it; the denser a metal, the harder it
should be for heat to penetrate it. Driver argues that it is quite wrong to
imagine that children can work out for themselves, merely through
observation and induction, what it has taken centuries of scientific
research to establish. A better approach, she argues, would be for
teachers to take a more active role ‘as mediators between the pupils’
experiences and understandings and that of the scientific community’
(Driver 1983:84) for example, by helping pupils to articulate the
significance of their observations. Thus, ‘by referring to the ideas and
investigations of past scientists, some of the powerful ideas of young
children can be explored in a way that treats them with respect’ (p. 67).
Valuable as Driver’s insights are, they are based upon two
assumptions about the relationship between language and thought
which pose difficulties for a social psychology of classroom
knowledge. First, there is the assumption of a straightforward
relationship between talk and thought. Children’s talk is taken as a
window upon their individual minds, as revealing how they think, how
they solve problems, how they understand the world. Second, the view
of science is one that pre-dates the new sociology of scientific
knowledge; depicting scientific truth in the classical manner, as
established, uncontroversial, arising from the proper application of
hypothetico-deductive methods. This has been termed the ‘storybook’
account of science (Mitroff 1974), and has emerged as one amongst a
series of ‘repertoires’ available to scientists for writing and talking
about their work: a post hoc way of accounting for a preferred version
of scientific truth, rather than an accurate description of how scientists
actually talk, think and work (see, for example, Gilbert and Mulkay
1984; Potter and Mulkay 1985). So, while we should hardly expect
children to reinvent in their classrooms the fruits of centuries of
54
scientific endeavour, we may still be asking too much, that they should
conform to an idealized version of scientific history and practice.
In Common Knowledge (1987), Edwards and Mercer offer an account
of classroom discourse in which, rather than taking the talk as a window
upon children’s thought processes, they examine it as contextualized
dialogue with the teacher. The discourse itself is the educational reality,
and the issue becomes that of examining how teacher and children
construct a shared account, a common interpretative framework for
curriculum knowledge, and for what happens in the classroom. This
involves teacher and pupils in the moment-to-moment construction of
‘context and continuity’, the shared basis of talk and action upon which
all subsequent talk and interpretation are built. Typically, this dialogue
turns out to be no simple negotiation between equals but a process that is
dominated by the teacher’s concerns, aims, expectations and prior
knowledge, via the teacher’s control of the dialogue.
Examining classroom discourse

Investigating shared understandings and how they’ develop requires a
close examination of classroom discourse. The first three extracts we
shall examine are from a transcribed video recording of the last few
minutes of a lesson on computer graphics. The teacher is coming to the
conclusion of her first lesson on the subject with a group of 9-year-olds.
They have been shown how to draw some straight lines and right angles
on the computer screen, using commands from the computer language
LOGO. Very briefly, this involves typing into the keyboard, instructions
which move a point (‘arrow’) on the screen: ‘FORWARD 100’ moves the
pointer 100 units in the direction it is pointing, drawing a line on the
screen as it goes. ‘RIGHT 90’ makes the pointer turn clockwise through
90 degrees, and a further ‘FORWARD 100’ would draw the second side
of a square. In extract 1 the teacher (‘T’) is talking the pupils through
how to complete a 180° turn, subtracting 60° to leave 120° (normal punc-
tuation is used here, with the addition of slashes for pauses, and some
contextual notes italicized in square brackets).
Extract 1:—Doing a 180° turn

T: You want to point that-a-way. You’ve got to turn from
there/right round to there. Can you tell me how many
degrees?
Susan: Measure it
T: You could measure it. Now how are you going to do
that?
Susan: With one of these [Susan picking up a protractor].
55
Derek Edwards
T: Make sure you’re measuring the right angle you’re

going to turn through. What do you turn it through?
Susan: Ninety.
T: Umm/you’re going to turn from that way right round
to point that-a-way. This is your turn from here to
there. Now you’ve got to measure the turn you’re
going to make. So what turn is that from there to
there?
Susan: A hundred and twenty.
T: Now how did you get that? You’re right Susan. How
did you get a hundred and twenty? Any idea Lara?
Lara: Sixty and sixty.
T: Not sixty and sixty/no. It’s got something to do with
the sixty but not quite. If you imagine/that line’s
continued/now that angle’s sixty. What do the
angles/ what’s a straight-lined angle? How many
degrees in a straight-lined angle? If I turn there/
right round/half turn or straight-line angle/straight
angle/how many degrees have I turned? There’s two
quarter turns look. That’s one. So one quarter turn is
how many degrees? What’s a quarter turn?[T
rotating her pencil through an angle of 90°, and
then 180°]
Lara: Ninety.
T: Ninety. Make another quarter turn. How many degrees
would that be? Another/
Lara: Half turn.
T: It’s a half turn but/the first quarter turn is ninety. How
many quarter turns do we need to make that half turn?
Lara: Four.
T: You think about it We’re pointing that way. OK we
want to point this way. We want to turn our pencil right
round so it’s pointing this way/one quarter turn which
is how many degrees?
Lara: Ninety.
T: Again/which is/another quarter turn is/
Lara: Ninety degrees.
T: It’s another ninety degrees. It’s two quarter turns that
way. If we do it one/watch the pencil/if we do one
quarter turn/it takes us to there. Yes/that’s ninety
degrees.
Lara: A hundred and eighty degrees.
T: A hundred and eighty degrees. Right. And that is/
what that measures from there right round to there/
56
so if to there is sixty degrees what’s that? It’s a

hundred and eighty altogether/and that’s sixty so
what’s that one?
Lara: Forty a hundred and forty.
T: Not a hundred and forty. What are you trying to do?
Lara: Add them together.
T: Not add them together. The whole lot is a hundred and
eighty. The whole turn/from there round is a hundred
and eighty degrees. Yes/that part of it is sixty degrees/
so what’s that part going to be? Think about that. All
right. Success [T looking at what Lara has written onto
the computer screen]/OK.
Extract 1 displays several of the features of classroom discourse that

we have been discussing, and especially, the issues of control, and of
the building of a context of shared understanding. The teacher is firmly
in control, both of the flow of speech, of who takes turns at speaking,
and also of its content. She does most of the talking, asks the questions
that determine speaker and topic, and defines the points at which each
question is satisfactorily answered. Her method is not simply to tell the
pupils things; she could easily have told them outright, that a straight
line represents 180°, and articulated the rule, for them to learn by heart,
that to draw a triangle, you turn the ‘arrow’ through 180° minus
whatever angle you wish to use (in the case of an equilateral triangle
180–60=120). Rather than that, she engages in the ‘eliciting’ style of
progressive, child-centred education. She asks questions, and guides
the pupils to the required answers by means of how her questions are
couched, and by her responses, reactions, pauses, gestures and mimes.
The pupils’ pursuit of the angle of a straight line is progressively
narrowed to working out, via the teacher’s prompts and demonstrations
with the pencil, what is twice ninety.
It is also a heavily contextualized piece of dialogue. One would need
to be there to see what exactly they are talking about; even the video
recording does not always reveal what people are looking at or referring
to. The point is, of course, that the participants themselves are required to
see and understand these things. But simply ‘being there’ is not enough.
It is not merely a physical context that they are working with. The teacher
is at pains to ensure that the significance of what is done and seen is
understood by everyone, and can be used as the basis for proceeding with
the lesson; the discourse of context and continuity is addressed to matters
of shared understanding. So, once Susan comes up with the answer,
120°, the teacher switches to her partner Lara, and makes sure, not only
that she also knows the answer, but that she can be guided towards
working it out, with however much help, for herself.1 The teacher’s
57
Derek Edwards
questions are not oriented towards finding things out—at least not in the
obvious sense. She does not need the pupils to tell her ‘answers’ that she
does not know. Rather, they are oriented towards finding out what the
pupils themselves know, serving to establish a commonality of
understanding between herself and the pupils she is teaching. Teachers’
questions also impart information, as well as requesting it, in that they are
devices for introducing the formulations, conceptualizations and links
between things that the pupils are meant to adopt (‘One quarter turn
which is how many degrees?’)
Let us pursue this notion of establishing shared formulations. Extract
2 continues from where extract 1 leaves off. The teacher, having spent the
last few minutes with the girls, now moves them over to where the boys
are working at another computer screen. The teacher has already (earlier
in the same lesson) worked through, with the boys, what an ‘equilateral’
triangle is, and how to draw one using LOGO, turning the screen pointer
through angles of 120° to draw the required 60° angles.
Extract 2: Talking geometry

T: OK will you come over and join this problem ’cause
we’ve got a bit of a problem here/and it’s one I want
you to think about for next week. We’re going to try
and draw a triangle. Now we’ve decided there’s
something different about it to what/we’ve been doing.
What do you think’s different?
Mark: The angles
T: The angles are different aren’t they.
Stuart: Instead of ninety degrees you turn forty five.
T: Well/DO you?/OK/I want you to try for next week/ to
draw a triangle. Now the girls have got an equilateral
triangle here. How about you two [Mark and Stuart]
doing an isosceles triangle?
Stuart: What’s that?
T: To try/what’s an isosceles triangle? Tell him Mark.
What’s an isosceles triangle? [Mark giggles, as T puts
her hand on his head]
Mark: It’s one with/not all equal sides.
T: Not all equal/are any of them equal? Isosceles/[T
holding up two figures]
Mark: Two.
T: Two equal sides/OK. One’s got two equal sides/that
one’s got three/so what’s that one called?
Mark: Called equal/[T begins nodding her head] equilateral.
T: Equilateral.
58
The appeal to a continuity of shared understanding, the building from

moment to moment of a context of common knowledge, is quite explicit
in extract 2. The problem to be solved has a future orientation—it is one
to pay special attention to, because it is to be taken up ‘next week’. The
starting point is the immediate past—the context of shared knowledge
that has just been created—‘now we’ve decided there’s something
different about it…’. What previously was the explicit, acted-out and
talked-about business of the lesson, is now offered by the teacher as part
of an implicit common understanding, ‘what we’ve been doing’, the
‘given’ basis upon which new understandings can be built.
The teacher proceeds in the familiar, eliciting style (‘What do YOU
think’s different?), together with clues, and prompts that signal when the
answer is not the required one (‘Well/Do you?’). Then comes the
introduction of new terminology, though again, apparently via elicitation
(‘What’s an isosceles triangle. Tell him Mark’). Through the introduction
of a common vocabulary, a common conceptual currency is
established—joint terms of reference, by which teacher and pupils can
express their common understandings, and also formulate these within
the recognized language of geometry. For their part, the pupils appear to
understand that, although they are being asked a question, the teacher
will invariably provide them with clues to its answer. The teacher is
understood to be in control of truth and definition. It having been
defined, earlier in the lesson, what an ‘equilateral’ triangle is, Mark now
offers the safe proposition that an ‘isosceles’ triangles must be something
other than that— ‘It’s one with/not all equal sides’. This is not
acceptable. But rather than simply put him straight, the teacher proceeds
to cue the required response, with words and gesture (the two fingers),
and her well-timed nod then helps Mark to recall the term ‘equilateral’.
We have termed these uses of prompts, gestures, reformulations of
questions and other devices ‘cued elicitation’ (Edwards and Mercer
1987). They serve, amongst other things, to maintain the appearance that
knowledge and understanding are being elicited from the pupils
themselves, rather than being imposed by the teacher. We shall discuss
later, some educational problems that this subterfuge may create.
Extract 3 is the remainder of the lesson, beginning where extract 2
leaves off. The teacher is continuing to orient the pupils forward, to what
will happen in the next lesson. The capitalized words are LOGO
commands.
Extract 3:‘So, remember’

T: Equilateral./What we’re going to do for next week is/
you’re going to go away and you’re going to try and
do it on your own/right You [T addressing the girls]
59
Derek Edwards
try and do a program for equilateral triangle. You two

[T looking at Mark and Stuart]/think about the turns
you’ve got to make/not the ones you’ve already got
That’s a problem because we’ve gone LEFT
NINETY haven’t we [T pointing to the diagram she
has drawn on paper]. FORWARD SEVENTY I think
it was. Now we want to turn to point in that direction.
We’re trying to work out what our turn would be. We
know that angle’s sixty degrees [pointing to an
internal angle of the triangle]/but is that the angle we
want to turn?
Susan: No
T: [Turning to Mark] Is it?
Mark: No.
T: No. We want to turn from there/right round to there/ so
we’ve got to decide the angle of turn. So remember
that when you’re writing your program/you’ve got to
write down the angle which you want the arrow to turn
through/not any of the turns you’ve already got. It’s
what you want it to turn through. So your problem [T
addressing the boys] is an isosceles triangle/ doesn’t
matter about the length of the sides/and yours [the
girls’] is an equilateral triangle/OK/and we’ll have a
go at that next week/and perhaps another thing/if you
manage to/you suss out the triangles/and we’ll try and
do our initials or something next week because/this
week you haven’t taken the pen off/the arrow off.
What you’ve done is come backwards if you’ve
needed to go in another direction. Next week I’ll show
you how to take the pen off/as if you lift the pen off the
paper [T gesturing lifting her pen off a sheet of paper]/
and how to put it back on again. OK/ so that’s your
problem this week/equilateral triangles. Think you’ll
manage that?
T: Yes
Mark: OK/fine.
The point about this future-orientation of the teacher’s talk is that she is
not merely providing information, or forward planning, about the next
lesson. She is taking the opportunity to recap and reformulate what it is
that the pupils are supposed to take away from this lesson, and bring back
with them to the next. And it is not merely some homework. It is some
shared understanding: ‘So we’ve got to decide the angle of turn/ so
remember…you’ve got to write down the angle which you want the
60
arrow to turn through…it’s what you want it to turn through’. The words
‘right’, ‘so’, and ‘so remember’ mark out the upshot of the lesson, the
point of it all, the knowledge to take away, think about, bring back, and
be ready to build upon. The lesson’s major point is made and repeated—
to draw a triangle on the computer screen you have to turn the pointer not
through the triangle’s internal angle, but through its complementary one.
In fact, teachers’ recaps of what has been ‘done’ or ‘covered’ in lessons,
are opportunities not only for repeating and reinforcing these shared
understandings, but also, for tidying them up, redefining them in terms
that better fit the teacher’s preferred version of the knowledge in
question. These ‘reconstructive’ recaps often occur at the beginnings and
ends of lessons, but also during them, at moments when some doubt has
arisen, that the required version of events is jointly understood (see
Edwards and Mercer 1987; and in press, for a detailed examination).
Discourse, rhetoric and school science

The development of classroom knowledge is the development of a
discourse, the creation of a shared conceptual framework, a common
language for the interpretation and communication of thought and
experience. We have seen that shared understandings are heavily
contextualized in terms of the unfolding talk and activity, with the
significance of experience and discovery being marked out in the talk,
conceptualized, reformulated, and forming the basis for subsequent
talk. Despite a superficial reliance upon pupils’ own ideas and
contributions, it is also very much a teacher-dominated affair. On close
examination, classroom education looks much more like socialization,
an inculcation of pupils into a predetermined culture of educated
knowledge and practice, than some unfolding development of
individual cognitions.
We may ask, therefore, what is going on, when teachers seem so much
at pains to display education as ‘e-ducare’, the classic definition, a matter
of drawing out, leading out, or eliciting from pupils, thoughts and ideas
(‘cognitions’), that are latent within. Why do they overtly elicit
everything from the children, while simultaneously gesturing, hinting at,
implying and cuing the required answers? This sort of classroom
discourse seems oriented towards the accomplishment of consensus,
with pupils and teacher coming as soon as possible (if not immediately)
to a shared understanding that accords closely with what the teacher
already knows. Teacher and pupils appear to collude in this consensus,
the content of which, while generally emanating from the teacher, is
displayed as if emanating from the children. John Holt (1969) has
documented a range of ways in which children learn to fake
understanding, seeking to achieve recognized ‘right answers’, at the
61
Derek Edwards
expense of any grasp of principle. It is important to emphasize that

examining classroom discourse in terms of joint understandings does not
mean that we must be solely concerned with consensual agreement.
Shared understanding is the issue, not the assumption. We are interested
in how teachers and children orient themselves in talk, to the issue of
shared understanding, in how it is displayed or accomplished, or fudged,
or ignored, or disputed. What we find, in the main, is that teacher and
pupils strive to accomplish an appearance of consensus, avoiding
resolved disagreements, and indeed, avoiding disagreements altogether.
Teachers are themselves trained and educated. They learn at college
and university, and in teaching practice, the current educational theories
and ideologies—the philosophy of e-ducare, of Piaget and Plowden, the
inadequacy of teaching things to children that they are not ready to learn,
that they may learn by rote but not understand. Teachers themselves can
articulate the reasons, though not everything that they do is explicitly
formulated. But the ideal of elicitation has to be practised against the
requirement that the job gets done. There is an overriding practical
concern with achieving order and control, both behavioural and
cognitive, and with doing the curriculum. They are faced with the
dilemma of having to elicit from children precisely what the curriculum
determines shall be taught (these issues are discussed in more detail, in
Edwards and Mercer 1987, and in Billig et al. 1988).
Another reason for all the subtle subterfuges may be that educationists
are operating with an inadequate notion of the subject to be taught—of
what doing ‘science’ is really like. We argued earlier that the popular
view of science, and the one favoured even by scientists themselves when
discussing their own, or some favoured work, is something of a
‘storybook,’ version. The lone scientist constructs knowledge through
hypothesis and strict empirical method, the proposed theory being that
which must be forced upon any rational mind by the evidence available.
Other ‘contingent’ repertoires are available for accounting for what are
considered to be false beliefs, opposing views and interpretations,
misleading research, false trails and so on (see Gilbert and Mulkay 1984;
Potter and Wetherell 1987). The point is, that science as it is practised is
a social activity, a discourse amongst scientists that is oriented towards
what is taken to be common knowledge. It has the character of a debate,
and makes full use of the devices of rhetoric and persuasion (see also
Yearley 1981, 1985).
Classroom science has some of the features of everyday scientific
practice. It also is a communal, discursive activity, a construction of
shared understanding, publicly communicated and embodied in shared
symbolic forms—conversation, writing and text, diagrams, drawings and
so on, with the same concern for establishing a common conceptual
framework for the encoding of experience, method, observation and
62
theory. The contrast lies in the lack of debate. Science has no ‘teacher’,
no ultimate power and authority who knows all the answers in advance.
There is, indeed, some of that quality in scientific practice. Events can go
unobserved until some new theory gives them meaning, or even until
some theory reifies them as observations at all—the development of
science is full of instances of disputed observations, disputed
phenomena, suspicions that the observed data of deepest space, of
electron microscopes or of sub-atomic particles, may be more revealing
of the equipment that produces them, than of the world they purport to
measure. But what will prevail as the orthodox theory, or as an exciting
alternative to it, or as an interesting or crucial discovery, has to be
demonstrated, achieved, argued for, and accepted as such by other
scientists.
The notion that pupils may be unable to understand what their
individual cognitive development has not yet prepared them for, has an
ironic parallel in classroom discourse. Readiness is indeed an issue, but
what often occurs is that pupils come up with ideas that the teacher is not
ready for. In their third lesson on computer graphics (two weeks later),
the children decided that it would be a good idea to draw their names or
initials on the screen (extract 4):
Extract 4:— ‘You can’t do curves can you?’

Karen: (to Lisa)…The S is going to be a bit difficult/though.
You could just write LC if you wanted to.
T: But you can’t do curves can you? You can’t do an S.
We haven’t learnt to do them…
Lara: We could do that [finger drawing a boxed S in the air].
T: That’s right…box writing as long as it’s got right
angles/as long as it’s not curved/you’re all right. It can
do curves/but we haven’t learned how to do them yet.
So/just use what you know. That’s the best way.
Again, we have an explicit appeal to a continuity of shared knowledge

that encompasses what has already been incorporated into their common
understanding, and which extends into the future: ‘you can’t do’ what
‘we haven’t learnt to do (yet)’. But here, it operates also as a constraint.
The teacher’s ‘can you?’ superficially elicits, but the teacher’s own
answer (‘you can’t’) closes down the possibility of Lara’s contradicting.
Rather than following up the pupils’ own thoughts and problem
proposals, we have to settle for what, at a point near the scheduled end of
the lesson, the teacher has taught so far.
The restrictions of teacher-dominated discourse point to the possibility
that pupil-pupil dialogue may be able to offer some compensatory
63
Derek Edwards
advantages. It was argued earlier that adult science is more open-ended and
uncertain than school science, more argumentative, free-wheeling, open to
disputation and reformulation. This suggests a role in the development of
educational knowledge, for children to argue things through with each
other, without the teacher’s constant supervision and control over where
the discussion has to lead. Pupils do, of course engage in such discussions.
But one still has to consider what their importance, if any, might be. Pupils’
dialogues may help to reveal how individual pupils think (this is the usual
use of such materials), but the possibility remains that they may be
important also, for the development of dialogical, argumentative forms of
thinking. Extract 5 is taken from the end of the third lesson on computer
graphics. The teacher has allowed the group to continue ‘playing’ with the
computer, while the researchers were packing up their recording
equipment. Stuart and Tracy continued to try to draw their names on the
screen. We left the camera running.
Extract 5: Arguing and thinking

Tracy: RIGHT NINETY/then we’ll be there and I don’t want
to be there.
Stuart: RIGHT NINETY/is just here…write your columns
down here…It’s still not finished yet.
Tracy: Then/it’s/RIGHT SIXTEEN again. Put RIGHT
SIXTEEN.
Stuart: Where?
Tracy: We’re/there we are. We’re facing that way/we want to
go/RIGHT SIXTEEN.
Stuart: Then we want/FORWARD.
Tracy: No/we don’t want to be there. We’re doing a RIGHT.
Look. We go down there/go back up there/go down
there.
Stuart: FORWARD
Tracy: Oh/come on/you do it (…)
Stuart: NINETY/that’s your first (…) No that’s wrong (…)
Tracy: LEFT NINETY.
Stuart: LEFT or RIGHT?
Tracy: LEFT.
Stuart: We’d be going that way then.
Tracy: Tell you what/we could go that way.
Stuart: No we’re going that way.
Tracy: Over there/you said/RIGHT NINETY we’d be going
that way. He don’t’ know his left and right. NINETY/
so then we’ll be facing that way. FORWARD TEN.
What’s that supposed to be?
64
Stuart: What/who I mean?

Tracy: TEN then we’ll be here/there then I want to go
Stuart: Miss [calling to the teacher]/what angle do we/what
are we (…) to be back up there again
T: Where are you? You’re there/and you want to turn
through that angle there?
Stuart: No/no/we don’t want to do that one Miss. We’re going
to do one of the smaller ones (…) was it LEFT or
RIGHT?
Tracy: See I told you.
T: He’s already done it.
It is clear that the children are not merely taking turns at thinking aloud,
revealing for an observer’s convenience, their processes of thought.
Rather, they are engaged with each other in joint action and argument,
such that what each says and does affects the other’s next move.
Furthermore, their verbal formulations of thought, are dialogical and
argumentative in form, addressed persuasively to the difference
between a proposed action or solution, and what the other person thinks
or assumes — ‘No/we don’t want to be there. We’re doing a RIGHT.
Look. We go down there…’ ‘No that’s wrong…we’d be going that way
then.’) Argument is a form of thought, indeed many and varied forms of
thought, irreducible to the mere adding together of individual
cognitions that may happen to be put into words (cf. Billig 1987, from
which the caption for extract 5 is borrowed). As with science proper,
once we abandon the notion that there is some higher, pervasive
authority that already knows everything, ideas have to be justified
against criticism, supported by argument and evidence against possible
refutation. They do not spring from the data, nor from some rigid logic,
evident and undeniable. They do not arise merely out of the perception
and cognition of individuals, finding form in words only as some final
part of the process, for the purposes of teaching them to someone else.
The common knowledge of science and of the school classroom is
intrinsically social, framed in a common language, dialogical in form
and process.
Ironically, given the powerful influence of Piaget upon the ‘child-
centred pedagogy’ (Walkerdine 1984), it is Piaget who provides us with
the clearest statement of the importance of argumentation between
children, in the development of their thought:
What then gives rise to the need for verification? Surely it is

the shock of our thought coming into contact with that of
others, which produces doubt and the desire to prove…The
social need to share the thought of others and to
65
Derek Edwards
communicate our own with success is at the root of our need

for verification. Proof is the outcome of argument…Logical
reasoning is an argument which we have with ourselves, and
which reproduces internally the features of a real argument.
(Piaget 1928:204)
The prevailing emphasis upon individual learning, individual cognitive

growth, coupled with the pervasiveness of teacher- dominated dialogue,
leads to some of the subterfuges and inadequacies of the sorts of teaching
we have examined here. There is a tendency in teachers to keep their
educational goals secret; pupils cannot be told too much, but have to
work things out for themselves. As Driver also observed:
It is common to see science lessons which end with the

clearing up after the practical work is finished. The time for
the important discussion of how the important experiences
gained relate to the new ideas is missed. Activity by itself is
not enough. It is the sense that is made of it that matters.
(Driver 1983:49)
We might add that discussion is important throughout the lessons, and

that must include discussion between pupils, as well as with the
teacher. Our analyses of classroom discourse suggest that teacher-pupil
talk and pupil-pupil talk are both important, but often for different
reasons. There are overriding asymmetries between teacher and pupil,
both cognitive (in terms of knowledge) and interactive (in terms of
power), that impose different discursive patterns and functions.
Teachers’ expertise lends itself to direct explanation and to assisted
learning, of the Vygotskian sort, in which the less competent child is
helped (‘scaffolded’ is Jerome Bruner’s term), towards increased
competence. But we should not ignore what pupils can learn from
others who know no more than themselves—the skills of disputation,
the notion that all knowledge is questionable, or in need of scrutiny and
justification, that we do not always have to agree. It is important that
the teacher’s knowledge should not be immune from such an approach.
Perhaps the most difficult achievement is not to let children argue with
each other, but to open up the teacher’s own understandings for
scrutiny in the classroom—her pre-established plans and assumptions,
her aims and methods, her own understandings and conclusions, for
disputation, without the whole exercise becoming merely another
guessing game of what the teacher wants us all to say. In any case, there
is clearly an important role for a social psychology of educational
knowledge, which emphasizes the social and communicative nature of
knowledge, which bases itself upon a sound understanding of the social
66
and communicative foundations of the very knowledge (‘science’, for

example) that education seeks to impart, and which has as its empirical
basis, the close examination of educational talk and texts.
Note
1 There is an intriguing similarity here between the teacher’s elicitation of
geometry from these pupils, and the famous sequence in Plato’s Meno,
in which Socrates, in a dialogue with a slave boy, appears to elicit from
him a version of Pythagoras’s theorem. Socrates claims that the boy
must have known the theorem innately (see Billig et al. 1988, Chapter 4
for a full discussion of the significance of this parallel, its relevance to
modern educational practice, and the doctrines of Piagetian learning and
of innate knowledge).
References
Billig, M. (1987) Arguing and Thinking, Cambridge: Cambridge
University Press.
Billig, M., Condor, S., Edwards, D., Gane, M., Middleton, D.J. and
Radley, A.R. (1988) Ideological Dilemmas: a Social Psychology of
Everyday Thinking, London: Sage.
Brook, A., Driver, R. and Johnston, K. (1988) ‘Learning processes in
science: a classroom perspective’, in J.J.Wellington (ed.) Skills and
Processes in Science Education: a Critical Analysis, London:
Methuen.
Bruner, J.S. (1986) Actual Minds, Possible Worlds, London: Harvard
University Press.
Butterworth, G. and Light, P. (1982) Social Cognition: Studies of the
Development of Understanding, Brighton: Harvester Press.
Carey, S. (1985) Conceptual Change in Childhood, Cambridge, Mass.:
Bradford Books/MIT Press.
Delamont (1987) ‘Three blind spots? A comment on the sociology of
science by a puzzled outsider’, Social Studies of Science 17: 163–70.
Donaldson, M. (1978) Children’s Minds, London: Fontana.
Driver, R. (1983) The Pupil as Scientist, Milton Keynes: Open University
Press.
Driver, R and Oldham, V. (1986) ‘A constructivist approach to
curriculum development in science’, Studies in Science Education 13:
105–22.
Edwards, D. and Mercer, N.M. (1986) ‘Context and continuity:
classroom discourse and the development of shared knowledge’, in
K.Durkin (ed.) Language Development in the School Years, London:
Croom Helm.
Edwards, D. and Mercer, N.M. (1987) Common Knowledge: the
Development of Understanding in the Classroom, London: Methuen.
Edwards, D. and Mercer, N.M. (in press) ‘Reconstructing context: the
conventionalization of classroom knowledge’, Discourse Processes.
Engineering Council and the Standing Conference on Schools’ Science
and Technology (1985) Problem Solving: Science and Technology in
Primary Schools, London.
67
Derek Edwards
Fodor, J.A. (1975) The Language of Thought, New York: Thomas

Y.Crowell.
Gilbert, G.N. and Mulkay, M. (1984) Opening Pandora’s Box: a
Sociological Analysis of Scientists’ Discourse, Cambridge: Cambridge
University Press.
Heritage, J. (1984) Garfinkel and Ethnomethodology, Cambridge: Polity
Press.
Holt, J. (1969) How Children Fail, Harmondsworth: Penguin.
Keil, F.C. (1986) ‘On the structure-dependent nature of stages of
cognitive development’, in I.Levin (ed.) Stage and Structure:
Reopening the Debate, New Jersey: Ablex.
Latour, B. and Woolgar, S. (1979) Laboratory Life: the Social
Construction of Scientific Facts, London: Sage.
Mehan, H. (1979) Learning Lessons: Social Organization in the
Classroom, Cambridge, Mass.: Harvard University Press.
Mitroff, I.I. (1974) The Subjective Side of Science, Amsterdam: Elsevier.
Nuffield Foundation (1967) Nuffield Maths Project, I Do and I
Understand, London: W. & R. Chambers and John Murray.
Olson, D.R. (1984) ‘Oral and written language and the cognitive
processes of children’, in A.Lock and E.Fisher (eds) Language
Development, London: Croom Helm/The Open University.
Piaget, J., (1928) Judgment and Reasoning in the Child, London:
Piaget, J. (1970) ‘Piaget’s theory’, in P.H.Mussen (ed.) Carmichael’s
Manual of Child Psychology, New York: John Wiley & Sons.
Plowden Report (1967) Children and their Primary Schools, 2 vols.
Report of the Central Advisory Council for Education in England,
London: HMSO.
Potter, J. and Mulkay, M. (1985) ‘Scientists’ interview talk: interviews as
a technique for revealing participants’ interpretative practices’, in
M.Brenner, J.Brown and D.Canter (eds) The Research Interview: Uses
and Approaches, London: Academic Press.
Potter, J. and Wetherell, M. (1987) Discourse and Social Psychology,
London: Sage.
Russell, J. (1978) The Acquisition of Knowledge, London: Macmillan
Press.
Sinclair, J.McH. and Coulthard, R.M. (1975) Towards an Analysis of
Discourse: the English used by Teachers and Pupils, London: Oxford
University Press.
Vygotsky, L.S. (1987) Thought and Language, Cambridge, Mass.: MIT
Press.
Walkerdine, V. (1982) ‘From context to text: a psychosemiotic approach
to abstract thought’, in M.Beveridge (ed.) Children Thinking through
Language, London: Edward Arnold.
Walkerdine, V. (1984) ‘Developmental psychology and the child-centred
pedagogy: the insertion of Piaget into early education’, in
J.Henriques, W. Hollway, C.Urwin, C.Venn and V.Walkerdine
Changing the Subject, London: Methuen.
Wertsch, J.V. (1985) Vygotsky and the Social Formation of Mind,
Wood, D. (1988) How Children Think and Learn, Oxford: Basil
Blackwell.
Woolgar, S. (1988) Science: the very idea, Chichester: Ellis Horwood.
68
Yearley, S. (1981) Textual persuasion: the role of social accounting in the

construction of scientific arguments’, Philosophy of the Social
Sciences 11: 409–35.
Yearley, S. (1985) ‘Vocabularies of freedom and resentment: a
Strawsonian perspective on the nature of argumentation in science and
the law’, Social Studies of Science 15: 99–126.
69
Chapter five
Self-concept and school achievement
Dale H.Schunk
For many years social psychologists have been concerned with a
study of the self-concept. Within a cognitive approach to social
psychology this concern has led to the view that the individual is an
active seeker of information, concerned with establishing a view of
themselves that is consistent with their lives and that is centrally
concerned with the degree to which they have control over important
life events.
Within educational research there has developed a particular tradition
that has seen self-concept as being closely involved with the individual’s
level of academic attainment. As Dale Schunk points out, while the
relationship between self-concept and academic attainment exists, it is
far from clear just what it means.
Schunk’s case is based on the assumption that the self-concept is
best regarded as being hierarchically organized. A general self-
concept, our fundamental view of ourselves, can be regarded as the
summation of a number of more particular beliefs relating to different
aspects of our range of activities. It is these more specific self-concepts,
Schunk argues, with which it is important for the teacher to be
concerned.
Schunk’s contribution illustrates how research into the operation of
the self-concept, using the core notion of self-efficacy (one’s beliefs in
what one can achieve) leads into a number of practical
implementations. The chapter is wide ranging in its scope and
introduces concerns with the ways in which a teacher might use peer
modelling and goal setting to further the development of pupils’ self-
efficacy and thereby enhance levels of attainment. The chapter,
together with the following chapters by Rogers on motivation (Chapter
6), Kutnick on pupil autonomy (Chapter 7), and Slavin on co-operative
learning strategies (Chapter 13), illustrates how concerns with the inner
child can be seen to sit happily alongside a concern with classroom
practices.
70
Introduction
Current theoretical accounts of learning view pupils as active seekers
and processors of information (Pintrich et al. 1986). Learners’
cognitions influence the instigation, direction, and persistence of
achievement behaviours (Winne 1985; Corno and Snow 1986; Schunk,
in press). Research conducted within various traditions emphasizes
pupils’ beliefs concerning their capabilities to control important
aspects of their lives (Nicholls 1983; Covington 1984; Weiner 1985;
Bandura 1986).
This chapter discusses the influence of the self-concept on children’s
motivation and learning in school. The self-concept comprises such
dimensions as self-confidence, self-esteem, self-concept stability and
self-crystallization (Rosenberg and Kaplan 1982). Much of the present
discussion is focused on perceived self-efficacy, or individuals’
perceptions of competence at academic tasks. Initially I give a brief
theoretical overview of self-concept to include its development and
differentiation. Research is reviewed that examines the influence of
general self-concept on pupil achievement. I then present theory and
research on domain-specific self-perceptions with the concentration on
self-efficacy. The chapter concludes with suggestions of ways to enhance
pupils’ academic self-concepts.
Self-concept: development and differentiation

Self-concept refers to one’s collective self-perceptions that are formed
through experiences with, and interpretations of, the environment, and
that are heavily influenced by reinforcements and evaluations by
significant other persons (Shavelson and Bolus 1982). Self-concept is
multidimensional. Self-esteem is one’s perceived sense of self-worth—
whether one accepts and respects oneself. Self-confidence denotes the
extent to which one believes one can produce results, accomplish goals,
or perform tasks competently. Self-esteem and self-confidence are
related. The belief that one is capable of performing a task competently
can raise self-esteem. High self-esteem might lead one to attempt a
difficult task, and subsequent success enhances self-confidence.
Other dimensions of self-concept are stability and self-crystallization
(Rosenberg and Kaplan 1982). Self-concept stability refers to the ease or
difficulty of changing the self-concept. Stability depends in part on how
crystallized or structured are one’s beliefs. Beliefs become crystallized
with repeated similar experiences. By adolescence people have relatively
well-structured perceptions of themselves with respect to such
characteristics as general intelligence, sociability and honesty. Brief
experiences providing evidence that conflicts with individuals’ beliefs
71
Dale H.Schunk
will not have much impact. In contrast, self-concept is readily modified

in areas where people have ill-formed notions about themselves, usually
because they have little if any experience (for example, ability to learn
Swahili or perform mental rotations).
The development of self-concept proceeds from a concrete view of
oneself to a more abstract one (Montemayor and Eisen 1977). Young
children perceive themselves concretely: they define themselves in terms
of their appearance, actions, name, possessions, and so forth. Children do
not distinguish between overt behaviours and underlying abilities or
personality characteristics. They also do not have a sense of an enduring
personality. Children’s self-concepts are rather diffuse and loosely
organized. Children acquire a more abstract self-view with development
and especially as a result of schooling. They develop separate
conceptions of underlying traits and abilities and understand that
behaviours do not always match capabilities. The self-concept becomes
better organized and more complex.
Development also brings greater differentiation of the self-concept
Although most investigators postulate the existence of a general self-
concept, recent evidence indicates that the self-concept is hierarchically
organized (Shavelson and Bolus 1982; Marsh and Shavelson 1985) with
a general self-concept located at the top of the hierarchy and specific sub-
area self-concepts towards its base. Self-perceptions of specific
behaviours presumably influence sub-area self-concepts (for example,
English, mathematics), which in turn combine to form the academic self-
concept. The general self-concept is formed by self-perceptions in the
academic areas combined with those in non-academic (social, emotional,
physical) domains.
The knowledge that children acquire about themselves occurs
primarily through their social interactions (Mead 1934). School children
receive many opportunities to evaluate their skills and abilities, and this
evaluative information contributes to the formation and modification of
their self-concepts. Performances are evaluated against absolute and
normative standards. Absolute standards are fixed; normative standards
are based on the performance of others. Both types are found in school.
Pupils whose goal is to complete six workbook pages in thirty minutes
gauge their progress against this absolute standard. Grading systems
often are based on absolute standards (for example, A=90–100, B= 80–
89, and so on).
Social comparison refers to comparing one’s performances with
those of others. Festinger (1954) hypothesized that, where objective
standards of behaviour are unclear or unavailable, people evaluate their
abilities and opinions through comparisons with those of others, and
that the most accurate self-evaluations derive from comparisons with
those who are similar in the ability or characteristic being evaluated.
72
The ability to use comparative information depends on higher levels of

cognitive development and experience in making comparative
evaluations (Veroff 1969). Festinger’s hypothesis may not apply to
children younger than age 5 or 6 because such children tend not to
relate two or more elements in thought (for example, their
performances with those of others) (Higgins 1981). This does not mean
that young children cannot evaluate themselves relative to others,
rather, they may not automatically do so. Children show increasing
interest in comparative information in the early school years and by age
8 to 10 regularly use such information to help form self-evaluations of
competence (Ruble et al. 1976; Ruble et al. 1980).
The meaning and function of comparative information change with
development. Pre-schoolers actively compare at an overt physical level
(for example, amount of reward). Other social comparisons involve how
one is similar to and different from others and competition is based on a
desire to be better than others without involving self-evaluation (for
example, ‘I’m the general; that’s higher than the captain’) (Mosatche and
Bragonier 1981). Later on social comparisons shift to a concern for how
to perform tasks (Ruble 1983). Early infants engage in peer comparisons,
but these often are directed toward obtaining correct answers. Providing
comparative information to young children increases their motivation for
practical reasons. Direct adult evaluation of children’s capabilities (for
example, ‘You can do better’) seems more influential in helping children
form self-evaluations of capabilities.
Recent research suggests that the self-concept is not passively formed
through environmental interactions; rather, it is a dynamic structure that
mediates significant intrapersonal and interpersonal processes (Cantor
and Kihlstrom 1987). Markus and her colleagues (Markus and Nurius
1986; Markus and Wurf 1987) hypothesize that the self-concept
comprises self-schemas or generalizations formed through prior
experiences. These schemas process personal and social information
much as academic schemas process cognitive skill information
(Rumelhart and Ortony 1977). Self-schemas differ in various ways.
Some are more elaborate (that is, include more information) than others;
some refer to the present self, whereas others refer to past or possible
future selves. Some schemas are positive, others are negative (for
example, ‘I’m no good’).
The multidimensional nature of the self-concept is captured in the
notion of working self-concept. Not all self-representations are equally
active at the same time. Working self-concept refers to those self-
schemas that are mentally active at the moment—one’s presently
accessible self-knowledge. This conceptualization allows for a relatively
stable core (general) self-concept, surrounded by domain-specific self-
concepts that are capable of being altered more easily. The latter are
73
Dale H.Schunk
activated by task circumstances. For example, pupils working on

mathematics problems have their mathematical self-concepts active, but
once they switch to a different task (for example, reading), different
schemas will be activated. Changes in domain-specific self-concepts that
occur as a result of learning have an impact upon more general self-
knowledge.
Self-concept and academic achievement

It is commonly believed that self-concept bears a positive
relationship to school achievement. The thinking runs something like
this. Pupils who are confident of their learning abilities and feel a
sense of self-worth display greater interest and motivation in school,
which enhances achievement. Higher achievement, in turn, validates
one’s self-confidence for learning and maintains a high sense of self-
esteem.
Although these ideas are intuitively plausible, they have not been
consistently supported by research. Wylie (1979) reviewed seventy-
eight research studies examining the relationship of self-concept to
academic achievement and ability. The average correlation between
achievement measures (grade point averages) and overall measures of
self-concept was +0.30. Correlation coefficients range from -1.00
through 0 to +1.00 and indicate the direction and strength of the
relation between two measures. Positive coefficients denote a direct
relationship: individuals high on one measure also are high on the other
measure. Negative coefficients denote an inverse relationship:
individuals high on one measure are low on the other measure.
Coefficients at or close to zero indicate no relationship: individuals
high on one measure are equally likely to be high or low on the other
measure. Regardless of sign (plus or minus), the higher the number the
stronger the relationship. For example, +0.10 is a weak positive
correlation, -0.82 a strong negative correlation. Wylie’s finding of
+0.30 is a moderate positive correlation.
Correlations were somewhat higher when standardized measures of
self-concept were employed, whereas they were lower (and often
nonsignificant) with idiosyncratic, researcher-developed self-concept
measures. Higher correlations—many around +0.50—were obtained
between grade point averages and measures of academic self-concept
(self-concept of ability). That higher correlations were found between
achievement and academic self-concept than between achievement and
overall self-concept supports the hierarchical organization of self-
concept (Shavelson and Bolus 1982). The highest correlations with
academic achievement have been obtained with subject-area self-
concepts (for example, English, mathematics).
74
Even assuming that self-concept and achievement are positively

related, researchers have not agreed on the direction of influence
(Scheirer and Kraut 1979; Shavelson and Bolus 1982). Different models
have postulated that self-concept influences achievement, that
achievement influences self-concept, that self-concept and achievement
reciprocally influence one another, and that self-concept and
achievement are each influenced by other variables (for example,
teaching).
For various reasons, self-concept studies have typically not explored
causal mechanisms. For one, many studies have not been based on
theoretical models postulating causal dominance. For another, methods
for exploring causal relations in correlational studies, until fairly
recently, were not widely used. A third problem concerns the stability of
self-concept Given that general self-concept presumably is an enduring
characteristic formed through many experiences over time, it should not
fluctuate much as a result of relatively brief experimental interventions.
Attempting to alter self-concept with a brief intervention to determine
whether changes in self-concept affect achievement may be difficult.
Similarly, changes in achievement may not have much immediate effect
on perceptions of one’s self-regard.
Bandura (1986) contends that general measures of psychological
functioning cannot predict with accuracy what people will do in specific
situations. Self-concept includes people’s perceptions of their
competencies (self-confidence) and their self-worth (self-esteem), both
of which vary for performances in different domains. One might, for
example, judge him- or herself highly capable in intellectual endeavours
and feel a great sense of self-worth, moderately competent in social
situations but feel inadequate, and low in competence in athletic pursuits
but not feel inadequate. Variability is often found among different
activities within the same domain. Within the intellectual domain, a pupil
might judge his or her competence as high in science and mathematics,
moderate in English and social studies, and low in French and Latin.
Further, there is no necessary relation between self-confidence and self-
esteem. Students may judge themselves competent in areas from which
they derive no pride; conversely, low perceived competence does not
necessarily result in low self-esteem.
As noted in the preceding section, self-concept researchers view the
self-concept as being in dynamic interplay with the environment—
reacting to it while simultaneously influencing it. This notion fits well
with Bandura’s (1986) contention that cognitions, behaviours and the
environment reciprocally influence one another (triadic reciprocality).
The best prediction of behaviour in a specific situation comes from
people’s self-perceptions within that domain, or their working self-
concepts in different areas of human functioning.
75
Dale H.Schunk
Domain-specific self-perceptions
Domain-specific self-concepts are far less crystallized than is academic
or general self-concept; therefore, the former are more readily modified
by experimental interventions. In our own research, young people
typically feel more self-confident about learning academic content after
participating in a training programme that includes instruction and
practice opportunities in the content The training programme makes
salient pupils’ perceptions of their capabilities in that academic area. A
broad construct such as academic self-concept is influenced by beliefs
about many different skills. Unless the skill taught is general and affects
many academic domains, academic self-concept is likely to show little
change due to brief experimental interventions.
I will illustrate the influence of a particular type of domain-specific
self-perception on school learning: perceived self-efficacy, or personal
beliefs about one’s capabilities to engage in activities necessary to attain
designated performance levels (Bandura 1982). Self-efficacy is
hypothesized to influence choice of activities. Pupils who hold a low
sense of efficacy for accomplishing a task may avoid it, whereas those
who believe they are capable should participate more eagerly. Self-
efficacy also affects motivation and achievement Especially when they
encounter difficulties, pupils who believe that they are capable of
learning ought to work harder, persist longer, and develop better skills,
than those who doubt their capabilities (Schunk, in press).
People acquire self-efficacy information from their actual
performances, vicarious (for example, observational) experiences, forms
of persuasion, and physiological indices. In general, one’s successes
raise efficacy and failures lower it, although once a strong sense of
efficacy is developed an occasional failure will not have much impact. In
school, pupils who observe similar peers perform a task are apt to believe
that they, too, are capable of performing it. Information acquired from
vicarious sources will have a weaker influence on self-efficacy than
performance-based information; when the two sources conflict, the
former will be outweighed by the latter. Pupils also receive persuasory
information from teachers (for example, ‘You can do this’). Positive
persuasory feedback enhances self-efficacy, but this increase will be
short-lived if pupils’ subsequent efforts turn out poorly. Children can
also derive efficacy information from such physiological indices as heart
rate and sweating. Anxiety symptoms before a test might signal that one
lacks the skills to succeed.
Information acquired from these sources does not influence self-
efficacy automatically, but rather is cognitively appraised (Bandura
1982). A self-efficacy model of cognitive skill learning is portrayed in
Figure 5.1. This model is derived from different theoretical traditions
76
including social cognitive learning, attribution, and instructional theory

(Weiner 1985; Winne 1985; Bandura 1986; Corno and Snow 1986).
Figure 5.1 Self-efficacy model of cognitive skill learning.
I have discussed this model in depth elsewhere (Schunk 1985b, and in

press). In this chapter I present the overall framework, along with some
supporting research evidence. I then discuss evidence for the predictive
utility of self-efficacy during cognitive skill learning. The article
concludes with implications of the research findings for teaching and
learning.
Self-efficacy and cognitive skill-learning

Pupils differ in aptitudes and prior experiences. Aptitudes include
general abilities, skills, strategies, interests, attitudes and personality
characteristics (Cronbach and Snow 1977). Educational experiences
derive from such influences as prior schools attended, interactions with
teachers and time spent on different subjects. Aptitudes and experiences
are related. For example, skilled readers typically perform well on
reading tasks, which earns them teacher praise and high grades. In turn,
these outcomes lead pupils to develop greater interest in reading, which
further enhances skills.
At the outset of a learning endeavour we may speak of self-efficacy
for learning, acquiring knowledge, developing skills or mastering
material. Aptitudes and prior experiences will affect pupils’ initial beliefs
about their learning capabilities. Pupils who previously have performed
well in a content area are apt to believe that they are capable of further
learning; those who have experienced difficulties may doubt their
capabilities. At the same time, efficacy is not a mere reflection of
aptitudes and prior experiences. Using pupils of high, average and low
mathematical ability, Collins (1982) found children of high and low
mathematical self-efficacy within each ability level. Pupils solved
problems and could rework those they missed. Ability was positively
related to skilful performance, but regardless of ability level, children
with higher efficacy solved more problems correctly and chose to rework
more than they had missed.
I discuss task engagement variables in a moment. While participating
in learning activities, children derive efficacy cues that signal how well
77
Dale H.Schunk
they are learning and that they use to assess efficacy for continued
learning. In turn, higher efficacy for learning enhances motivation and
skill acquisition. Some important cues are performance outcomes,
outcome patterns, attributions, social comparisons, persuader credibility
and bodily symptoms.
Performance outcomes are influential cues. Successes general raise
self-efficacy and failures lower it; however, an occasional failure after
many successes may not have much impact, nor should one success
after many failures (Schunk, in press). Early learning is often fraught
with failures, but the perception of progress promotes efficacy.
Outcome patterns are important; efficacy will not be aided much if
pupils believe that their progress is slow or that their skills have
stabilized at low levels.
Attributions, or perceived causes of successes and failures, influence
efficacy in important ways. Achievement outcomes are often attributed to
such causes as ability, effort, task difficulty and luck (Frieze 1980;
Weiner 1985; Rogers, this volume). Children view effort as the prime
cause of outcomes. With development, ability attributions become
increasingly important influences on expectancies, and the role of effort
declines in importance (Nicholls 1978; Harari and Covington 1981).
Success achieved with great effort should raise efficacy less than if
minimal effort is required, because the former implies that skills are not
well developed (Bandura 1982).
Children also derive cues from social comparisons. As stated
earlier, Festinger (1954) hypothesized that, where objective standards
of behaviour are unclear or unavailable, observers evaluate
themselves through comparisons with others, and that the most
accurate self-evaluations derive from comparisons with those who are
similar in the ability or characteristic being evaluated. Pupils
frequently compare their performances with those of their peers, and
feel more (less) efficacious when they believe that they are
accomplishing more (less) work. Peers are also important models,
and observing models is a form of social comparison. Observing
similar peers improving their skills instills a sense of efficacy for
learning, whereas observed failures cast doubt on pupils’ capabilities
to succeed (Schunk 1985b). Similarity is based on perceived
competence or on such personal attributes as age, sex, and ethnic
background (Rosenthal and Bandura 1978).
Persuader credibility is important because pupils may experience
higher efficacy when they are told they are capable of learning by a
trustworthy source (for example, the teacher), whereas they may
discount the advice of less credible sources. Pupils may also discount
otherwise credible sources if they believe that the sources do not fully
understand the nature of the task demands (for example, difficult for
78
pupils to comprehend) or the effect of contextual factors (for example,

too many distractions).
Bodily symptoms serve as physiological cues for appraising
efficacy. Sweating and trembling may signal that children are not
capable of learning. Children who notice that they are reacting in less-
agitated fashion to academic tasks may feel more efficacious about
learning.
Task engagement refers to pupils’ cognitive activities (attending,
rehearsing, processing and integrating information), along with their
verbalizations and behaviours, that are focused on academic tasks
(Corno and Mandinach 1983). While pupils are engaged in tasks,
variables associated with the classroom instructional context give rise
to cues that they use to appraise their self-efficacy for learning. Some
influential task engagement variables are the purpose of instruction,
content difficulty, methods and materials used in instruction,
strategies taught to pupils, performance feedback, models, goal
setting, attributional feedback, and rewards. Rather than discuss each
of these in depth, I will focus on research investigating the effects of
attributional feedback, goal setting, strategy training and peer
models.
Research evidence
Attributional feedback.
Attributional feedback, which links pupils’ successes and failures with
one or more causes, is a persuasive source of efficacy information.
Young children stress the role of effort. Although ability information
becomes more important with development (Nicholls 1978), effort
feedback motivates children of different ages. Being told that one can
achieve better results through harder work (that is, effort feedback for
prior difficulties) motivates one to do so and conveys that one possesses
the necessary capabilities to succeed (Dweck 1975; Andrews and
Debus 1978).
Effort feedback for prior successes supports children’s perceptions of
their progress in learning, sustains motivation and increases self-efficacy for
continued learning. In an early study (Schunk 1982) children periodically
received effort feedback as they solved problems during a subtraction
instructional programme. Some children had their prior achievement linked
with effort (‘You’ve been working hard’), others had their future
achievement linked with effort (‘You need to work hard’), and those in a
third condition did not receive effort feedback. Linking prior achievement
with effort led to the highest motivation (rate of problem-solving during the
instructional programme), post-test subtraction skill and self-efficacy.
79
Dale H.Schunk
The effects of ability feedback were compared with those of effort

feedback using a similar methodology (Schunk 1983a). Children
periodically received ability feedback for prior successes (‘You’re good
at this’), effort feedback (‘You’ve been working hard), ability and effort
(combined) feedback, or no feedback. Children who received only
ability feedback demonstrated higher post-test self-efficacy and skill
compared with the effort-only and the ability-plus-effort conditions. The
latter subjects judged their effort expenditure during training greater than
did ability-only children. Children in the combined condition may have
discounted some ability information in favour of effort. They might have
wondered how able they really were if they had to continue to work hard
to succeed.
Schunk (1984) tested the idea that the sequence of attributional
feedback influences achievement outcomes. Children lacking
subtraction skills received instruction and solved problems over sessions.
One group periodically received ability feedback, a second group
received effort feedback, a third condition was given ability feedback
during the first half of training and effort feedback during the second
half, and for a fourth condition this sequence was reversed. Ability
feedback for early successes led to higher ability attributions, post-test
self-efficacy and subtraction skill, compared with effort feedback. These
results suggest that early ability feedback instilled higher expectations
for continued learning.
Schunk and Cox (1986) explored how the sequence of effort feedback
affected learning disabled pupils’ motivation, self-efficacy, and
subtraction skill. During an instructional programme, pupils received
effort feedback during the first half of the programme, effort feedback
during the second half, or no effort feedback. Each type of effort
feedback promoted post-test self-efficacy and skill better than no
feedback, but effort feedback during the first half enhanced effort
attributions for success. Given pupils’ learning disabilities, the effort
feedback for early or later successes may have seemed highly credible,
because they realistically had to expend effort to succeed. They may have
interpreted the effort feedback as indicating that they were becoming
more skilful and were capable of further learning. Over an extended
period effort feedback might lower self-efficacy, because as pupils
become more skilful they might wonder why they still have to work hard
to succeed.
Goal-setting
Goal-setting involves comparing one’s present performance against a
standard. When children pursue a goal, they experience heightened self-
efficacy for attaining it as they observe their goal progress. A sense of
learning efficacy sustains task motivation. Goals exert their effects
80
through their properties: specificity, difficulty level, proximity (Locke, et

al. 1981; Bandura and Cervone 1983). Goals that incorporate specific
performance standards are more likely to raise learning efficacy because
progress toward an explicit goal is easier to gauge. General goals (for
example, ‘Do your best’) do not enhance motivation. Although pupils
initially may doubt their capabilities to attain goals they believe are
difficult, working toward difficult goals builds a strong sense of efficacy,
because they offer more information about learning capabilities than
easier goals. Goals are also distinguished by how far they project into the
future. Proximal goals, which are close at hand, result in greater
motivation than more distant goals. As pupils observe their progress
toward a proximal goal, they are apt to believe that they are capable of
further learning.
To test the hypothesis that proximal goals enhance children’s
achievement strivings, Bandura and Schunk, (1981) presented children
with a subtraction instructional programme over a number of sessions.
The programme consisted of sets of material that included instruction
and problems to solve. Some children pursued a proximal goal of
completing one set each session; a second group was given a distant goal
of completing all sets by the end of the last session; a third group (general
goal) was advised to work productively. Proximal goals heightened task
motivation, and led to the highest post-test subtraction skill and self-
efficacy. The distant goal resulted in no benefits compared with the
general goal.
These findings support the idea that when children can easily gauge
their progress against a goal, the perception of improvement enhances
self-efficacy. Similar results were obtained during a division
instructional programme in which children individually solved problems
over sessions (Schunk 1983b). Half of the children were given goals each
session, whereas the other half were advised to work productively. This
study also explored how providing social comparative information about
other children’s performances affected achievement outcomes. Within
each goal conditions, half of the children were informed of the number of
problems that other similar children had completed (which matched the
session goal); the other half were not given comparative information. The
comparative information was designed to convey that the goals were
attainable.
Goals led to higher post-test self-efficacy; comparative information
promoted motivation during the instructional programme. Children who
received both goals and comparative information demonstrated the
highest post-test division skill. These results suggest that providing
children with a goal and information that it is attainable raises self-
efficacy for learning, which contributes to more productive instructional
performance and greater skill development.
81
Dale H.Schunk
Schunk (1985a) tested the idea that self-set goals promote

achievement striving better than externally-imposed goals and no goals.
The rationale for this study derived from the idea that participation in
goal-setting raises goal commitment, which is necessary for goals to
enhance performance (Locke et al. 1981). Children classified as learning
disabled in mathematics received subtraction instruction and practice
opportunities over sessions. Some children set performance goals each
session, others had comparable goals assigned, and children in a third
condition received instruction but no goals. Children in the two goal
conditions judged their expectancy of goal attainment at the start of each
session.
Self-set goals led to the highest post-test self-efficacy and subtraction
performance. Although children in the two goal conditions demonstrated
higher task motivation during the instructional sessions, children in the
self-set condition judged their expectancy of goal attainment higher. The
latter finding suggests that allowing pupils to set their learning goals
enhanced their self-efficacy for attaining them.
Strategy instruction
Much research shows that pupils benefit from instruction on strategies,
or cognitive plans oriented toward improving performance (Baker and
Brown 1984; Paris et al. 1984). Strategy instruction also influences self-
efficacy. The belief that one understands and can apply a strategy that
enhances achievement leads to a perceived sense of control over learning
outcomes and higher self-efficacy (Licht and Kistner 1986; Schunk, in
press).
In learning a strategy, pupils benefit from verbalizing aloud the
component steps while applying them to a task. Verbalization
facilitates learning because it directs attention to important task
features, assists strategy encoding and retention, and helps pupils
work in a systematic fashion (Fuson 1979; Harris 1982). Verbalization
is most beneficial for pupils who typically perform in a deficient
manner, for example, impulsive, remedial, learning disabled or
emotionally disturbed children (Borkowski and Cavanaugh 1979;
Hallahan et al. 1983).
To test these ideas, Schunk and Rice (1985) presented language-
deficient children in grades four and five (9- and 10-year-olds) with
reading comprehension instruction. Within each grade, half of the
children verbalize the steps in a strategy prior to applying them to
passages; the other half received strategy instruction but did not verbalize
the steps. This six-step strategy was as follows:
What do I have to do? (1) Read the questions; (2) read the
story; and (3) look for key words; (4) reread each question,
82
and (5) answer that question; (6) reread the story if I don’t
know the answer.
Strategy verbalization led to higher reading comprehension, self-

efficacy, and ability attributions across grades. The latter finding
suggests that strategy verbalization may enhance self-efficacy through its
effect on ability attributions, by leading to an increase in one’s beliefs
that one has the ability required for success.
Similar results were obtained by Schunk and Rice (1984). Remedial
readers in the early years of schooling received instruction in listening
comprehension. Strategy verbalization led to higher self-efficacy
across grades, and promoted performance among the older children in
the sample but not among the youngest (aged around 7 years). Perhaps
the demands of verbalization, along with those of the comprehension
task itself, were too complex for the youngest children. These children
may have focused their efforts on the comprehension task, which
would have interfered with strategy encoding and retention.
Subjects in the Schunk and Cox (1986) study received one of three
verbalization treatments. One group verbalized aloud subtraction
solution steps and their application to problems (continuous
verbalization), a second group verbalized aloud during the first half of
the instructional programme but not during the second half
(discontinued verbalization), and a third group did not verbalize (no
verbalization). Continuous verbalization led to higher post-test self-
efficacy and skill than discontinued and no verbalization, which did not
differ.
It is possible that when instructed no longer to verbalize aloud,
discontinued verbalization children had difficulty internalizing the
strategy; they may not have used covert instructions to regulate their
performances. They may also have believed that although the strategy was
useful, other factors (for example, effort) were more important for solving
problems (Fabricius and Hagen 1984). We have found that emphasizing to
children the value of the strategy as a means of improving performance
produces greater increases in children’s self-efficacy and comprehension
performance compared with strategy instruction without information on
strategy value (Schunk and Rice 1987).
Peer models
Models constitute a vicarious source of efficacy information. Perceived
similarity of observers and models is an important cue. Models who are
similar or slightly higher in competence provide the best information.
Students who observe similar peers learn a task are apt to believe that
they can learn as well (Schunk 1987). Peer models may exert more
beneficial effects on self-efficacy than teacher models, especially among
83
Dale H.Schunk
students with learning problems who doubt that they are capable of
attaining the teacher’s level of competence.
These ideas were tested with primary age school children who had
experienced learning problems in mathematics (Schunk and Hanson
1985). Children observed videotapes portraying an adult teacher and a
same-sex peer model. The teacher repeatedly provided instruction on
subtraction with regrouping, after which the model solved problems.
Other pupils viewed videotapes that portrayed only the teacher
presenting instruction, and some pupils were not exposed to models.
After viewing the tapes, children who had observed a peer model judged
perceived similarity in competence to the model. All pupils assessed self-
efficacy for learning to subtract, and then received subtraction
instruction.
This study also investigated the effects of mastery and coping
models: children exposed to a peer observed either a mastery or coping
model. The peer mastery model easily grasped subtraction operations,
solved all problems correctly, and verbalized positive achievement
beliefs reflecting high self-efficacy, high ability, low task difficulty, and
positive attitudes. The peer coping model initially made errors and
verbalized negative achievement beliefs, but over time made fewer
errors and began to verbalize coping statements (for example, ‘I’ll have
to work hard on this one’). The coping model’s problem-solving
behaviours and verbalizations eventually matched those of the mastery
model.
Observing a peer model enhanced self-efficacy for learning more than
observing the teacher model or not observing a model; teacher model
children judged self-efficacy higher than no-model children. Similar
results were obtained on post-test measures of self-efficacy and
subtraction skill. There were no differences on any measure between the
peer mastery and coping conditions, which may have been due to pupils’
prior classroom experiences with subtraction. Even though their
successes had been limited to problems without regrouping, they none
the less had these experiences to draw on and may have concluded that if
the peer could learn to regroup, they could as well.
Follow-up research comprising two experiments further explored the
effects of mastery and coping models (Schunk et al. 1987). Low
achieving children observed videotapes portraying a peer model
demonstrate either rapid (mastery model) or gradual (coping model)
acquisition of skill in adding and subtracting fractions. These children
had experienced little, if any, prior success with fractions. We expected
that children would perceive the coping model’s gradual learning to be
more similar to their typical performances than the rapid learning of the
mastery model, and thereby experience higher self-efficacy for learning
after observing coping models.
84
Experiment 2 also investigated the effects of a number of models:

children observed either one or three same-sex peer models display
mastery or coping behaviours. Multiple models presumably increase
the probability that observers will perceive themselves as similar to at
least one of the models, which should enhance self-efficacy. Because
subjects had encountered difficulties learning mathematical skills, we
felt that they might be more likely to discount the successes of a single
model than the diverse instances of successful learning displayed by
multiple models.
In both experiments children who observed coping models judged
themselves more similar in competence to the models than did children
who observed mastery models. In Experiment 1, observation of a coping
model led to higher self-efficacy, skill, and instructional session
performance, compared with observation of a mastery model. In
Experiment 2, children in the single coping model, multiple coping
model and multiple mastery model conditions demonstrated higher self-
efficacy, skill and instructional session performance, compared with
children who observed a single mastery model. These latter results show
that the benefits of multiple models did not depend on perceptions of
similarity in competence. Similarity in competence may be a more
important source of efficacy information when children are exposed to a
single model and have a less-diverse set of modelled cues to use in
judging self-efficacy.
Predictive utility of self-efficacy

The predictive utility of self-efficacy for learning is determined by
relating this measure to the number of problems that children complete
during the instructional sessions. Positive correlations have been
obtained (range=+0.33 to +0.42), which indicate that the higher that
children judge self-efficacy for learning, the more problems they
complete during the sessions. More rapid problem-solving has not been
attained at the expense of accuracy. Similar correlations have been
obtained using the proportion of problems solved correctly. Self-efficacy
for learning also relates positively to post-test self-efficacy and skill
(range=+0.46 to +0.90).
The predictive utility of pre-test efficacy often is inadequate because
students lack skills and judge efficacy low. In contrast, there is greater
variation in students’ post-test judgements and in their demonstrated
skills. Studies have yielded positive correlations (range=+0.27 to +0.84)
between post-test efficacy and skill.
We have also tested a causal model of achievement to determine the
interrelationship of instructional treatment, self-efficacy, persistence and
skill. Not surprisingly, we have found that instruction produces gains in
85
Dale H.Schunk
pupils’ skills and self-efficacy and that higher self-efficacy leads to

greater skill and persistence.
Educational implications
The preceding results suggest that teachers help to promote children’s
self-concepts by structuring the curriculum so that pupils work on tasks
they can accomplish and receive positive information from teachers and
others (peer models) that they are capable of performing well. Teaching
pupils to use learning strategies and to set challenging but attainable
goals also fosters motivation, skill development and perceived
competence.
Many of the procedures discussed in this article can be implemented
easily in schools. For example, the strategy instruction procedures were
applied in children’s regular reading groups (Schunk and Rice 1984,
1985, 1987). Teaching pupils to use a comprehension strategy by having
them verbalize steps fits well with the suggestion by researchers to teach
strategies to pupils, especially those with learning problems (Borkowski
and Cavanaugh 1979; Raphael and McKinney 1983; Brown et al. 1984;
Paris et al. 1984).
Attributional feedback is easily delivered by teachers while pupils are
engaged in seatwork activities. Feedback that signals progress in
learning validates pupils’ beliefs that they are acquiring skills and
enhances motivation for further learning. It is important that the feedback
be viewed as credible by pupils. Effort feedback for success at a task that
they believe is easy may lead them to wonder whether the teacher thinks
they are low in ability (Weiner et al. 1983). Pupils will discount ability
feedback after they have had to struggle to succeed. Effort feedback is
credible when pupils have to work hard to succeed, whereas ability
feedback is credible when they succeed with little effort.
Goal-setting is incorporated by teachers with lesson goals for pupils.
Contracts specify learning or performance goals. Goal-setting
conferences, in which teachers meet periodically with pupils to discuss
their goal attainment and to set new goals, enhance achievement and
capability self-evaluations (Gaa 1973). Short-term goals are especially
beneficial with young children, as well as with pupils with learning
problems because they provide concrete standards against which to
gauge progress.
Peer models are often incorporated into classroom procedures when
teachers select one or more pupils to demonstrate a skill to other class
members. The typical practice is to choose peers who master skills
readily—mastery models. Among pupils who have experienced some
learning difficulties, peers who also have encountered learning problems
but who have mastered the present skill may make better models. Peers
86
also could model such coping behaviours as increased concentration and

hard work. While pupils are engaged in seatwork, teachers might provide
social comparative information (for example, ‘See how well Kevin is
doing? I’m sure that you can do just as well’). Teachers need to ensure
that learners view the comparative performances as attainable; judicious
selection of referent pupils is necessary.
Peers also enhance observers’ self-efficacy in small groups.
Successful groups in which each member is responsible for some aspect
of the task and members share rewards based on their collective
performance reduce negative ability-related social comparisons (Ames
1984). Teachers need to select tasks carefully, because unsuccessful
groups will not raise efficacy.
Strain and his colleagues have successfully used peers as social skill
trainers with withdrawn children (Strain et al. 1981). Peers are trained to
initiate social contacts with verbal signals and motor responses. Such
initiations increase withdrawn children’s subsequent social initiations
and gains often generalize to classrooms. A less formal application
involves pairing a socially competent peer with a less competent child to
work on a task. The opportunity for social interaction within the dyad
promotes the less competent child’s social skills (Mize et al. 1985).
The use of peers as instructional agents is found in tutoring
programmes. Despite some methodological problems in studies,
research shows that tutoring results in academic gains by tutor and
tutee (Feldman et al. 1976). Peer instructors are also helpful where
their teaching strategies fit well with learners’ capabilities or the skills
being taught. Adult teachers typically employ verbal instruction and
relate information to be learned to other material, whereas peer
teachers use non-verbal demonstrations more often and link instruction
with specific items (Ellis and Rogoff 1982). Peer instruction seems
beneficial for students who may not process verbal material
particularly well.
Given that diverse domain-specific self-concepts contribute to the
general academic self-concept, we might expect that the development of
self-confidence in one domain would show little transfer to other
domains and have little impact on academic self-concept. One means of
improving transfer is to teach pupils strategies that can easily be
modified for use on different types of content. This point fits well with
the suggestion that transfer is facilitated when pupils are trained using
multiple tasks (Borkowski and Cavanaugh 1979). Pupils who believe
that they have learned strategies that will improve their performance in
different domains are apt to demonstrate higher achievement and hold
higher academic self-concepts. Many task engagement variables (for
example, goal-setting, attributional feedback) are general in the sense
that they are not limited in use to only one academic domain.
87
Dale H.Schunk
References
Ames, C. (1984) ‘Competitive, co-operative, and individualistic goal structures: a
cognitive-motivational analysis’, in R.Ames and C.Ames (eds) Research on
Motivation in Education: Student Motivation, vol. 1, Orlando: Academic
Press, pp. 177–207.
Andrews, G.R. and Debus, R.L. (1978) ‘Persistence and the causal perception of
failure: modifying cognitive attributions’, Journal of Educational Psychology
70: 154–66.
Baker, L. and Brown, A.L. (1984) ‘Metacognitive skills and reading’, in P.D.
Pearson (ed.) Handbook of Reading Research, New York: Longman, pp. 353–
94.
Bandura, A. (1982) ‘Self-efficacy mechanism in human agency’, American
Psychologist 37: 122–47.
Bandura, A. (1986) Social Foundations of Thought and Action: A Social
Cognitive Theory, Englewood Cliffs, NJ: Prentice-Hall.
Bandura A. and Cervone, D. (1983) ‘Self-evaluative and self-efficacy
mechanisms governing the motivational effects of goal systems’, Journal of
Personality and Social Psychology 45: 1017–28.
Bandura, A. and Schunk, D.H. (1981) ‘Cultivating competence, self-efficacy, and
intrinsic interest through proximal self-motivation’. Journal of Personality
and Social Psychology 41: 586–98.
Borkowski, J.G. and Cavanaugh, J.C. (1979) ‘Maintenance and generalization of
skills and strategies by the retarded’, in N.R.Ellis (ed.) Handbook of Mental
Deficiency, Psychological Theory and Research, 2nd edn, Hillsdale, NJ:
Erlbaum, pp. 569–617.
Brown, A.L., Palincsar, A.S. and Armbruster, B.B. (1984) ‘Instructing
comprehension-fostering activities in interactive learning situations’, in
H.Mandl, N.L.Stein and T.Trabasso (eds) Learning and Comprehension of
Text, Hillsdale, NJ: Erlbaum, pp. 255–86.
Cantor, N. and Kihlstrom, J.F. (1987) Personality and Social Intelligence,
Collins, J. (1982) Self-efficacy and Ability in Achievement Behavior, paper
presented at the meeting of the American Educational Research Association,
New York, March.
Corno, L. and Mandinach, E.B. (1983) ‘The role of cognitive engagement in
classroom learning and motivation’, Educational Psychologist 18: 88–
108.
Corno, L. and Snow, R.E. (1986) ‘Adapting teaching to individual differences
among learners’, in M.C. Wittrock (ed.) Handbook of Research on Teaching,
3rd edn, New York: Macmillan, pp. 605–29.
Covington, M.V. (1984) ‘The self-worth theory of achievement motivation:
findings and implications’, Elementary School Journal 85: 5–20.
Cronbach, L.J. and Snow, R.E. (1977) Aptitudes and Instructional Methods, New
York: Irvington.
Dweck, C.S. (1975) ‘The role of expectations and attributions in the alleviation of
learned helplessness’, Journal of Personality and Social Psychology 31: 674–
85.
Ellis, S. and Rogoff, B. (1982) ‘The strategies and efficacy of child versus adult
teachers’, Child Development 53: 730–5.
Fabricius, W.V. and Hagen, J.W. (1984) ‘Use of causal attributions about recall
performance to assess metamemory and predict strategic memory behavior in
young children’, Developmental Psychology 20: 975–87.
88
Feldman, R.S., Devin-Sheehan, L. and Allen, V.L. (1976) ‘Children tutoring

children: a critical review of research’, in V.L. Allen (ed.) Children as
Teachers: Theory and Research on Tutoring, New York: Academic Press, pp.
235–52.
Festinger, L. (1954) ‘A theory of social comparison processes’. Human Relations
7: 117–40.
Frieze, I.H. (1980) ‘Beliefs about success and failure in the classroom’, in J.H.
McMillan (ed.) The Social Psychology of School Learning, New York:
Academic Press, pp. 39–78.
Fuson, K.C. (1979) ‘The development of self-regulating aspects of speech: a
review’, in G.Zivin (ed.) The Development of Self-regulation Through Private
Speech, New York: Wiley, pp. 135–217.
Gaa, J.P. (1973) ‘Effects of individual goal-setting conferences on achievement,
attitudes, and goal-setting behaviour’, Journal of Experimental Education 42:
22–8.
Hallahan, D.P., Kneedler, R.D. and Lloyd, J.W. (1983) ‘Cognitive
behaviour modification techniques for learning disabled children: self-
instruction and self-monitoring’, in J.D.McKinney and L.Feagans (eds)
Current Topics in Learning Disabilities, vol. 1, Norwood, NJ: Ablex, pp.
207–44.
Harari, O. and Covington, M.V. (1981) ‘Reactions to achievement behavior from
a teacher and student perspective: a developmental analysis’, American
Educational Research Journal 18: 15–28.
Harris, K.R. (1982) ‘Cognitive-behavior modification: application with
exceptional students’ Focus on Exceptional Children 15: 1–16.
Higgins, E.T. (1981) ‘Role taking and social judgment: alternative developmental
perspectives and processes’, in J.H.Flavell and L.Ross (eds) Social Cognitive
Development: Frontiers and Possible Futures, Cambridge: Cambridge
University Press, pp. 119–53.
Licht, B.G. and Kistner, J.A. (1986) ‘Motivational problems of learning-
disabled children: individual differences and their implications for
treatment’, in J.K.Torgesen and B.W.L.Wong (eds) Psychological and
Educational Perspectives on Learning Disabilities, Orlando: Academic
Press, pp. 225–55.
Locke, E.A., Shaw, K.N., Saari, L.M. and Latham, G.P. (1981) ‘Goal
setting and task performance: 1969–1980’, Psychological Bulletin 90:
125–52.
Markus, H. and Nurius, P. (1986) ‘Possible selves’, American Psychologist 41:
954–69.
Markus, H. and Wurf, E. (1987) ‘The dynamic self-concept: a social
psychological perspective’. Annual Review of Psychology 38: 299–
337.
Marsh, H.W. and Shavelson, R. (1985) ‘Self-concept: its multifaceted,
hierarchical structure’. Educational Psychologist 20: 107–23.
Mead, G.H. (1934) Mind, Self and Society, Chicago: University of Chicago Press.
Mize, J., Ladd, G.W. and Price, J.M. (1985) ‘Promoting positive peer relations
with young children: rationales and strategies’. Child Care Quarterly 14: 211–
37.
Montemayor, R. and Eisen, M. (1977) ‘The development of self-conceptions
from childhood to adolescence’. Development Psychology, 13: 314–19.
Mosatche, H.S. and Bragonier, P. (1981) ‘An observational study of social
comparison in preschoolers’. Child Development 52: 376–8.
89
Dale H.Schunk
Nicholls, J.G. (1978) ‘The development of the concepts of effort and ability,
perception of academic attainment, and the understanding that difficult tasks
require more ability’, Child Development 49: 800–14.
Nicholls, J.G. (1983) ‘Conceptions of ability and achievement motivation: a
theory and its implications for education’, in S.G.Paris, G.M.Olson and
H.W.Stevenson (eds) Learning and Motivation in the Classroom, Hillsdale,
NJ: Erlbaum, pp. 211–37.
Paris, S.G., Cross, D.R. and Lipson, M.Y. (1984) ‘Informed strategies for
learning: a program to improve children’s reading awareness and
comprehension’. Journal of Educational Psychology 76: 1239–52.
Pintrich, P.R., Cross, D.R., Kozma. R.B. and McKeachie, W.J. (1986)
‘Instructional psychology’, Annual Review of Psychology 37: 611–51.
Raphael, T.E. and McKinney, J. (1983) ‘An examination of fifth- and eighth-
grade children’s question-answering behavior: an instructional study in
metacognition’, Journal of Reading Behavior 15: 67–86.
Rogers, C.G. (1990) ‘Motivation in the primary years’, this volume.
Rosenberg, M., and Kaplan, H.B. (1982) Social Psychology of the Self-concept,
Arlington Heights, Ill.: Harlan Davidson.
Rosenthal, T.L. and Bandura, A. (1978) ‘Psychological modeling: theory and
practice’, in S.L.Garfield and A.E.Bergin (eds) Handbook of Psychotherapy
and Behavior Change: An Empirical Analysis, 2nd edn. New York: Wiley, pp.
621–58.
Ruble, D.N. (1983) ‘The development of social-comparison processes and their
role in achievement-related self- socialization’, in E.T.Higgins, D.N.Ruble
and W.W.Hartup (eds) Social Cognition and Social Development, New York:
Cambridge University Press, pp. 134–57.
Ruble, D.N., Boggiano, A.K., Feldman, N.S. and Loebl, J.H. (1980)
‘Developmental analysis of the role of social comparison in self-evaluation’,
Developmental Psychology 16: 105–15.
Ruble, D.N., Feldman, N.S. and Boggiano, A.K. (1976) ‘Social comparison
between young children in achievement situations’, Developmental
Psychology 12: 191–7.
Rumelhart, D.E. and Ortony, A. (1977) ‘The representation of knowledge
in memory’, in R.C.Anderson, R.J.Spiro and W.E.Montague (eds)
Schooling and the Acquisition of Knowledge, Hillsdale, NJ: Erlbaum,
pp. 99–135.
Scheirer, M.A. and Kraut, R.E. (1979) ‘Increasing educational achievement via
self concept change’. Review of Educational Research 49: 131–50.
Schunk, D.H. (1982) ‘Effects of effort attributional feedback on children’s
perceived self-efficacy and achievement’, Journal of Educational Psychology
74: 548–56.
Schunk, D.H. (1983a) ‘Ability versus effort attributional feedback: differential
effects on self-efficacy and achievement’, Journal of Educational Psychology
75: 848–56.
Schunk, D.H. (1983b) ‘Developming children’s self-efficacy and skills: the roles
of social comparative information and goal setting’, Contemporary
Educational Psychology 8: 76–86.
Schunk, D.H. (1984) ‘Sequential attributional feedback and children’s
achievement behaviors’, Journal of Educational Psychology 76: 1159–69.
Schunk, D.H. (1985a) ‘Participation in goal setting: effects on self-efficacy
and skills of learning disabled children’, Journal of Special Education 19:
307–17.
90
Schunk, D.H. (1985b) ‘Self-efficacy and classroom learning’, Psychology in the

Schools 22: 208–23.
Schunk, D.H. (1987) ‘Peer models and children’s behavioral change’, Review of
Educational Research 57: 149–74.
Schunk, D.H. (in press) ‘Self-efficacy and cognitive skill learning’, in C.Ames
and R.E.Ames (eds) Research on Motivation in Education, vol. 3, San Diego:
Academic Press.
Schunk, D.H. and Cox, P.D. (1986) ‘Strategy training and attributional feedback
with learning disabled students’, Journal of Educational Psychology 78: 201–
9.
Schunk, D.H. and Hanson, A.R. (1985) ‘Peer models: influence on children’s
self-efficacy and achievement’, Journal of Educational Psychology 77: 313–
22.
Schunk, D.H., Hanson, A.R. and Cox, P.D. (1987) ‘Peer model attributes and
children’s achievement behaviors’, Journal of Educational Psychology 79:
54–61.
Schunk, D.H. and Rice, J.M. (1984) ‘Strategy self-verbalization during remedial
listening comprehension instruction’, Journal of Experimental Education 53:
49–54.
Schunk, D.H. and Rice, J.M. (1985) ‘Verbalization of comprehension strategies:
effects on children’s achievement outcomes’, Human Learning 4: 1–10.
Schunk, D.H. and Rice, J.M. (1987) ‘Enhancing comprehension skill and self-
efficacy with strategy value information’, Journal of Reading Behavior 19:
285–302.
Shavelson, R.J. and Bolus, R. (1982) ‘Self-concept: the interplay of theory and
methods’, Journal of Educational Psychology 74: 3–17.
Strain, P.S., Kerr, M.M. and Ragland, E.U. (1981) ‘The use of peer social
initiations in the treatment of social withdrawal, in P.S. Strain (ed.) The
Utilization of Classroom Peers as Behavioral Change Agents, New York:
Plenum Press, pp. 101–28.
Veroff, J. (1969) ‘Social comparison and the development of achievement
motivation’, in C.P.Smith (ed.) Achievement-related Motives in Children, New
York: Russell Sage Foundation, pp. 46–101.
Weiner, B. (1985) ‘An attributional theory of achievement motivation and
emotion’, Psychological Review, 92: 548–73.
Weiner, B., Graham, S., Taylor, S.E. and Meyer, W. (1983) ‘Social cognition in
the classroom’, Educational Psychologist 18: 109–24.
Winne, P.H. (1985) ‘Cognitive processing in the classroom’, in T.Husen and
T.N.Postlethwaite (eds) The International Encyclopedia of Education, vol. 2,
Oxford, England: Pergamon Press, pp. 795–808.
Wylie, R.C. (1979) The Self-Concept, vol. 2Theory and Research on Selected
Topics, Lincoln, NE: University of Nebraska Press.
91
Chapter six
Motivation in the primary years
Colin Rogers
The following chapter offers a review of recent work on motivation and
its development throughout the primary years. The last decade or so has
seen a marked increase in interest among researchers in this topic,
following from the development of a number of cognitive models. With
their emphasis on the style of thinking characteristic of different
individuals and the impact of the environment upon these styles, such
theories lend themselves to further development within an educational
context.
The work reviewed by Rogers in the following chapter draws
attention to a number of key aspects of the individual child’s experience
of schooling by placing an emphasis on the qualitative aspects of the
child’s motivational responses. The articulation of the theories that are
being developed can serve as a useful aid to teachers engaged in the
complex task of developing their own responses to the pupil.
Some of the work discussed here also illustrates the ways in which
motivational theory might be used to investigate some of the functions
and modes of operation of the hidden curriculum, in this instance by
investigating the child’s acquisition of certain basic attitudes towards
different parts of the school curriculum. It is important to note that
Rogers’s conclusion is that modes of presentation can have an effect on
the development of these attitudes that is by no means unimportant when
set beside the effects of the content of the curriculum area itself. Even
with a totally prescribed curriculum, teacher action matters.
There are a number of links here that the reader ought to look out for.
In particular the chapter by Schunk (Chapter 5) dealing with the self-
concept will contain a number of close parallels. The reader might also
like to consider some of the implications of the work reported here for the
development of gender differences discussed by Croll and Moses
(Chapter 11). The impact of various educational experiences might not
always be immediately apparent, but remains instead, dependent upon
the development of appropriate cognitive structures to set beside them.
92
Introduction
I suspect that for most primary school teachers a major perceived
advantage of their work conditions, in comparison with their colleagues
in the secondary sector, is that motivating their pupils is not seen to be a
major problem. The typical primary school child is seen to be interested,
lively and enthusiastic. As a primary teacher said to me recently:
The great thing about kids at this age is that they want to
learn. They do not have to be bullied or cajoled. They might
not all succeed in learning as much as I would like them to,
but the desire to try is there. I have always been put off of
secondary work by the pupils. So many of them seem to have
lost interest, they are detached, unco-operative. I don’t know
why that is, but it does seem to be a major difference.
Of course, it is clearly the case that there will exist considerable degrees
of variation between pupils at the primary level in terms of their
motivation. Some will be seen to be more strongly motivated than others,
but the general style of the majority of pupils is seen to be a positive and
desirable one. These impressions are substantially supported by research
(Nicholls 1984; Stipeck 1984). There is a well-documented decline in
motivation that seems to be related to the age of the child. Details of this
will be examined later, but at this stage it is worth noting that some of the
decline seems to stem from the child’s increasing capacity to analyse the
situation that he/she finds themselves to be in. However, it is also worth
noting (Stipeck 1984) that the organization of schooling for children at
different ages may in itself be partly responsible for these changes.
However, before we can begin to examine the changes that may occur in
motivation within a child over time, it is necessary to examine just what
is meant by the term motivation itself.
Again it is useful to begin with some impressions of the ways in which
teachers themselves might conceptualize motivation within their pupils.
My own experience of talking to teachers about motivation leads to the
conclusion that motivation in pupils is often assessed in quantitative
terms and in terms of the degree to which a pupil matches up to the
requirements of the school. Thus pupils are described as having a high
level of motivation, or as lacking in motivation. The well-motivated pupil
is described as being easy to work with, obedient, unlikely to be late or to
miss periods of schooling. The poorly-motivated pupils are the ones who
will be absent (in spirit if not in body), likely to be in and out of trouble
on a regular basis and so on. Pen portraits of well-motivated pupils
written by practising teachers could almost have been the result of a
request to describe the ideal pupil from a teacher’s point of view.
93
Colin Rogers
Recent developments in motivational theory go some way in

challenging much, but not all, of the conceptions alluded to above
(Ames 1984; Weiner 1986). These recent developments are varied in
detail but share as a common reference point the view that the
perceptions of the individual child are a key element in defining and
determining the nature of the motivational process. Motivation
becomes more than just a fixed disposition that some people are lucky
to have more of than others and becomes instead an active and dynamic
process. The outcomes of this process will have an effect upon the way
in which the individual relates to the environment around him or her
but will at the same time be affected by that environment. Indeed Ames
(1987) has concluded from her work that any attempt to change the
motivation of pupils must give prominence to the role played by the
classroom environment. The implication is that motivation cannot
simply be regarded as something that the pupils bring with them to
school which, at best, the school might enhance and, at worst damage.
Rather motivation is the result of interaction between pupil and school
and all those elements that make up the child’s environment.
Motivation is created by the child’s experience of school.
A qualitative approach to motivation

In the development of my own thinking about the nature of motivation
during the primary years, the work of Atkinson (Atkinson 1964;
Atkinson and Raynor 1978; Weiner 1980) has been influential.
Atkinson and his colleagues have been concerned to develop a model
of the motivation process that would ultimately allow precise
predictions to be made about the type of behaviour that an individual
might engage in, and how that behaviour results from changing
motivational states. In many respects, the theory that has been
produced is of greater relevance to the psychological laboratory than it
is to the school classroom, and for this, and other reasons, it is not
likely to be have a major, direct bearing upon classroom practice.
However, Atkinson’s work is important for the way in which it draws
attention to the complexities inherent in motivational processes.
One particular aspect of Atkinson’s work is relevant here. Atkinson
has claimed that the motivational forces that direct behaviour in any
particular situation are a combination of elements of that person’s
personality and elements of the situation. Thus we would expect to find
fairly consistent differences between individuals across a range of
situations, to the degree that those people have differing personalities.
However, we would also expect to find variation in the responses of one
person as he or she moved from one context to the next. Some people are
94
generally well motivated, but even they will still appear to be less
motivated in some situations than others.
More specifically, Atkinson argued that the relevant personality
structure was itself complex and needed to be thought of in terms of a
number of related components. For Atkinson, each individual is
influenced by one aspect of personality concerned with their need to
achieve success. The stronger this personality trait, the more eager a
person will be to experience the fruits of success. The individual also
brings a second personality trait, however, one that is concerned with
failure, and the anxiety that this produces. The stronger this ‘fear of
failure’, the less likely the person will be to engage in tasks where
success, and therefore failure, is possible. The most certain way to avoid
failure is never to attempt to achieve anything. While nothing being
ventured leads to nothing being gained, it also ensures that we have no
reason to fear not reaching our targets.
Any situation in which success or failure are possible, that is, any
situation where one’s performance may be measured (by oneself or
others) against some standard of excellence, will thus give rise to
conflict. To the degree that our behaviour is governed by the need to
achieve success, we will wish to engage in the task in order to enhance
the prospect of being successful. To the degree that our behaviour is
governed by fear of failure we will wish to avoid the task in order to
diminish the prospect of failure. Atkinson’s work goes on to specify in
detail how this conflict may be resolved and how the nature of the
resolution will vary from person to person, and from situation to
situation. There is not the space here to discuss these details, but the
following should be noted. First, Atkinson is primarily concerned here
with the workings of intrinsic motivation. He is concerned with the fear
of failure and the desire to succeed in their own right. The failure may
be a completely private one that no-one else will be interested in, but
the fear of failure can still be invoked. The experience of failure can
still provide a threat to our self-esteem and sense of competence. In
many situations success and failure will also give rise to concerns with
aspects that are extrinsic to the task itself. If we are successful others
may reward us, others may hold us in higher esteem. If we are
unsuccessful we may be punished or lose out on some of the material
benefits of success. These concerns will also have an effect upon our
behaviour, and there will be other elements of personality structure that
will be relevant to these.
Second, Atkinson emphasizes the importance of attempting to
understand the perspective of the individual whose behaviour is of
interest to us. In order to avoid the worst consequences of failure, a pupil
may chose to persist at tasks that are clearly far too difficult. Persistently
undertaking and failing very difficult tasks is not educationally adaptive,
95
Colin Rogers
but Atkinson draws our attention to the possibility that such behaviour
may well be adaptive for the individual in that the very difficulty of the
task provides the pupil with some protection against the personal
consequences of failure. If the task is so difficult that failure has to be
reasonably expected, then it cannot be too threatening. How much more
threatening it would be to risk failure at a task that one ought to be able to
succeed at. Avoiding challenging tasks by sticking at those that are very
difficult, or very easy, may make little educational sense, but if one’s
concern is with avoiding failure, rather than with gaining success, it
might appear to be a very sensible strategy to adopt. In short, it would be
wrong to assume that the pupil whose behaviour does not seem to lend
itself very readily to the requirements of the educational system is not
motivated. Atkinson’s work suggests that such a pupil may be very
strongly motivated, but that their style of motivation is not one that the
school system will find congenial.
Perceptions of success and failure

While Atkinson’s work has drawn attention to the importance of
analysing situations from the perspectives of the individuals within them,
his theory does not really succeed in making available the means by
which this might be readily done within the classroom context. To this
extent the work of Bernard Weiner (1979, 1984, 1985, 1986) has already
proved to be significant and is likely to continue to influence research
into motivational processes within the classroom in the years to come.
Weiner has developed a view of the nature of the dynamics of
achievement motivation that is based upon one of the major theories
within social psychology—attribution theory. As this perspective
provides the basis for the research to be reported here later it will be
attended to in some detail.
Attribution theory is generally regarded as having its origins in the
work of Heider (1958) and in developments provided by people such as
Kelley (1983) and Jones and Davis (1965). The basic concerns of the
theory are to explore what has been referred to as ‘naïve psychology’,
that is, the everyday rules of thumb that people use in order to make sense
of the world around them and the people within it with whom they have
to deal. The sense that we make of the world around us, and of our own
role within it, will, according to attribution theorists, play a large part in
determining our own behaviour.
The major assumption made by attribution theorists is that a central
part of this process of making sense of the world involves causal
attributions. A trivial example can readily illustrate the point. If, in a busy
shopping street, someone steps on my foot, I will, of course, experience
some degree of pain. But what else might I expect or do? If I assume the
96
action to have been unintended, caused merely by the crowded condition

of the pavement, my reaction will be different than if I conclude that the
passer-by had stepped on my foot with the deliberate intention of causing
me harm. My own action (doing nothing in the first instance, running
away before the other person gets really nasty in the second) will owe
more to the attribution that I make about the cause of the other person’s
behaviour than it will to the action itself.
Weiner’s analysis of achievement motivation has been based on very
similar assumptions. In this case the event to be understood, in terms of
making attributions for its causes, will be an instance of success and
failure. As with the example above, the assumption is made that while the
success or failure in itself will produce given reactions often of an
emotional nature (Weiner 1985, 1986), it will be our understanding of the
cause of that success or failure that will determine many other aspects of
our response.
Clearly there are a number of causes to which a success or failure
event could be attributed. A large number of causes have been identified
by researchers (see, for example, Weiner 1985; Little 1985, and the
research of the present author to be discussed below). Even allowing for
the fact that two causes, ability and effort, generally emerge as the most
frequently cited ones, at least among adult, western samples, the number
of individual causes employed is large enough to make analysis of the
effects of different types of attributions difficult without some attempt at
categorization. Several such categorizations have been suggested and the
reader who is interested in observing the development of these is advised
to study the reviews published by Weiner himself (Weiner, et al. 1972;
Weiner 1979, 1983, 1984 and 1985).
For present purposes it is sufficient to dwell on two causal
dimensions that have been most frequently and most reliably identified,
those of stability and locus. Stability refers to the degree to which the
cause assumed to be responsible for an event is judged to have a
consistent and repeatable effect over time. Ability and effort, the two
most commonly-used individual causes, are judged to lie towards
opposite ends of this dimension. Ability is a stable cause of success or
failure. To the adult at least, one’s ability is seen to be stable and
enduring. If I perceive myself to lack a certain ability now, then I am
likely to believe that I will lack that ability on later occasions. It is
important to be clear as to the difference between ability and
performance. As a possible cause of performance differences, ability is
seen as something that underlies one’s performance level. If one’s
performance level rises as the result of practice, this will be due to
one’s now being able to make greater use of the ability that one
possesses. The underlying ability itself has not changed, just the degree
that one is able to make use of it.
97
Colin Rogers
It follows from this that performances are seen to have been primarily
the result of ability are performances that will themselves be expected to
be repeated. On the other hand, performances that are seen to be the
result of a less stable cause, such as effort, are believed to be less certain
to be repeated as the cause itself is liable to fluctuation. This has been
summarized by Weiner (1985:559) as the expectancy principle.
Atkinson’s work, briefly discussed above, indicated the importance of
the expectations that an individual has for success and failure. An
individual’s response to a success or failure situation would be determined
by their personality structure and the degree to which success was
expected. For Weiner, expectations are also important and are seen to be
directly related to the pattern of attributions that an individual makes. Take
the individual who, for whatever reason, is likely to make attributions for
success outcomes to the presence of high ability, and for failure outcomes
to low effort. What might be the main consequences of this? Under these
circumstances Weiner’s theory would predict that each success would lead
to higher expectations for further success as success is being attributed to a
stable cause. Each instance of failure, however, has a less marked effect on
expectations as these are attributed to unstable causes. Thus the overall
trend is for increasing expectations of success. The individual becomes
increasingly confident that engagement in achievement-related tasks will
lead to success outcomes. Conversely, the individual who attributes
success to high effort and failure to low ability will increasingly come to
expect failure. For it is now each instance of failure that leads to increased
expectations of the same, as these are the outcomes that are now assumed
to have been caused by stable factors. Successes do not do much to alter the
gloomy prognosis as they are seen to have been caused by an unstable
factor, effort. Such an individual is therefore likely to come increasingly to
expect to fail, and therefore to become less motivated to engage in
achievement-related tasks.
While effort and ability differ in terms of their degree of stability, they
are alike in that they are both seen to reside within the individual. In
terms of the second major dimension they are both said to have an
internal locus. Causes with an external locus are seen to reside outside of
the individual concerned. One might believe that one has failed at a task
because one lacks ability (an internally-located cause) or one might
believe that the task set was too difficult (an externally-located cause). In
many respects these two statements may appear very similar (they both
seem to imply that the level of ability that one has was insufficient to
perform a task of a given order of difficulty) but Weiner has claimed that
there is an important psychological difference between them. The nature
of this psychological difference has shown some development and
change as the theory itself has developed, but is currently held to concern
the degree to which one’s feelings of self-esteem will be involved.
98
Outcomes that are attributed to internally-located causes are more likely

to affect one’s feelings of self-esteem (positively for success and
negatively for failure) than are outcomes attributed to externally-located
causes. A failure due to task difficulty might be seen as one that is really
somebody else’s fault (the teacher should have known that I would not be
able to perform this task and therefore should never have set it) while one
that is internally attributed implies that the task was legitimate and that
one’s own inadequacies led to the failure.
There are other dimensions that have been examined and these too
will have their effects. The dimensions of stability and locus are,
however, the most important. As was again the case with Atkinson
earlier, it is possible to identify styles that are more or less adaptive
from the educational point of view. The individual who typically
attributes success outcomes to stable and internal causes will both
come to expect more success with greater confidence and to reap
greater gains in terms of enhanced self-esteem from those successes. If
they also attribute failures to unstable and external causes they can
reduce the expectation of failure and minimize the negative impact that
failure is always likely to have upon their self-esteem. The opposite
pattern, attributing success to unstable and external causes, and failure
to stable and internal ones will have the opposite effect. Here, one
would come increasingly to expect failure and to suffer the more
serious consequences for self-esteem.
The antecedents
Clearly if we were all able to choose our attributional patterns, in the light
of the above argument, we would all choose to attribute our successes
rather than our failures to stable and internal causes. This does not always
happen. Attributions are made in accordance with a series of constraints
that effectively determine the nature of the process. Again it is not the
present intention to give anything like a full review of the research work
that has been done here (for further details see Ames and Ames 1984;
Ames and Ames 1985; Weiner 1986).
One obvious consideration is the past history of the individual. There
is a clear connection between the degree to which an event is expected
and the degree to which it is likely to be attributed to stable causes
(Weiner 1986). Highly-expected events are more likely to be attributed to
a stable cause than ones that were unexpected. Thus the pupil who
expects success, through having had plenty of it in the past, is more likely
to attribute that success to a stable cause. We have already seen that such
an attribution is likely itself to have an effect on the further expectations
that the individual has. Expecting success promotes attributions to stable
causes, attributions to stable causes lead to increased expectations of
99
Colin Rogers
success. Conversely, low expectations for success are likely to lead to

attributions to unstable causes and these in turn further promote low
expectations. In both cases cyclic processes are established which are set
to become more deeply entrenched each time they operate, in the first
case beneficently, in the second viciously.
Of more interest to the classroom teacher, however, is the idea that the
pattern of attributions that a child makes, and the pattern of motivational
behaviour that follows from this, can be influenced by the activities that
take place within the school setting. One prime candidate for generating
any such effects is likely to the classroom teacher.
A useful introduction to the idea that the teacher can influence the
attributions made by the pupil is to be found in Bar-Tal (1982) and
further information and ideas can be obtained in Dusek (1985). Bar-Tal’s
summary of the research shows that teachers can influence the
attributions made by pupils via a number of different processes. The five
major categories into which these processes can be classified are given
by Bar-Tal as verbal appeals, instructions, reinforcements, verbal
feedback and direct references to causality. Some of these are explored
further below.
The impact of direct references to causality has been best
demonstrated by a number of studies (Andrews and Debus 1978; Dweck
1985; Maehr and Kleiber 1987) that have sought to assess the
effectiveness of attributional retraining. In these studies individual
children are selected on the grounds that they have some motivational
problem. The problem is then attended to by endeavouring to get the
child to attribute their failures to a lack of effort rather than to a lack of
ability. To the extent that a child can be persuaded to view a lack of effort
rather than ability as the causes of their failure, motivational
improvements are likely to take place. However, it is not yet entirely clear
just how important the direct attempt at changing the attributions actually
is. What does seem to be important is that the child comes to accept that
increased effort can produce increased performance outcomes. Under
some circumstances different experiences of success and failure might in
themselves be sufficient to bring about this change in the child’s
perception.
Dweck is cited by Bar-Tal in support of his claim that the verbal
appeals made by teachers can influence the attributions made by pupils.
In this study (Dweck, et al. 1978) the interactions that took place
between teacher and pupil were observed and these observations then
related to the attributions that pupils made. In short, Dweck’s claim was
that in cases where the teacher relatively frequently criticized a pupil for
non-academic aspects of their work (attendance, classroom behaviour,
neatness, grooming, and so on), the pupils were in turn less likely to see
ability as a prime cause for their failures. Dweck’s argument has a
100
compelling ring to it. Pupils who receive relatively little criticism for the
non-academic aspects of their work are less likely to be able to find
plausible reasons for their failures other than their lack of ability. What
made many people attend to Dweck’s study was the fact that she claimed
that the patterns of teacher-pupil interactions she had observed to be
related to attributions, were also related to the gender of the pupil. Girls
were held to be less likely to receive criticism for the non-academic
aspects of their work and therefore more likely to come to conclude that
their relative lack of ability was responsible for any lack of success they
experienced.
In spite of the compelling ring that Dweck’s argument has, there is
other evidence available that suggests that events may not always be as
she suggests. Eccles and Wigfield (1985) draw attention to a number of
studies that have failed to support aspects of the claim made by Dweck,
often referring to the sex-linked nature of the process but also
challenging the more basic link between teachers’ actions and pupils’
attributions. In their own review Eccles and Wigfield point to the
possibility that more general aspects of classroom life may have an effect
on the attributions made by pupils. Included in their considerations is the
work of Cooper (Cooper and Good 1983; Cooper 1983, 1985). Cooper’s
interest in attributions arose from a concern with the teacher expectancy
effect (Rogers 1982; Dusek 1985). Since the publication of Pygmalion in
the Classroom by Rosenthal and Jacobson (1968), considerable interest
has been shown by educationalists in the idea that teachers may, quite
unwittingly, influence pupil performance, perhaps unfairly, by virtue of
the expectations that the teachers hold for them. Cooper’s concern was to
attempt to identify some of the processes involved in the translation of
expectations into pupil performance. Essentially Cooper has claimed that
teachers are likely to want to exercise greater control over the
interactions that they have with pupils considered to be of relatively low
ability in order to be able to give greater assistance to those pupils.
However, an unintended consequence of this apparently sensible policy
decision is that the pupils concerned find it very difficult to initiate any
meaningful interactions with the teachers and to exercise any substantial
control over the work that they themselves do. This in turn can give rise
to the pupils’ coming to see themselves as lacking efficacy and therefore
coming to make attributions that match this. They are increasingly more
likely to take the view that their own efforts count for little and that it is
their lack of ability, about which they can do virtually nothing, that
causes their lack of progress. An important implication of Cooper’s
work, and one that requires further enquiry, is that a useful focus for the
researchers’ interests will be upon the conditions that are most likely to
lead to teachers believing that they ought to exercise such control over
certain of their pupils.
101
Colin Rogers
A developmental perspective
From the primary teacher’s perspective, the role of attributions in
motivation needs to be understood against a background of
developmental change. Nicholls has been an important researcher in
this respect (Nicholls 1984) and before going on to report details of
some of my own primary based work I wish briefly to examine some of
his claims.
It will be recalled that the causes of ability and effort are given a
prominent role by Weiner and his colleagues in their consideration of
the way in which attributions for success and failure might affect the
behaviour of individuals. Ability and effort are held to have different
effects from each other as they lie on different parts of the stability
dimension. Ability is seen to be a relatively stable cause of success
and failure while effort is relatively unstable. It is primarily for this
reason that attributions of failure to lack of ability are likely to have
more harmful effects than attributions to lack of effort. A lack of
ability will be seen to be an enduring characteristic of oneself and
will therefore continue to depress performance in the future. Effort
levels can be increased, however, so present low attainment levels
might also rise.
One of Nicholls’ major claims is that young children do not
necessarily share this view of the nature of ability. Nicholls has claimed
that the child’s conceptions of ability will progress though a number of
stages from an initial starting point at which the relationship between
cause and effect was not really understood at all. The child sees ability
and effort as being essentially similar in that both are liable to change. By
trying harder one can become cleverer. Effort is clearly seen as the major
cause of success and failure. At this stage, as some of my own research
shows, young children will assume that if everybody tries equally hard
then everybody will do equally well. It is only later, sometimes not until
the secondary years of schooling, that children come to acquire the adult
view of ability as a fixed and enduring entity that acts to set an upper limit
on what we can achieve. From this point onwards, attributions to ability
for failure begin to carry with them the implication that future failure has
to be expected. Prior to this point attributions to ability carry no
particular threat.
For this and other reasons (see the collection edited by Nicholls
1984 for a much fuller discussion, and Rogers, 1989) the primary
years are likely to be extremely important. It is during this period that
a child’s general motivational style will be developing. Ames (1986)
draws attention to a distinction between children who are ego- or task-
involved. Ego-involved children are more concerned with the
consequences of success and failure for their own self-esteem (and are
102
therefore likely to experience anxiety when confronted by difficulty),

while task-involved children concentrate on the completion of the task
itself. Dweck and her colleagues (Diener and Dweck 1978; Dweck
and Wortman 1982) have made a similar distinction between children
who are mastery oriented and those who experience helplessness.
When experiencing difficulty with a task the former group are likely to
assume that they are adopting an ineffective strategy, and will
therefore begin to look for ways to improve it, while the latter, helpless
children, assume that they lack the required ability and therefore begin
to give up.
It is likely that these patterns of response, while general in nature, will
not be completely uniform with respect to different types of tasks. Due to
the beliefs that children have about the nature of certain tasks, a task or
mastery-oriented approach may be more likely under some
circumstances than others. Work of the present author to be reported
below explores aspects of children’s understanding of the nature of task
requirements related to the curriculum area from which the task was
drawn and the way in which it was presented.
Children’s understanding of success and failure for different

types of work
Children from the top infant and second and fourth year junior classes
from two primary schools took part in the study. Each child was asked to
nominate one classmate that they considered to perform at a higher level
than themselves at each of three types of work and then one that
performed at a lower level. Having done this each child was then asked to
explain, in their own words, why these higher or lower performances
were obtained. The children’s responses were recorded and then later
categorized.
The three types of work that the children were asked to think about
were chosen so as to juxtapose curriculum area with style of
presentation. Two curriculum areas were chosen, mathematics and
reading. The two styles of presentation selected were tests based on
classwork, and schemes in which each child would make his or her way
through a series of graded activities at their own pace. Reading was
represented only in the schemes version while mathematics was
represented by both forms. Children therefore made their judgements in
respect to three sets of work, (1) tests—mathematics tests taken by the
whole class based on work done on a whole class basis, (2) mathematics
schemes, (3) reading schemes. These recorded responses were
categorized into a coding system that was based upon work reported by
Little (1985). The final coding system contained twenty-one categories
and these are summarized in the following Table.
103
Colin Rogers
Table 6.1 Brief definitions of categories used in the analysis of children’s

responses
Note All examples can be considered to be responses to the question ‘Why does
he/she do better than you in maths?’
104
Children within this age range make use of a wide range of different
types of explanations. It is not intended to make reference here to all of
the points that could be made about the findings summarized in the
Table, but rather to concentrate upon some of the more notable ones.
As children grow older they are more likely to make use of ability as
an explanation of better or worse performance. In particular, the oldest
children in the sample were relatively more likely to make use of
references to abilities that relate to a specific skill (‘He’s got a
mathematical brain’) as well as to more general ability (‘He’s just very
clever’). Both of the junior years were more likely to refer to effort as a
cause of relative success and failure, while the second year juniors were
particularly more likely to make reference to behaviour (in the sense of
the other displaying or not displaying teacher-sanctioned behaviour,
being well behaved being an explanation of success and poor behaviour
an explanation for failure). The sex of the pupil made very little
difference to the use of the different categories, which is generally in
keeping with other findings (Covington and Omelich 1979) which
suggest that sex differences in attributions may be relatively rare when
the data has been collected in natural classroom settings. Perhaps
surprisingly, the clearest tendency towards a difference between the
sexes was found with the youngest children. This might suggest that as
children have longer experience of the school system the effects of
gender on attributions begin to decline.
The effects of school experience are certainly implicated in the
findings related to the different curriculum areas and forms of work
organization. Reading seemed to be understood as making different
demands on the child than either version of mathematics. In particular,
Voluntary Time Spent on the Task was seen as being implicated in both
success and failure in reading. Children who did well at reading did so
because they spent more time at it, and this was the time which they could
have devoted to other activities (‘In the evening she reads instead of
playing out’). In addition to Voluntary Time Spent, a lack of a Specific
Ability was also frequently used to explain a lack of success in reading. It
is worth noting here that explanations for failure were generally not just
the opposite of the explanations given for success.
While reading success was clearly seen to be related to the amount of
practice that the children were judged to be getting, success in
mathematics was more likely to be explained by reference to ability,
provided the mathematics in question involved the use of tests. Children
were particularly likely to use references to General Ability, together
with references to Effort and Behaviour, to explain success in
mathematics tests. Failure at these tests was seen as being particularly
related to a lack of effort. When mathematics was approached via a
scheme, however, the pattern of attributions seems to be quite different.
105
Colin Rogers
There was no apparent pattern of use for General Ability here when the
children were attempting to explain relative success, while references to
Specific Ability were conspicuous by their absence. Failure at work
within a mathematics scheme was seen to be particularly associated with
behaviour, specifically the presence of aspects of behaviour that would
not be sanctioned by the teacher.
Conclusion
More extensive research along the lines of that described above needs to
be carried out before any firm conclusions can be drawn. However,
having given the usual caveat, it is worth briefly commentating on the
implications that might be said to exist The results of this research clearly
seem to indicate that the developing attributions of primary school
children are related to both the curriculum area that they are concerned
with and the way in which that work is presented. As well as learning
their mathematics and developing their skills as readers, children are also
developing their understanding as to what it is that one needs to be in
order to be able to perform well at these, and other, subjects. The way in
which these attitudes develop is likely to be crucial in determining the
pupil’s later motivation for that subject. If mathematics comes to be seen
as a subject that can only be succeeded at if one has a high level of
general ability, it is likely to pose a bigger threat to self-esteem in the later
years (see, for example, the work of Covington (1984) and his
colleagues) and therefore more likely to be avoided. The results of the
work reported above suggest that a child is more likely to develop such
an attitude if his or her early experiences with mathematics involve
frequent class-based tests than if they do not.
It would be clearly imprudent to advise teachers not to carry out such
tests in the light of these few results, but the strategy involved in this
research is one that could be sensibly taken up on a larger scale.
Attribution theory provides a framework within which it is possible to
analyse the motivational style of individual pupils. It is already clear that
there are possibilities for remedial work on this basis (Maehr and Kleiber
1987). It is now also apparent that attribution theory can provide a means
by which the effects of different classroom practices and the means by
which classroom material is presented to children can also be assessed.
Work has already been done on the effects of co-operative versus
competitive learning (Ames 1984). Such systematic analysis may
provide us with some further insights into the decline of motivation
throughout the school years, and hopefully some suggestions as to how it
might be arrested. More importantly, it also provides a means by which
teachers are able to begin to analyse some of the effects of their own
decisions regarding the ways in which classrooms are to function.
106
References
Ames, C. (1984) ‘Competitive, cooperative and individualistic goal structures: a
cognitive-motivational analysis’, in R.E.Ames and C.Ames (eds) Research on
Motivation in Education, vol. 1, Student Motivation, London: Academic
Press.
Ames, C. (1986) ‘Effective motivation: the contribution of the learning
environment’, in R.S.Feldman (1986) The Social Psychology of
Education: Current Research and Theory, Cambridge: Cambridge
University Press.
Ames C. (1987) ‘The enhancement of student motivation’, in M.L.Maehr and
D.A.Kleiber (eds) Advances in Motivation and Achievement, vol. 5,
Enhancing Motivation, Greenwich Conn.: JAI Press.
Ames, R.E. and Ames, C. (eds) (1984) Research on Motivation in Education, vol.
1, Student Motivation, London: Academic Press.
Ames, C. and Ames, R. (eds) (1985) Research on Motivation in Education, vol. 2,
The Classroom Milieu, London: Academic Press.
Andrews, G.R. and Debus, R.L. (1978) ‘Persistence and the causal perception of
failure: modifying cognitive attributions’. Journal of Educational Psychology
70: 154–66.
Atkinson, J. (1964) An Introduction to Motivation, Princeton, NJ: von
Nostrand.
Atkinson, J. and Raynor, J. (1978) Personality, Motivation and Achievement,
Washington, DC: Hemisphere.
Bar-Tal, D. (1982) ‘The effect of teachers’ behaviour on pupils’ attributions: a
review’, in C.Antaki and C.Brewin (eds) Attributions and Psychological
Change, London: Academic Press.
Cooper, H.M. (1983) ‘Teacher expectation effects’, in Bickman L. (ed.) Applied
Social Psychology Annual, vol. 4, London: Sage, pp. 247–75.
Cooper, H.M. (1985) ‘Models of teacher expectation communication’, in J.B.
Dusek, (ed.) Teacher Expectancies, London: Lawrence Erlbaum
Associates.
Cooper, H.M. and Good, T. (1983) Pygmalion Grows Up: Studies in the
Expectation Communication Process, New York: Longman.
Covington, M.V. (1984) ‘The motive for self-worth’, in R.E.Ames and C.Ames
(eds) Research on Motivation in Education, vol. 1, Student Motivation,
London: Academic Press.
Covington, M.V. and Omelich, C.L. (1979) ‘Are causal attributions
causal? A path analysis of the cognitive model of achievement
motivation’, Journal of Personality and Social Psychology 37: 1487–
1504.
Diener, C. and Dweck, C. (1978) ‘An analysis of learned helplessness:
continuous changes in performance, strategy and achievement cognitions
following failure’, Journal of Personality and Social Psychology 36:
451–62.
Dusek, J.B. (ed.) (1985) Teacher Expectancies, London: Lawrence Erlbaum
Associates.
Dweck, C. (1985) ‘Intrinsic motivation, perceived control and self-evaluation
maintenance: an achievement goal analysis’, in C.Ames and R.Ames (eds)
Research on Motivation in Education, vol. 2, The Classroom Milieu, London:
Academic Press.
Dweck, C.S., Davidson, W., Nelson, S. and Enna, B. (1978) ‘Sex differences in
learned-helplessness’; II, The contingencies of evaluative feedback in the
107
Colin Rogers
classroom’; III, ‘An experimental analysis’, Developmental Psychology 14:

268–76.
Dweck, C.S. and Wortman, C.B. (1982) ‘Learned helplessness, anxiety
and achievement motivation’, in H.W.Krohne and L.Laux (eds)
Achievement, Stress and Anxiety, London: Hemisphere Publishing
Corporation.
Eccles, L. and Wigfield, A. (1985) ‘Teacher expectations and student motivation’,
in J.B.Dusek (ed.) Teacher Expectancies, London: Lawrence Erlbaum
Associates.
Heider, F. (1958) The Psychology of Interpersonal Relations, New York:
Wiley.
Jones, E.E. and Davis, K.E. (1965) ‘From acts to dispositions: the attribution
process in person perception’, in L.Berkowitz (ed.) Advances in
Experimental Social Psychology, vol. 2, New York: Academic Press, pp.
219–66.
Kelley, H.H. (1983) ‘Perceived causal structures’, in J.Jaspars, F.Fincham and
M.Hewstone Attribution Theory and Research: Conceptual, Developmental
and Social Dimensions, London: Academic Press.
Little, A. (1985) ‘The child’s understanding of the causes of academic success
and failure: a case study of British schoolchildren’, British Journal of
Maehr, M.L. and Kleiber, D.A. (eds) (1987) Advances in Motivation and
Achievement, vol. 5, Enhancing Motivation, Greenwich, Conn.: JAI
Press.
Nicholls, J. (ed.) (1984) Advances in Motivation and Achievement, vol. 3, The
Development of Achievement Motivation, London: JAI Press.
Peterson, P.L., Wilkinson, L.C. and Halliman M.T. (1985) The Social Context of
Instruction: Group Organisation and Group Processes, New York: Academic
Press.
Rogers, C.G. (1982) A Social Psychology of Schooling, London: Routledge &
Kegan Paul.
Rogers, C.G. (1989) ‘Early admission. Early labelling’, British Journal of
Educational Psychology, Special Monograph Series, no. 4, Early Childhood
Education.
Rosenthal, R. and Jacobson, L. (1968) Pygmalion in the Classroom, New York:
Holt, Rinehart & Winston.
Stipek, D.J. (1984) ‘Young children’s performance expectations: logical analysis
or wishful thinking’, in J.Nicholls (ed.) Advances in Motivation and
Achievement, vol. 3, The Development of Achievement Motivation. London:
JAI Press.
Weiner, B. (1979) ‘A theory of motivation for some classroom experiences’,
Journal of Educational Psychology 71: 3–25.
Weiner, B. (1980) Human Motivation, New York: Holt, Rinehart & Winston.
Weiner, B. (1983) ‘Speculations regarding the role of affect in achievement-
change programmes guided by attributional principles’, in J.M.Levine and
M.C.Wang, (eds) Teacher and student perceptions: Implications for learning,
Hillsdale, NJ: Lawrence Erlbaum Associates.
Weiner, B. (1984) ‘Principles for a theory of student motivation and their
application within an attributional framework’, in R.E.Ames and C.Ames
(eds) Research on Motivation in Education, vol. 1, Student Motivation,
London: Academic Press.
Weiner, B. (1985) ‘An attributional theory of achievement motivation and
emotion’. Psychological Review 92: 548–73.
108
Weiner, B. (1986) An Attributional Theory of Motivation and Emotion, New York:

Springer-Verlag.
Weiner, B., Frieze, I., Kukla, A., Reed, L., Rest, S. and Rosenbaum, R.M. (1972)
‘Perceiving the causes of success and failure’, in E.E.Jones, D.E. Kanouse,
H.H.Kelley, R.E.Nisbett, S.Valins and B.Weiner (eds) Attribution: Perceiving
the Causes of Success and Failure, New York: General Learning Press.
109
Chapter seven
Social development of the child and

the promotion of autonomy in the
classroom
Peter Kutnick
‘Social development of the child and the promotion of autonomy in the
classroom’ combines two separate issues and opens various scenarios
through which teachers may more effectively structure the learning
context of their classrooms. The issues are not separate from one another.
Theories of social development describe the potential and experiences
for the behaviour and understanding displayed by children. Autonomy
describes various expectations of the ‘classroom performance’ of the
primary pupil and ways in which these expectations may be met.
This chapter explores the relational context of the classroom and the
effects (on relationships) of various teaching strategies. This chapter
intends to dispel as myth that autonomy is a simple or easily achievable
classroom product. If autonomy is described as self-directed
pupilhood, how is that self-direction achieved? It cannot arise from
purely individualized instruction, for studies show that this is most
likely to lead to pupil dependence on the teacher. Rather the self is seen
as a social product, developed through diverse relationships with adults
and peers. The actual relationships may vary in quantity and quality.
The focus here is on quality, and it is the closeness engendered in
relationships with teachers and classroom peers that encourage the
possibility of autonomy.
In reviewing a number of classroom and experimental studies, this
chapter identifies various uses of autonomy and the particular types of
relationships that they are based upon. All of the studies show that the
teacher is in a central role in structuring the experiences that are offered
to pupils. Structuring of experience includes the traditional academic
presentation of learning materials and the learning groups in which
children participate. Based on earlier chapters in this volume, present
discussions further explore classroom groups, social interaction and the
developing child. The chapter leads into specific consideration (in
110
Social development of the child and autonomy
chapters of Wright (Chapter 8) and Slavin (Chapter 13) of the roles and
relationships that can be structured between teacher and pupils and
provides a rationale for the integrated use of co-operative learning
techniques.
Introduction
This chapter considers the potential of the child from a social
developmental perspective and questions how this development may be
encouraged in the primary school classroom. Most readers will have
considered overt stages of cognitive development (in their various guises
of Piagetian, Brunerian and other schemes) which provides a broad
backdrop for this discussion. This chapter will review major features in
the child’s social development and definitions of autonomy will be
described. Observational and experimental research into the real world
of classrooms will then be drawn upon to question whether and how
classroom context promotes social development and autonomy. The
discussion concludes by reviewing a number of ways in which the
classroom can be structured for autonomy. The heart of the argument
considers the teacher’s role in a tightly-constrained educational system;
whether a child-centred approach is self-contradictory, and if
innovations such as co-operative learning should be integrated into the
classroom.
Autonomy: movement away from simple models

For many years it has been acknowledged that the role of primary
education should include (see DES 1985) cognitive, social and
emotional aspects. In providing such an education for children,
primary schools must encourage the psychological realities of
autonomy and competence coincidently with the development of
knowledge and understanding in their pupils. These realities may be
identified simply as promoting the child’s realization that they are each
separate and individual units (a relative autonomy from others), and
that they are able to master tool, object, word usage and
conceptualization of knowledge. In a simplistic sense, teachers and
psychologists would agree that autonomy and primary education are
intertwined and their duty towards children is to find (and use) a
concept of motivation which will allow the development of cognitive,
social and emotional skills while encouraging the child to undertake
these skills as an individual and self-guided unit.
There is always a ‘but’ to add to simple definition agreements; and
this is where the role of social psychology of primary schools becomes
more fully explained. Within this world of primary education and
111
Peter Kutnick
development of autonomy we acknowledge that the relationships

engendered in schools and the social context of the classroom define how
development will take place. On one hand, we can name a number of
theories and exhort that teachers naively adhere to them if autonomy is to
be successfully promoted. On the other hand teachers and classroom
researchers find autonomy development hindered by pupil dependence,
ineffective social relationships in groups and a central focus on practical
aspects of the core curriculum in primary classrooms. Even in the post-
Plowden years with the design of numerous curricula to promote
individual autonomy, the likelihood of autonomy actually taking place is
limited by the context and structure of intellectual and social relations in
the classroom (see Simon 1981).
Any consideration of autonomy in primary schools must ask what is
actually meant by autonomy and how this definition may be put into
practice. Theories and approaches to autonomy are based on various
ideas of the developing child. The simple expectation that children will
become autonomous by adhering to any (developmentally-oriented)
curriculum is quickly dispelled when we look at the various definitions
of autonomy in the literature; especially those generated in psychology
and sociology.
Psychological theories include consideration of ‘learning’ and
‘cognitive development’. Autonomy, as explained in ‘learning’ theory,
identifies children’s understanding as the result of a system of positive
or negative sanctions (or rewards) that are administered in co-
ordination with some event undertaken by the child. Both teachers and
peers are sources of rewards and serve as models for children.
Autonomy can be said to take place when the child has undergone a
succession of positive rewards (designed to progressively encourage
the individual) so that children may provide their own reward for
learning. Learning theorists find it difficult to explain how the
necessary movement from an ‘external’ reward (provided by the
teacher) to an autonomous state whereby the child can undertake and
regulate an ‘internal’ system for rewards can be encouraged. Principles
of reinforcement are evident in a number of classroom-based teaching
approaches (see Deci, Nezlek and Sheinman 1981), where teachers
reward information-oriented and self-controlled behaviours by pupils;
and there are successful results within the strict experimental
parameters of these studies. An alternative conceptualization of
autonomy is based upon an understanding of ‘cognitive development’
derived from theorists such as Piaget (see Piaget 1971). This approach
identifies a series of stages through which the child’s intellect may
progress (and may appear to assume children to be autonomous if the
stage of formal operations is achieved). The stage approach has been
used in many curricula and individualized teaching techniques
112
currently used in science, mathematics and other areas in primary

schools. This understanding of autonomy as intellectual development
has been under much criticism lately, a criticism which forces the
realization that cognitive development takes place in a social
relationship context, and learning and autonomy are embedded in the
experiences offered to the child (see Kutnick 1988). A cognitive
developmental view of autonomy would encourage the child into a
position of sufficient intellectual and social relational skills where the
motivation to learn and understand would come from within the
individual and would be unconstrained by lack of immediate
information and personnel around the child (the child is not purely
dependent on the teacher for support and knowledge). A rough
approximation of the progression towards autonomy identified by
psychologists is present in Table 7.1.
Table 7.1 Development of social relationships, with logical-mathematical, moral

and social perspective (developmental) parallels
Source Kutnick, 1988:53.
Children’s initial development moves socially and intellectually

away from dependence on people and objects. Development progresses
to include relations with other adults (teachers) and peers which facilitate
a more critical appreciation of events around the child, thus encouraging
autonomous as opposed to dependence learning.
Apart from the psychological theories, a range of social approaches to
autonomy and development is evident. Social approaches include
113
Peter Kutnick
functional behaviours and the description of interactional processes

leading to autonomy (or its inhibition). Functional approaches identify
an end-product (as fitting into society) and establish patterns of action
within the school to produce that end-product. On the surface, functional
approaches are closely aligned to learning theories and may disguise
serious flaws in promoting autonomy. In identifying an end-product,
functional approaches demand a conformity to social standards set by
school authority. These standards may not include ‘self-direction’ by the
pupil if it conflicts with teacher and school requirements. Phenom-
enological approaches observe actual happenings in the classroom; the
establishment of rules, rituals and negotiations that take place between
interactors in the classroom. Information is provided concerning abilities
of children, teachers and variations in the way individuals act
Interactional approaches include the understanding that: intellectual as
well as social developments take place in a social environment,
interactors bring particular qualities to the environment, and that
interactions and context can be so structured as to promote particular
variations in the development of autonomy.
Both psychological and social approaches present a range of
autonomy models; conformity to a set standard (by teacher or school
system) or an active involvement in the learning context. Any simple
assumption that unplanned participation in the classroom will lead to
autonomy should be quickly quashed. Current psychological theories
have shown that the child is more than an ‘empty jug’ or ‘tabula rasa’.
The development of competence leading to autonomy must take place in
a planned context (by the teacher) and should not be left to chance.
Teachers’ desire for effective learning may be combined with the
functional need to operate in an overcrowded classroom and an
individualized conceptualization of learning, although a developmental/
interactional view of autonomy occurring through processes of the
classroom is at odds with a functional approach. Autonomy is neither
simple nor easy to implement.
Social developments in childhood, a background for autonomy

Children are born into a number of social relationships and demonstrate
many competencies with objects and people that surround them. The
amount and type of experience that they are offered structures children’s
understanding of the world and creates the possibility for further
development. It should be accepted, as an understatement, that all early
development of the child takes place in or near the presence of other
people (usually parents or caregivers in development terminology). It is a
misconception to assume that a child’s development can take place in
isolation or that development is so purely ‘cognitive’ that it is not
114
affected by the presence of others and the quality of cultural experiences

offered at home or in society.
Social development is concerned with the stimulating and expanding
world around the child through social interactions. Initially, some
qualifications should be identified. It is only common sense to
acknowledge that differences may exist between the potential for
interaction and understanding, and the reality of that interaction and
understanding taking place (which is affected by the amount, type and
quality of experience offered to the child). Many social relationships
within which the child interacts are ‘close’ relationships of love and
affection; providing security to participate in the trial and error process of
development without having to ‘do it right’. Close relationships are, thus,
different from general social relationships of acquaintance or loose
friendship. What the child derives from close interactions is most basic to
both intellectual (cognitive) and social development—movement away
from an egocentrically dominated world (see Furth 1980). Realization
that others may have a different point of view from oneself and the ability
to adjust to this moves the child into a ‘social-centric’ world. Thus,
knowledge is seen as a social process (as observed and described by
Edwards and Mercer 1987) resulting from actions with others;
development is not just simple discovery by the child, it is a discourse
between the child and others. Others, with whom the child interacts and
develops close relationships, are usually archetypically termed as adult
and peer. Each of the archetypes has a particular quality of relationship
with the child; yet the child needs to develop quality relationships with
both types to provide a background for autonomy (see Youniss 1980).
Shantz (1982) states that the child’s main social responsibility in
development is to construct a workable social theory that will allow her/
him to participate in the actual social environment.
The world of adults

The child’s relationship with adults is obviously the first relationship to
develop and the most long-lasting. This relationship lays the background
for future ‘sociability’ and establishes the basis for acceptance of and
reaction to authority (facilitating relationships with teachers, social
conventions and early moral development). Classically, adults have been
seen as the vital and only link between the child and society, although
this view seriously belittles the potential relationship with peers
(according to Youniss 1980). In their position as adults, parents have
been shown to structure the child’s social, emotional and cognitive
experiences. Within this relationship the child is socialized into an
expected social conformity.
Probably the best example of the experience offered by parents is
‘attachment’. Attachment is ‘an affectional tie that one person forms
115
Peter Kutnick
between himself and another specific one—a tie that binds together in
space and endures over time’, according to Ainsworth et al. (1974). The
child’s realization of attachment usually occurs between 9 and 12
months of age. The child, who previously was quite settled in the
presence and care of many others, begins to show a distinct preference
for one or very few individuals. This tie to particular others has been
preceded by a quality relationship of care and stimulation between the
child and caregiver (usually parent or very near relative). While
occurring early in life, the existence of attachment will affect the
child’s later cognitive and social development. Ainsworth has found
that children with the most secure attachment relationships (measured
by the amount of acceptance, sensitivity and co-operation as opposed
to rejection, insensitivity and interference between child and caregiver)
are more likely to explore unfamiliar surroundings and objects than
children with poor attachment. Exploration and curiosity lead to
cognitive stimulation and development of understanding. Upon
achieving school age, children with good attachment are more
competent at exploring their new environments and more effective in
their interactions with teachers and peers (see Waters and Srouf 1983).
Affection in attachment also takes place with an adult who is bigger
than the child and who controls resources (food, comfort and
stimulation). Children’s understanding is parentally dominated by
regularities such as ‘knowing’ what is right and wrong, when to get up
and go to bed, and assertion of gender roles within the home.
Attachment is thus a relationship of hierarchical authority and love.
The regularities of behaviour leading to attachment co-occur with
authority and rule interactions. Piaget (1932a) describes this social
regularity as introducing a ‘morality of constraint’ to the child; the
basis for further experiences with rules and others which should,
eventually, lead to autonomy.
In their way, parents (or caregivers) provide the social experience
upon which children are able to form their understanding of rules,
conventions, morality, and ways to interact with others. Obviously
variations in the way that parents interact with their children will affect
future behaviours displayed by the child. Research by Baumrind
(1971) has shown that the authority style adopted by the parent will
affect the child’s ability to control their own behaviour and ability to
initiate and follow through activities in the nursery school (for
example, parents who set, explained and negotiated rules tended to
have children best adapted to starting school). And children, for their
part, will show variations in the way that they interact with the
authority situation. One type of interaction is known as dependence/
independence proneness— whereby children may demonstrate greater
or lesser need for teacher direction, approval and feedback. Children’s
116
affiliation to adults (as opposed to peers) may also be determined by

birth order, with first-born and only children having a stronger tie to
adults than later born children.
Parents or caregivers play a major role in the stimulation of thought
and understanding for the nursery-aged child according to Tizard and
Hughes (1984). Home-based relationships allow the child to be
interacted with as an individual and to participate in quality
conversations about sequences of household events, mathematical and
logic skills. Dissimilarly, children in nursery schools are often conversed
with (by teachers) in the presence of other children. Teacher participation
is often limited to initiating conversations and giving signs of approval or
disapproval before being drawn into other conversations. Parents
stimulate the child’s activity by providing intensive discourse with their
children concerning ‘embedded’ concepts (that are real and meaningful
rather than abstract).
Underlying the world of adults is the constant presentation of a
hierarchical relationship, whether in power-subordination, expert-
novice, knower-seeker. The relationship with adults is an obvious model
for the child’s entry into primary school, in which the teacher is focused
upon as a source of security and attachment while structuring the
classroom environment, rules and knowledge for the child. This
hierarchical relationship explains only part of the child’s social
development and only one step towards autonomy.
The world of peers

Relationships with peers have been noted for encouraging co-operation
and close relationships of intimacy and mutuality. Peers also have been
described as providing models to be copied, social reinforcement of
approved behaviour and educational tutors for one another (according to
Field 1981). Older-aged peers have been found to teach ‘socialized’
behaviour of manners, norms and dominance hierarchies to younger
children (see Hartup 1978). Studies of early infancy have found that
children will prefer to look at and interact with one another (as opposed
to adults) from as young as two months. If allowed opportunities to
consistently interact with a particular peer, evidence of friendship has
been found as early as 12 months (see Vandell and Mueller 1980). Peers
can help overcome fear of strangers. There have been cases in which
peers form attachments with one another in preference to adults (see
Freud and Dann 1951).
The world of peers presents two distinct features:
1. They engage in preparation for adult life. Children practise classic

patterns of dominance and subordination (leader and follower
games) and sex-role differences (around Wendy houses and play
117
Peter Kutnick
hospitals). They replicate adult behaviour in play; a non-

threatening interaction where children do not have to be absolutely
correct and will often switch roles between themselves. They
participate in these activities with some intensity but invariably
with a smile on their faces (see work on rough-and-tumble play by
Blurton-Jones 1976).
2. Peer activity is distinctive from adult behaviour. Children,
especially young children, are able to compare themselves with
one another and find similarity of size, strength and experience.
Youniss (1980) notes that sharing goods and pleasurable activities
amongst peers can lead to a sharing of private thoughts and
feelings, and development of a sense of mutual consideration. This
ability to be ‘the same as’ is the basis for Piaget’s (1932a) assertion
that only peers can develop the ‘morality of co-operation’; an
ability to consider other children’s points of view and come to joint
decisions. The morality of co-operation necessitates movement
beyond an egocentric world to one of ‘sociocentrism’ (rules can be
adapted and changed to maintain play). Children as young as 4 and
5 can resolve social conflicts and help one another if they are in a
stable social group according to Strayer (1980). Pullantz and
Gottman (1981) found that antisocial children can learn how to
become accepted by their peers by observing and adopting
appropriate strategies used by other peers. Recent research in the
UK has found that children with pre-school experience are more
socially and cognitively advanced than children without preschool
experience, as measured at the end of primary schooling (see
Osborne and Millbank 1985).
The potential for various relationships (with social and intellectual

outcomes) does not necessarily mean that enhanced social awareness and
cognitive development will come to fruition simply by being placed in
the presence of peers. Readers with experience of young children may
not find (the ideal of) co-operative activity occurring frequently.
Conflicts of possession and current friendship are more likely to occur in
the intensity of peer interactions. Conflicts can cause disruption in
relationships between peers and tend to spill over into classroom
disruption. Conflicts have two divergent results:
1. Antagonism between children may lead to constant changes of

friends, the possibility of fights and general disruption.
2. On the other hand, conflicting perspectives between peers lead to
argumentation or discourse, the resolve of which enhances social
and intellectual understanding (see Edwards and Mercer 1987).
118
Piaget (1932b) states:
If we watch the child when he is not with adults…we will see

him develop a social life with other children, and we will
notice that while he is developing socially he is also
intellectually acquiring the power of thought and criticism.
Recent cognitive experiments (see Bearison et al. 1986) have paired non-
conserving peers and assigned them a conservation task. The result was
much discussion leading to intellectual movement originally beyond the
capability of either of the peers.
Interaction between peers potentially lays the background for
cognitive development. But not all relationships between peers will
promote cognitive development or even friendship. There will be
variation between childrens’ ability to get on with one another. Peer
relations such as friendship may be inhibited by classic social
psychological constraints of proximity (availability of children to
interact with each other), similarity (such as sex, race, ability and age
distinctiveness), and quality of play objects (single child toys will
encourage competition while two child toys encourage co-operation).
The actual occurrence and role that peers can play will be sharply defined
by the society and structured interactions in which they are allowed to
participate (see Kutnick 1988, for further discussion).
Bringing together the social developmental worlds of adults and peers
two final points can be made. First, the reader is reminded that human
thought and understanding are based in social relations and
communication, and that knowledge is constructed through joint activity
and discourse (from Edwards and Mercer 1987). Adults and peers each
have unique and overlapping contributions to make to this development.
A second point demonstrates the unique contributions from adult and
peer, and provides a diagrammatic display of the development of
autonomy. Figure 7.1 displays the social authority characteristics and
interpersonal relations generated in relationships with adults and peers.
The figure is an adaptation of Piaget’s research concerning moral
development. It represents the contributions of constraint and co-
operation applied to relations with adults and peers, and how the
combined aspects of authority and relationships are essential bases for
the development of autonomy. In this diagram the importance of adults in
generating an understanding of ‘constraint’ and peers in ‘co-operation’ is
stressed as well as the balance between co-operation and constraint. The
balance is fundamental for autonomy, where the child has the initiative
and freedom to interact fully with adults and peers without being
dominated by either.
119
Figure 7.1 Reinterpretation of Piaget’s stages of moral development as a stair-like matrix
120
Peter Kutnick
Source Kutnick, 1988:85.

Social developmental potential: points concerning action and

structure
The social worlds provided by adults and peers describe the potential for
development and question what is necessary to bring that potential about.
The potential for social development is often at odds with actual
occurrence in the child’s social world. Previously it was found that peers
working together (co-operatively) are able to promote conservation skills
in advance of individual skill levels (see Perret-Clermont 1980; Bearison
et al. 1986; and others). It would be ideal if children’s groupings in
primary schools supported this development. Instead of cognitive
potential being enhanced, pupil groups often show children working
only for themselves or competing with one another for limited resources
or toys. While there may be potential for co-operation amongst children,
they do not usually take advantage of it (in western societies). Obviously
other societies whose life practices are dependent on co-operation will
use it with greater frequency (as the kibbutz in Israel, and less-developed,
and agrarian societies, according to Shapira and Madsen 1969 and
Whiting and Whiting, 1975).
A distinction between general social and close relationships is vital
for the consideration of co-operation, learning and autonomy. While
children generally need social interaction, their most meaningful
learning and development takes place in close relationships. Closeness is
associated with affectional ties, mutuality, sensitivity and security
between the child and specific others. This closeness is generally
characteristic of the attachment relationship with parents and transfers to
teachers in the early stages of primary school. Closeness amongst peers
is not generally found until the child is able to achieve mutuality and
intimacy with a ‘best friend’ (which may not take place until
adolescence, according to Youniss 1980) unless teachers make an active
effort to promote closeness by structuring co-operative learning actions
(see Slavin in this volume) or some form of trust exercises (see Kutnick
and Brees 1982). If the developmental route to autonomy found in Figure
7.2 is to be followed, children must be allowed to experience closeness
with peers as well as adults. This experience of closeness should take
place as early as possible in their school careers.
In promoting autonomy we must consider that the development of
thought and the actions in which children participate are intimately
related. Piaget has described this relationship as the ‘law of conscious
realization’. Children entering the primary school (pre-operatively) need
to involve themselves in activities and repeat this involvement over and
again before they come to realize that their knowledge and behaviour is
changing. There are many examples of the constant activity of water play
(leading to conservation of liquid) and social play (leading to shared
121
Peter Kutnick
understanding of rules and co-operation) that can be cited; the child must
be fully involved in these activities. Children who are more able to use
their reflexive (concrete) thought are able to come to an immediate
understanding of that activity—but they still need to participate in action
before coming to understanding. As children ‘learn’ from the actions in
which they participate, teachers much consider that the structuring of
these actions is an essential practice. It is primarily the experience of
particular activities and with whom/how they are undertaken that are the
bases for the child’s learning and development. Jerome Bruner (1986),
citing research by Vygotsky (1978), speaks of ‘scaffolding’ intellectual
and social experience for children by the teacher. Scaffolding may
appear a complex term, but it graphically represents the need to structure
activities in the classroom for types of participation (whether discourses
between child and teacher or between peers). On the social
developmental side, we find that children involved in enjoyable activities
tend to form friendships with their co-participants (promixity leading to
similarity). Any grouping established for children’s participation in
events should be considered critically. Children will become attached to
groups (especially classroom ability-based groups established by
teacher) in which they act and tend to disassociate themselves from other
groups who are perceived as ‘different’ from themselves (see Tajfel
1978). Action and structure are key elements in bringing about social and
cognitive development.
To what extent is social development catered for in the classroom?

Recommendations from the Plowden Report (1967) have been cited as
providing legitimacy for a more child-centred primary education with
emphasis on social development, individualized curricula and mixed
ability classroom groups. However, recent research on classroom
activity finds that child-centred, social development is rare. To explain
the rarity of the social developmental classroom (with implications for
autonomy) let us question to what extent social development has
occurred in the classroom and how it may be creatively integrated.
Reviewing social development and autonomy-inducing aspects of
educational research is a complex task. There are two distinct types of
study that can be cited, with little or no overlap between them. The
studies may be termed ‘naturalistic’ and ‘experimental’. These studies
attempt to describe the educational experiences of a number of
classrooms or set out to prove a particular point respectively.
Naturalistic
Naturalistic studies of the classroom provide a holistic picture of
interactions and educational outcomes that characterize the
122
behaviour of teachers and pupils. (These studies are mainly labelled

as ‘process— product’ in that they do not fully account for the
backgrounds of teacher and pupils or the broad context of the
school.) In focusing on classroom interactions and their outcomes,
some studies have considered whether child-centred or progressive
teaching methods promote better outcomes than traditional methods.
Other studies simply provide a general picture of the classroom and
explain certain outcomes by the classroom regime. In both types of
study classroom organization is a central concern, and better
organized classrooms correlate with higher outcome for the pupil.
Regarding social development, we can ask three questions: Does
child centredness make a difference? Do teacher actions actually
promote a progressive education? And do children actually display
behaviour that will promote their social and autonomous
development? Insights into these questions are provided by the
following:
1. Bennett (1976) published results purporting that traditional

classrooms produced better academic results than informal and
mixed-style classrooms (although the highest scoring classroom
happened to be the most progressive). Bennett’s results caused
much controversy, leading to one clear conclusion— ‘the central
factor emerging from this study is that a degree of teacher direction
is necessary, and that this direction needs to be carefully planned,
and that the learning experiences provided need to be clearly
sequenced and structured’ (Bennett 1976:162).
2. The ORACLE team (see Galton, et al. 1980) challenged Bennett’s
assertion that a formal teaching style produced better results than an
informal style. They observed teacher and pupil behaviour in junior
school classrooms. None of the teaching styles that they observed
could be labelled child centred. In the main, pupils were taught
traditional subjects by traditional (didactic control) methods. Pupils,
for their part, undertook their ‘learning’ as an individualized activity
—even though they were placed in groups; pupils did not
cooperative with peers.
3. A similar lack of social cohesion and co-operation was described in
infant classrooms by Bennett (Bennett et al. 1984). In these
classrooms teachers mainly adopted an individualized approach to
learning and relied on individualized 123 curricula to encourage
development. They found that the pupils responded to the curricula,
although when a problem arose children had to rely on the teacher
for support, direction, and remedial attention. Large queues build up
around the teachers’ desk. Teachers took on a functional and
controlling role to get through the number of questions and maintain
123
Peter Kutnick
order in the classroom. Guidance and diagnosis by the teacher

became subverted by pupil dependence and the need to organize the
classroom.
4. Interactions in the classroom do provide the structure for
relationships and what comes to be known by pupils. Kutnick (1983)
observed teacher and pupil behaviour in infant and junior schools
and questioned children about teachers and classrooms. Evidence of
the ‘law of conscious realization’ was a main finding; with infants
interacting with their teachers and slowly coming to describe
teachers by the learning and caring activities that they participated
in. In addition to expected learning activities of the classroom,
juniors showed increased interaction with peers. They were warned
by teachers that ‘learning’ was not taking place while they ‘chatted’
to peers. Correspondingly, juniors described their teachers as having
both learning and control responsibilities in the classroom.
5. A final study shows that children may not naturally display co-
operation in the classroom, but they are very aware of social
processes. Amongst many results cited, Pollard (1985) found that
pupils often worked in groups in class and developed group-based
friendships. The groups, whether teacher appointed or child
preferred, tended to be based on ability, with children quickly
discovering a classroom hierarchy of ability and choosing friend/
groups on a basis of perceived similarity to one’s own ability.
Where does naturalistic research lead? It informs us that many

primary school teachers tend to use traditional control methods,
focusing on children’s behaviour rather than stimulating cognitive
aspects of pupil development. Children learn from what is actually
done in the classroom. While they may be placed in groups in an
informal setting, they take little responsibility or control over their
own learning. In classroom learning, children tend to approach
problems as individuals; they do not make use of peers or co-
operation. They depend on the teacher to resolve intellectual and
social problems.
Experimental
Experimental studies ask that a class be taught by a specific method
(usually to enhance child autonomy), and compare performance to a
parallel class that has been taught by traditional methods. The most
frequently cited studies have been undertaken in the United States, and
assume that children who demonstrate independence in the classroom
will learn more (conceptually) and achieve better academic results. In
these studies independence is equated to autonomy and autonomy is
achieved through the (behaviourist) reinforcement of some learning
124
behaviours in preference to other behaviours. In studies undertaken in

elementary schools, results tend to confirm the development of
autonomy and neglect to account for academic achievement. Examples
of this research include:
1. Deci, et al. (1981) instructed experimental teachers to use more

autonomy-oriented behaviour (use of rewards to promote an
information-oriented classroom rather than a control orientation).
They found pupils increased on measures of personal responsibility
and self-control in these classrooms.
2. Ryan and Grolnick (1986) studied children’s self-perception of
control, and found those who felt responsible for their own
behaviour (as opposed to those who felt their behaviour) was
controlled by teachers) were more likely to have positive concepts of
self-worth and an ability to demonstrate cognitive competence than
other pupils.
3. Grolnick and Ryan (1987) withdrew children from their
classrooms, provided them with one of three types of teaching
sessions (traditional, structured/child-centred and unstructured),
and tested the children for rote learning, conceptual learning and
interest. Results showed more rote learning took place with
traditional methods. More interest, conceptual learning and
longevity of learning took place in the structured and unstructured
sessions.
These studies conclude that the greater control pupils feel they have over
their own learning (greater ‘autonomy’) the greater are their feelings of
self-worth and conceptual learning. The studies do not question the
extent to which conceptual learning should dominate the classroom nor
do they provide any but the simplest behavioural definitions of
autonomy.
As a result of the above studies, two divergent recommendations for a
socially-structured classroom can be made:
1. The traditional classroom may be the most effective situation for

promoting ‘learning’. Croll and Moses’ (1988) observations of
junior classrooms found that whole class instruction provided the
greatest amount of control for the teacher, and teachers were able to
use this for effective questioning. Pupils, correspondingly, spent
more ‘time on task’ (which correlates to classroom learning).
2. An alternative recommendation comes from Ingram and Worrell
(1987), who experimentally introduced negotiation and pupil self-
determination (an individualized form of autonomy) into a primary
classroom. They found that pupils were capable of sharing
125
Peter Kutnick
responsibility and providing effective intellectual opportunity/

challenge for learning without autonomy being reduced to teacher
dependence.
The divergent recommendations are each supported by different areas

of classroom and psychological literature: the traditional, well-ordered
classroom calls for a ‘functionally’ well-behaved pupil with a
conformist definition of autonomy; the negotiating classroom allows
for discussion, argumentation and cognitive conflict supported in a
cognitive definition of autonomy. One’s definition of learning outcome
and autonomy must be carefully considered before choosing classroom
structure.
Promoting social development and autonomy in the classroom
The traditional view

From the previous discussion we find that the well-ordered, traditional
classroom is able to support academic and creative learning, at least to
a similar extent of progressive classrooms. Traditional classrooms, for
the most part, are happy places where children listen to and work with
the teacher. Teacher control may be didactic, but this disguises a range
of authority types. Teacher authority may span authoritarian,
authoritative and permissive types (similar to parental authority types
found by Baumrind 1971), and children tend to respond most
positively to the authoritative style (where they can negotiate and
assume some responsibility). This authority style is mirrored in
children’s preference for teachers who are ‘firm, but fair’ and ‘serious,
but human’ (see (Nash 1973; Pollard 1985). Children appear to feel
secure and able to learn in the traditional but caring classroom. Pupil
preference for a particular style of teacher will be based on a range of
experiences that they have been offered in their schooling, and few
pupils will have actually experienced the progressive alternative. We
are unable to say what pupils think of progressive teachers or whether
this style is preferred. It should be noted that within the traditional
classroom there are a number of practices which may inhibit social
development. Inhibition of social and autonomous development may
be a result of individualization in learning (cutting the child off from
the cognitive enhancing potential of peer interaction), ineffective
groupwork (with dependence on the teacher for direction and support),
generation of classroom hierarchies and cliques (differentiating pupils
from one another), and the perception that the responsibility for
learning is solely controlled by the teacher. Dependence-based learning
has been ‘effective’ in the past and still works today; especially in areas
126
of rote learning. In these circumstances, teachers who care,

communicate well, establish organized and sequential practice, and set
good examples, will be most effective with their pupils.
Alternative views
Moving away from the traditional classroom, we note that the sharing
of planning and decision-making by teacher and pupil will increase
feelings of involvement by pupils. One view of classroom-based
autonomy calls for greater pupil involvement. Greater amounts of
negotiation provide grounds for consensus between teacher and pupil
(see Pollard and Tann 1987) and lead to a non-traditional
understanding of learning and behavioural obligations, shifting
responsibility from teacher to include the pupil. Research undertaken
by Ryan and colleagues has found that pupil responsibility leads to
enhanced feelings of self-control and self-worth, and pupil learning is
seen as intrinsically motivating. ‘Autonomy’ identified by Ryan and
colleagues is based upon individual encouragement by the teacher.
Positive results for this individualized presentation of autonomy have
been found in experimental settings but remain to be verified in the
classroom over time.
This review of social development also leads to a more dynamic
approach to autonomy. A dynamic approach requires that pupils
develop effective interaction with teachers and peers as necessary bases
for autonomy. With effective peer interaction, the child will have a
viable alternative to weigh against teacher dependence and control.
Effective interaction with peers has not generally been found in
primary classrooms (where pupils are most likely to co-act rather than
cooperate). A number of studies have shown that work with peers can
be effective in co-operative learning (see Slavin in this volume). Co-
operative learning is not a simple exercise and requires active planning/
structuring by the teacher and the development of a group reward
system. Results from co-operative learning studies in classrooms show
that cognitive and social development is enhanced and pupils are less
disruptive. Pupils working effectively with one another (and
developing a ‘close’ relationship) may help to ease teacher dependence
and add a co-operative dimension to their actions (see Figure 7.2),
providing firm bases for autonomy.
In both the traditional and alternative views of autonomy, teachers
play a central role. They must be aware of the social and developmental
potential of children as well as understanding how this potential can be
brought into the classroom context. Teachers must plan and structure the
action and activities of the classroom for pupil participation. Planning
and structuring is not synonymous with control. To provide for
127
Peter Kutnick
autonomy the teacher must focus on the individual or the group and
generate a viable alternative to teacher dependence. Without this
devolution of control, classrooms will remain as effective as they
traditionally have been—and dependence, differentiation, limited
cognitive enhancement and effective control will remain as problems of
schooling. The autonomous classroom (which incorporates social
development into its structure) is able to draw upon effective
relationships with teacher and peers to promote security amongst pupils,
more effective cognitive development (through discourse with peers) and
co-operation.
References
Ainsworth, M., Bell, S. and Stayton, D. (1974) ‘Infant-mother attachment
and social development: “socialization” as a product of reciprocal
responses to signals’, in M.P.Richards (ed.) Integration of a Child into a
Social world, London: Cambridge University Press.
Baumrind, D. (1971) ‘Current patterns of parental authority’,
Developmental Psychology Monographs 4: 1–103.
Bearison, D., Magzaman, S. and Filardo, E. (1986) ‘Socio-cognitive and
cognitive growth in young children’, Merrill-Palmer Quarterly 32: 51–72.
Bennett, S.N. (1976) Teaching Style and Pupil Progress, London: Open
Books.
Bennett, S.N., Desforges, C., Cockburn, A. and Wilkinson, B. (1984) The
Quality of Pupil Learning Experiences, London: Lawrence Erlbaum
Associates.
Blurton-Jones, N. (1976) ‘Rough-and-tumble play among nursery school
children’, in J.Bruner, K.Sylva and A.Jolly (eds) Play, London: Penguin.
Bruner, J. (1986) Actual Minds, Possible Worlds, London: Harvard
University Press.
Croll, P. and Moses, (1988) ‘Teaching methods and time on task in junior
schools’. Educational Research 30: 90–7.
Deci, E., Nezlek, J. and Sheinman, L. (1981) ‘Characteristics of the
rewarder and intrinsic motivation of the rewardee’, Journal of
Personality and Social Psychology, 40: 1–10.
Department of Education and Science (DES) (1985) The Curriculum from
5–16, Curriculum Matters no. 2, London: HMSO.
Edwards, D. and Mercer, N. (1987) Common Knowledge: The Development
of Understanding in the Classroom, London: Methuen.
Field, T. (1981) ‘Early peer relations’, in P.S.Strain (ed.) The Utilization of
Classroom Peers as Behavioral Change Agents, New York: Plenum
Press.
Freud, A. and Dann, S. (1951) ‘An experiment in group upbringing’.
Psycho-analytic Study of the Child 6: 127–68.
Furth, H. (1980) The World of Grown-ups, New York: Elsevier Press.
Galton, M., Simon, B. and Croll, P. (1980) Inside the Primary Classroom,
Grolnick, W.S. and Ryan, R.M. (1987) ‘Autonomy support in education:
Creating the facilitating environment’, in N.Hastings and J.Schwieso
(eds) New Directions in Educational Psychology, Lewes: Falmer Press.
128
Hartup, W. (1978) ‘Children and their friends’, in H.McGurk (ed.) Issues in

Childhood Social Development, London: Methuen.
Ingram, J. and Worrell, N. (1987) ‘The negotiating classroom: child self-
determination in British primary schools’. Early Child Development and
Care 28: 401–15.
Kutnick, P. (1983) Relating to Learning, London: Allen & Unwin.
Kutnick, P. (1988) Relationships in the Primary School Classroom, London:
Chapman.
Kutnick, P. and Brees, P. (1982) ‘The development of cooperation:
explorations in cognitive and moral competence and social authority’,
British Journal of Educational Psychology 52: 361–5.
Nash, R. (1973) Classrooms Observed, London: Routledge & Kegan Paul
Osborne, A.F. and Millbank, J.E. (1985) ‘Long-term effects of pre-school
education’, paper given at Preschool Education Conference at the
University of London, Institute of Education.
Perret-Clermont, A.N. (1980) Social Interaction and Cognitive
Development in Children, London: Academic Press.
Piaget, J. (1932a) The Moral Judgement of the Child, New York: Free Press.
Piaget, J. (1932b) as cited in A.N.Perret-Clermont (1980) Social Interaction
and Cognitive Development in Children, London: Academic Press.
Piaget, J. (1971) The Science of Education and the Psychology of the Child,
London: Longman.
Plowden Report (1967) Children and Their Primary Schools, 2 vols. Report
of the Central Advisory Council for Education in England, London:
HMSO.
Rinehart & Winston.
Pollard, A. and Tann, S. (1987) Reflective Teaching in the Primary School,
London: Cassell.
Pullantz, M. and Gottman, J.M. (1981) ‘Social skills and group acceptance’,
in S.R.Asher and J.M.Gottman (eds) The Development of Children’s
Friendships, London: Cambridge University Press.
Ryan, R.M. and Grolnick, W.S. (1986) ‘Origins and pawns in the classroom:
Self-report and projective assessment of individual differences in
children’s perceptions’, Journal of Personality and Social Psychology
50: 550–8.
Shantz, C. (1982) ‘Children’s understanding of social rules and the social
context’, in F.C.Serafica (ed.) Social-cognitive Development in Context,
New York: Guildford Press.
Shapira, A. and Madsen, W. (1969) ‘Cooperative and competitive
behaviour of kibbutz and urban children in Israel’, Child Development
40: 609–17.
Simon, B. (1981) ‘The primary school revolution, myth or reality?’, in
B.Simon and J.Willcocks (eds) Research and Practice in the Primary
School, London: Routledge & Kegan Paul.
Strayer, F. (1980) ‘Childhood ecology and the study of pre-school social
relations’, in H.C.Foot, A.J.Chapman and J.R.Smith (eds) Friendship
and Social Relations in Children, Chichester: Wiley.
Tajfel, H. (1978) Differentiation between Social Groups: Studies in the
Social Psychology of Intergroup Relations, London: Academic Press.
Tizard, B. and Hughes, M. (1984) Young Children Learning, London:
Fontana.
129
Peter Kutnick
Vandell, D.L. and Mueller, E.C. (1980) ‘Peer play and friendship during the
first two years’, in H.C.Foot, A.J.Chapman and J.R.Smith (eds)
Friendship and Social Relations in Childhood, Chichester: Wiley.
Vygotsky, L. (1978) Mind in Society: The Development of Higher
Psychological Processes, London: Harvard University Press.
Waters, E. and Srouf, F. (1983) ‘Social competence as a developmental
construct’, Developmental Review 3: 79–97.
Whiting, B. and Whiting, J.M.W. (1975) Children of Six Cultures,
Youniss, J. (1980) Parents and Peers in Social Development, Chicago:
University of Chicago Press.
130
Chapter eight
Towards an adequate conception of

early moral development
Derek Wright
Initiating the focused section, Derek Wright combines a theoretical
background with commonsense realities to stimulate consideration of the
teacher’s role in encouraging moral development in pupils. The chapter
is written distinctively from a teacher’s point of view. It stresses the
importance and quality of relationships that can be engendered in the
classroom. The chapter opens an important issue in consideration of
moral development
Wright has worked with teachers and others in the field of moral
development over many years. His concerns about relationships should
be familiar with many teachers. His discussion moves away from
criticism of the overpresence of didactic teaching methods in the
primary classroom. He introduces the possibility that teachers may
engage in two distinctly different types of relationship with pupils, that
of constraint and that of collaboration. He points out that teachers have
been successful in the past for engendering the development of
morality in their charges. What he does not say is that the teacher’s role
is even more important in today’s society as school and teacher may
provide the only consistent experience of moral values that the child
encounters; especially if we accept that past incarnations of moral
experience (of the church, extended and nuclear family) are declining
in importance.
This chapter also links the quality of relationship between teacher and
child with particular conceptions of moral development. The didactic
teacher does promote a certain moral understanding of rules and
children’s obligation of acceptance. Rule knowledge and the generation
of moral habits are important for young children but they also have
implications for the child’s ‘personhood’. The collaborative or co-
operative relationship that can develop between teacher and pupil
provides a ‘richer’ sense of morality, and is seen at the root of discourse
and fairness. Thus Wright encourages teachers to consider the relation-
131
Derek Wright
ships in which they interact with pupils and effects on the child’s moral
and personal development. These working and classroom relationships
are particularly important when considering children with special needs
(Galloway, Chapter 12) and the need for the teacher to structure actions
for themselves and the pupils (further amplified in the context of play in
Chapter 9 by Smith).
Introduction
The focus of this chapter is upon how we, as teachers and parents, can
best conceive of the nature of moral development in children. No attempt
will be made to review the relevant psychological theories and research
findings, though the writer’s knowledge of this work lies behind the
discussion, and two theoretical perspectives, those of Freud and Piaget,
will be briefly outlined. Not only would it be impossible to do justice to
this theory and research in a short chapter, but the interested reader can
find such accounts elsewhere (for example, Wright 1971; Sapp 1986).
More important, the present aim is to explore our conception of moral
development, by which I mean our overall idea of how it occurs, or, more
accurately, to contrast two basic ideas which have strong echoes in
commonsense. Particular psychological theories are always partial,
limited in scope, and focused upon certain aspects of the process, and
empirical findings are always piecemeal and designed to test particular
predictions.
As adult moral agents we already have some conception of how
moral development occurs, and this conception shapes the way we treat
children within our living relationships with them. Moreover, we
cannot be so far off target, because overwhelmingly children grow up
to be moral agents themselves in greater or lesser degree—as we have.
However, it is my experience, both of myself and of the many teachers
I have worked with, that our conceptions of moral development are
largely implicit and unexamined, and that therefore we may be less
effective in promoting moral growth in the young than we might be.
The hope behind the chapter is that it will edge the reader towards
greater clarity and understanding of the process, and therefore of how
to promote it.
In my own efforts towards clarity I shall be drawing on certain ideas
developed in detail by Freud, and in much less detail by Piaget, in
order to contrast the two fundamental conceptions of how moral
agency evolves in children. I have chosen these two theorists, partly
because of their enormous influence, but mainly because their ideas
resonate strongly with commonsense and with our ordinary
experience of the moral life. It is this aspect of their work I want to
stress.
132
Early moral development
The teacher’s experience

It must surely be the experience of all teachers that they find themselves
at different times relating to children in one of two different ways or
modes. For convenience I will call them the control mode and the
equality mode. Though they may alternate frequently in the course of a
single lesson, it is difficult to see how they might occur at the same time.
Because of their extensive responsibilities and the complex network
of their accountabilities, teachers have no choice but to spend a lot of
time controlling, directing, and even dominating their pupils. Without
such control and direction the ordinary business of school learning could
not take place. Children themselves need, and indeed want, to be
controlled by teachers. It creates for them a sense of security and safety
and relieves them of the stress of responsibilities they are not ready for.
Of course, the ways in which such control is exercised can vary widely,
from the crudely aggressive at one end to the subtle and sensitive at the
other.
Teachers are not just concerned with the professional business of
teaching and learning, however, As moral agents themselves they
inevitably feel that children should not steal, lie, bully or cheat, and
should be helpful to each other, work hard, be truthful, and so on. In the
control mode of relating, the moral sense of the teacher finds its
expression in prescribing how children should behave, in reprimanding
and perhaps punishing for wrong behaviour, and in some way rewarding
the right behaviour. In other words the teacher asserts control in order to
ensure as much as possible that children behave in ways she or he judges
they ought to behave. The children experience morality as, so to speak,
imposed from outside through the teacher’s assertion of authority and
power though, of course, they can experience it as more or less
aggressively and compellingly imposed.
It is when teachers feel able to relax out of the controlling mode that
the equality mode of relating becomes possible. It is this aspect of their
work with children which many teachers find the most richly rewarding.
It includes moments of shared sadness or laughter, occasions of
democratic decision-making, times when there is real and authentic
exchange of thoughts between teacher and child in which each listens
seriously to the other, and that kind of co-operative activity in which the
teacher is equally a participant. In short it means those times in the
classroom when teacher and child experience each other as ordinary
persons on a more or less equal footing.
In regard to moral issues, the equality mode is present when there is
genuine reciprocity between teacher and child. Children feel as free as
teachers to give expression to their own moral voice in relation to their
own behaviour and that of the teacher, and in relation to the rules and
133
Derek Wright
organization of the school. Teachers themselves are able to acknowledge

their own unfairnesses and mistakes, and are seen to try to put things
right. Moral issues arising in the life of the school or outside it are open to
free discussion between teacher and pupils with the goal of reaching
agreement.
Piaget’s terms (Piaget 1932) for these two forms of relatedness that
have been briefly outlined are unilateral respect and constraint on the one
hand, and mutual respect, co-operation and reciprocity on the other.
From the point of view of the teacher, each carries an underlying and
implicit conception of how moral development is fostered in the child,
and each has implications for how we understand the moral life. Since it
may be assumed that all teachers experience both kinds of relatedness
with children in some degree, both conceptions are present in the
teachers themselves. In what follows the attempt will be made to bring
both conceptions out into the light of day so that we can ask to what
extent they are compatible (Piaget is emphatic that they are not), and
which of the two takes us to the heart of the matter (Piaget argues that it is
the second). It must be stressed, though, that we shall be concerned with
conceptions of relationship, moral development and morality, and not
with descriptions of the actual relationships teachers have with their
children, for these will always be unique to particular teachers and the
children they teach. The goal is to help teachers, in their efforts to nurture
moral growth in children, to be a little clearer about what they are doing
and why.
The control conception of moral development

The main features of this conception are very familiar. It is what many
people seem to mean by ‘teaching children to be moral’. Children are
initiated into, and sustained within, the moral world when adult
caretakers exercise their natural power and authority in a firm and kindly
way by telling them how they ought and ought not to behave and why,
rewarding them with approval and affection when they comply, and
expressing disapproval and perhaps punishing when they do not
conform. The underlying message from the adult is ‘I know how you
ought to behave and I am going to do my best to ensure that you do
behave in that way’. From the children’s point of view, morality is
experienced as external to themselves, as deriving from adults, and as
having the authority of adults behind it. Their function is to respect and
obey what adults tell them to do.
It is probably inevitable, and possibly necessary, that to some extent
children should experience morality in this way. However, the more the
relationship between children and adults is characterized by unilateral
respect, that is, the more strongly and rigidly adults assert their
134
controlling power in morally constraining and directing children, the

more their obedience will be conditioned by fear and anxiety, and the
more they will experience frustration and conflict. When such moral
training is taken to extremes, we may well agree with Piaget that ‘one
realizes most keenly how immoral it can be to believe too much in
morality, and how much more precious is a little humanity than all the
rules in the world’ (Piaget 1932:189).
Implicit in this idea of how morality develops in children is a
conception of morality as itself derived from some external authority,
either that of society, or of God, or of a church or religious culture, or of
a sacred scripture. The assumption is that morality consists of a number
of rules of conduct which have behind them an authority external to all
individual human beings and to which the individual is obliged to
submit. The function of adults in relation to children is to transmit these
rules and convey the sense of authority behind them by asserting their
own authority over the children. We might label this understanding of
morality as vertical or hierarchical since it implies that morality is
imposed on the individual from ‘on high’.
It is also implicit in this concept of moral development that without
the controlling authority of adults children would not become moral
beings. Moral agency is not intrinsic to being human but a set of attitudes
and beliefs which have to be somehow engrafted on to the child. The
problem in this view has been expressed by Piaget when he asserts that
‘the adult’s command, in spite of the nimbus which surrounds it, will
always remain external, “stuck on” as it were, to a mind whose structure
is of a different order’ (Piaget 1932:163).
However, it does appear to be the case, as Freud has argued, that,
given the child’s dependence on adults, and given that the unilateral
respect relationship is softened to some extent by reciprocal affection,
the child will ‘identify’ with adults and ‘internalize’ their prescriptions,
and thus form a conscience or super-ego. In Freud’s account the
aggression provoked in the child by the adult’s controlling power is
turned inwards upon the self, and makes possible the familiar experience
of self-blame and guilt, and also the tendency to blame others who break
the rules. As Freud put it, ‘the external restrictions are introjected, so that
the super-ego takes the place of the parental function, and thence-
forward observes, guides and threatens the ego in just the same way as
the parents acted to the child before’ (Freud 1933:85).
Whether or not we go along with Freud’s psychodynamic
explanation (and there is an alternative, behaviouristic account of
essentially the same process), it seems clear that within relationships of
unilateral respect which are moderated by a degree of reciprocal caring
and attachment, the rules imposed by adults can become more or less
permanently lodged within the child’s mind. Young children can
135
Derek Wright
become very scrupulous about observing the rules of adults in their

absence, can obviously experience guilt when they do not, and may be
quite strict in their insistence that younger children conform to them.
Indeed, the moral voice of authoritarian adults can become so
embedded in the child’s mind that it can become more demanding than
the adults themselves were. Some children can be too conscientious for
their own good.
This very brief account of the control conception of moral
development will, I hope, be sufficient, for it is likely to be very familiar
to the reader. It appears to be the dominating conception in the minds of
many people and probably deserves the label ‘traditional’. Most of us
have some experience of its effectiveness, for it relates directly to our
experience of conflict between an inner moral demand and wishes,
desires and wants. At the very least, children learn how adults expect
them to behave, and up to a point useful habits or regularities of
behaviour may be developed. It can be argued that the moral domain is
far too important and complex for children to be left to work it out for
themselves.
However, there are important questions to be asked. Is moral
development simply the internalizing of adult rules and prescriptions? Is
the concept of morality implicit in this approach an adequate one?
Though it appears to work with many children, it manifestly does not
with others. Certainly teachers are all too familiar with the fact that it has
little effect with some. Before turning to an exploration of what I have
called the equality approach, it is worth looking more critically at the
conception I have outlined.
A critique of the control conception

First of all, since I have drawn on Freud’s ideas to support the claim that the
control conception can be effective in certain ways, it is worth exploring
his thinking a little further in order to bring out some of the limitations he
saw in the development of a conscience or super-ego morality. For it is
fundamental to his account that the process of identification with, and
introjection of, the powerful parental moral voice involves for the child the
repression, or denial, in greater or lesser degree, of ‘immoral’ desires and
motives. Thus, on this theory, the more the child internalizes the ‘high
moral standards’ prescribed by adults so that they become a semi-
autonomous agency within the child’s personality, the more the child will
live with a state of inner conflict and tension. Add to this that the more or
less compulsively internalized rules may be far from rationally justifiable,
and the scene is set for personal problems in living later on.
It is worth mentioning another phenomenon to which Freud has
drawn attention, which is perhaps especially relevant for teachers,
136
namely, what he calls ‘identification with the super-ego’. The state of

being identified with the super-ego is one in which the individual’s
conscience is, for the moment, turned outwards onto others who have
clearly done something of which that conscience strongly disapproves.
This creates a momentary resolution of the inner state of tension and
conflict, for ego and super-ego are one, and the condemnatory and
repressive aggression usually turned inwards as an autonomous braking
mechanism for wrongdoing is turned outwards on someone else. When
this happens it is, for the moment, virtually impossible for the individual
to feel that ‘there but for the grace of God go I’, or, perhaps more
important, ‘there despite the grace of God went I’. Compassion is swept
away by ‘righteous’ anger, which can easily issue in unjustly punitive
action when the individual has power over another.
Though Freud clearly thought that some degree of super-ego
development was essential if the child is to become a civilized adult, his
therapeutic experience made him acutely aware of how destructive to
full and creative living a severe conscience could be. Indeed, it could be
said that in his account of the therapeutic experience he has given us the
germ of an alternative theory of moral development which finds fuller
description in Piaget. For our purposes the relevant aspect of the
therapeutic process is the progressive transformation of the super-ego,
or compulsive parental morality, into an ego morality; a morality
consciously chosen by the self through the reflective experience of
understanding the self in relation to others. As Freud put it in his last,
unfinished publication, ‘The analytic physician and the patient’s
weakened ego, basing themselves on the real external world, have to
band together into a party against the enemies, the instinctual demands
of the Id and the conscientious demands of the super-ego’ (Freud,
1986: 406). The striking feature of the psychotherapeutic process is
that the therapist goes to the very opposite extreme from the moralizing
adult of the child’s experience, and is as completely nonjudgemental,
in both the moral and nonmoral sense, as it is possible to be, whatever
the patient may report having done or not done. In the presence of
someone who refuses to pass moral judgements, the patient is able to
find her or his own moral voice, and separate from the introjected
moral voice of parents.
Freud thus presents us with two conceptions of moral development, a
super-ego conception which he developed in great detail, and an ego
conception which remains embryonic in his writing. The former has the
implication of a relatively split-off and compulsive conscience which
represents internally the moral voice of adult authorities, and the latter a
more or less rational and integrated moral sense which represents the
individual’s own, unique moral voice. They correspond more or less to
the control and equality conceptions as described here.
137
Derek Wright
Clearly, within relationships of unilateral respect, adult moral

prescriptions may ring true to the child’s own undeveloped and
precarious moral voice. When this happens, the child’s own moral voice
is strengthened and supported, and given clear verbal structure. But the
two can easily be in conflict Bonhoeffer (1965:367) gives a brief
illustration. A teacher asks a boy in front of the class whether his father
came home drunk last night. The boy denies it, though in fact his father
did come home drunk. As Bonhoeffer points out, the boy has lied, but in
doing so is bearing witness to a deeper truth, namely that it is none of the
teacher’s business. The boy’s own intuitive moral sense in conflict with
the teacher’s demand leads him to break a moral ‘rule’. An important
question to ask of any school is the extent to which the assertion of
control by teachers compromises the nascent moral integrity of pupils.
This simple example raises a much more fundamental limitation of
the control conception. It is that morality itself is altogether too rich and
complex a phenomenon to be reduced to a set of rules and prescriptions
that can be communicated by adults. Central moral notions like
truthfulness and fairness cannot be spelled out in any final set of rules to
be obeyed. The way such notions influence behaviour in relation to
others must always be a matter of individual intuitive and reflective
judgement
As a final point, consider the following imaginary dialogue between a
teacher and Bertrand Russell, aged seven. It is intended to illustrate a
simple point of logic, or moral mathematics as some philosophers have
called it. I have chosen Russell aged seven because it is likely he would
have responded this way, and if his moral mathematics is valid for him, it
must also be valid for other children aged seven, whether or not they can
realize or understand it.
Teacher: Russell, you should always tell the truth.

Russell: Sir, you must know as well as I do, that/cannot
possibly conclude from what you have just said that/
should always tell the truth unless I assert a second
premiss, namely that I should do what you tell me to
do. I cannot deduce that second premiss from anything
you might say. If I am to assert it, I must have a reason.
There is only one good reason I could have. That is that
I too think I should always tell the truth.
This logical point is given a more psycho-logical slant by Piaget when he

says:
For conduct to be characterised as moral there must be

something more than outward agreement between its content
138
and that of commonly accepted rules: it is also requisite that

the mind should tend towards morality as an autonomous
good and should itself be capable of appreciating the value
of the rules that are proposed to it.
(Piaget 1932:409)
Let us for a moment imagine the exchange continuing. The teacher

would seem to have two options. The first is to enter into a rational
dialogue with Russell as between two equals, sharing reasons and
reflections in order to reach agreement. The second is to switch into
controlling power assertion and say, in effect, ‘I know what is right and
you will do as I say’. When adults adopt the second response to
children, there are two possible outcomes. The children may submit
with an inner sense of protest, or they may suppress their own moral
voice, at least for the time being, and, in Freud’s phrase, identify with
the aggressor.
The equality conception of moral development

This conception can be easily summarized. Its goal is the nurturing and
strengthening of the child’s own moral voice, or her or his ‘ego
morality’. Behind this is the assumption that having a moral voice is as
intrinsic and emergent a feature of being human as consciousness or
thinking or loving. There are two conditions which encourage its
development. The first is as much experience of relationships of mutual
respect, co-operation, and reciprocity, founded on reciprocal sympathy,
with both peers and adults, as possible. The second is the stimulation to
consciously realize and generalize the morality implicit in such
relationships. The child’s understanding of morality will then not be
dependent upon adult authority, but will have about it a self-evidence
authenticated by the child’s own experience. The function of the adult is
to respect and support the child’s own moral realizations.
There are at least three basic assumptions about morality and the
moral life which underlie this conception. As with all assumptions they
are to be questioned and explored, though that cannot be done here.
The first is that though persons are factually unequal in an indefinite
number of ways, as moral agents (originators of judgements of what
should and should not be done) they are on an equal footing, and it is
morality itself which asserts this. Persons, as persons, are to be valued
equally, and moral agency is central to being a person. Second,
morality is between persons in the sense that it represents the ideal way
in which relationships between persons should be regulated. Third, the
irreducible heart of morality is to be found in a limited number of core
moral ideas, not in a set of rules of behaviour. Rules are frequently in
139
Derek Wright
conflict and it is usually possible to think of situations in which they

ought to be broken.
It is for the reader to explore what these core moral ideas might be.
But there are surely certain central ones. These are: respect for persons
as persons; fairness and justice; truthfulness; and that keeping of
promises and contracts which is essential to all community life and co-
operative activity. These ideas are rich and complex. To illustrate, in
numerous workshops with teachers and pupils in which they have been
asked to unpack what they understand by respecting another person as
a person, the writer has always found the following emerging: listening
to others with that kind of attention which accords the same status to
their thought as to one’s own, though of course not necessarily
agreeing; not intruding upon their body space, personal possessions
and private life without their agreement; being sensitively supportive of
their self esteem; and so on.
In discussing relationships of mutual respect it must again be stressed
that I am speaking of a conception of how relationships might be, rather
than giving a description of how they ever are. Piaget is sceptical whether
they ever exist in pure form: ‘It can even be maintained that mutual
respect and cooperation are never completely realized. They are not only
limiting terms, but ideals of equilibrium’ (Piaget 1932:90).
This relates to another feature of the equality conception of
morality, namely, that failure is intrinsic to it. If morality is understood
in the impoverished sense as a set of relatively simple rules imposed by
external or internal authority, then it is at least possible that an
individual might obey them without fault. When morality is understood
in the richer sense, as the embodiment of core moral ideas like fairness
and truthfulness in human relationships, then it becomes the idea
towards which we struggle rather than a state that we achieve. It is
sometimes said that parents and teachers should ‘set an example’ of
moral rectitude for the young. If adults do try to ‘set an example’, then
inevitably they are creating a false image of themselves which in time
children will come to realize. In the equality conception of moral
development, the important example that adults need to be is that of
people who cope creatively and constructively with their own moral
failure in relating to others.
However, in so far as children experience such relationships with their
peers, and are drawn into and sustained within such relationships with
adults, to that extent the optimum conditions for the flowering of their
personalities and self-confidence exist. It is for them, says Piaget, ‘a
necessary condition of autonomy under its double aspect, intellectual
and moral’, and ‘from the moral point of view, it replaces the norms of
authority by that norm immanent in action and in consciousness
themselves, the norm of reciprocity and sympathy’ (Piaget 1932:103).
140
The critical point here is that core moral ideas of the kind mentioned
are, in Piaget’s term, ‘constitutive’ of relationships of mutual respect and
co-operation. That is to say, such relationships cannot be sustained,
unlike relationships of constraint and unilateral respect, unless such ideas
are at least to some extent implicitly shaping them. For children, living
with such relationships is living the moral life whether they know it or
not. Each party to such relationship exercises a degree of moral
autonomy, and therefore each is answerable to the other for the moral
judgements they originate and express. The inevitable controlling
influence that adults must at times assert over children needs to be
understood by both child and adult; it is not the imposition of rules by
those who know what is right, but a necessary consequence of the
contractual and co-operative relationship between them which results
from the considerable factual inequalities between them, and their
reciprocal and complementary responsibilities.
It is worth looking briefly at how fairness and justice issues might be
dealt with within mutual respect relationships between teachers and
children. It may well be that responsibility for a particular decision rests
with the teacher, say, in regard to assessment of work or a sanction for
misbehaviour, and teachers want to be fair. The equality conception of
morality asserts that we can only have confidence that a decision is fair
when all parties to it, in this case the children, also see it as fair. Within
the mutual respect relatedness, if the children see it as unfair, they would
be supported in saying so. Not only would they be listened to, but the
issue would be explored together in detail, and if necessary and possible,
the decision modified in order that they see it as fair. The point is that
within such relationships both parties actively participate in making
fairness. Children’s understanding of fairness and justice, and their
confidence in it, can only really come about through their experience of
trying to make it.
This brings us to the other major aspect of the equality conception of
moral development, namely the stimulation of their conscious realization
of their own intuitive moral sense, and its progressive structuring in
conscious thought. This happens through discussion with others, in
which the child tries to articulate her or his own thoughts, listens to others
and learns that they think differently, and tries through reflection to
rethink and reach some kind of agreement with others. The issues can be
immediate and living moral issues within relationships, or more
artificially constructed tasks initiated by the teacher. The teacher’s role is
through sensitive questioning and the putting of considerations that have
not been mentioned in order to extend the children’s thinking. The
process of consciously realizing and making rational our own moral
institutions is a lifelong task. But it is an essential task if we are to be able
to give meaning and structure to our own moral voice. If the reader is
141
Derek Wright
interested, Kohlberg has given us a preliminary account of how the

child’s moral reasoning develops and a full account of his work can be
found in Kohlberg 1981 and 1984.
Experienced teachers who have read thus far are likely to be
impatiently aware as a reality of their daily life that some children are
disruptive, antisocial, violent and hostile to peers and teachers alike.
It is this experience which is most likely to trigger the control
conception, and indeed be held to justify it. Since we are human, this
may at times be inevitable, and in the short term be expedient. But it
is also likely to be part of the teacher’s experience that in the long
term the more or less aggressive assertion of the teacher’s
disciplinary control is ineffective in producing a real change in such
children. The reason is, of course, that for such children the
conditions are not propitious for the internalization of the teacher’s
prescriptions, though they may learn as a matter of prudence to
modify their behaviour.
However difficult and slow it may be, the equality conception invites
us to set up situations in which we can offer our own respect for such
children as persons because they are persons. This means listening with
sympathy, and sensitively questioning them in order for them to give full
voice to their own confused feelings and thoughts—probably the first
time an adult has done that. And if the teacher listens carefully, she or he
is likely to hear as a kind of base line to what is being said, a muddled and
inarticulate sense of having been treated unfairly all along the line.
Indeed, much antisocial behaviour can be construed as at least partly an
inarticulate protest against being treated unfairly, though of course it is
due to much more than that. The equality conception would suggest that
the teacher try to get such children to articulate what they see as unfair in
their school experience, explore with them what would make things more
fair, and do something positive to bring that about. This small step would
at least create the beginnings of a relationship of mutual respect that
might subsequently be built upon. But I have no illusions about the
difficulty in doing this, especially under the stressful conditions in which
teachers work, and especially when the childs’ antisocial stance has
hardened. The equality conception would claim, however, that
something of the kind must happen if that stance is not to become
permanent.
Concluding remarks
The emphasis of this chapter has been upon teachers rather than
children. The reason is that the experience children have within school
depends so much upon how teachers conceive their task. The aim has
been to help teachers a little towards clarifying their own conceptions
142
of moral development, and therefore how they relate to children in that

respect.
Two contrasting conceptions of moral development have been
outlined. Neither will be new, since both, it is hoped, have a clear
resonance with commonsense and teachers’ experience with children.
The bias of the writer is clear enough. It is not claimed that the control
conception is wholly invalid. It has its own echo in our experience and is
related to a certain conception of the nature of morality itself. But it is my
view that the equality conception takes us close to the heart of the process
of moral development, and implies a more adequate conception of
morality. There is, of course, a great deal more to be said. If this chapter
stimulates the reader to pursue these matters further for herself or
himself, it will have achieved its task.
References
Bonhoeffer, D. (1965) Ethics; London: Macmillan paperback edition.
Freud, S. (1933) New Introductory Lectures, London: Hogarth Press.
Freud, S. (1986) ‘An outline of psychoanalysis’, in vol. 15, Historical and
Expository Works on Psychoanalysis, London: Penguin Books.
Kohlberg, L. (1981) Essays on Moral Development, vol. I, New York: Harper and
Row.
Kohlberg, L. (1984) Essays on Moral Development, vol. II, New York: Harper
&Row.
Piaget, J. (1932) The Moral Judgment of the Child, London: Routledge & Kegan
Paul.
Sapp, G.L. (ed.) (1986) Handbook of Moral Development, Birmingham,
Alabama: Religious Education Press.
Wright, D. (1971) The Psychology of Moral Development, London: Penguin
Books.
143
Chapter nine
The role of play in the nursery and

primary school curriculum
Peter K.Smith
The role of play is accorded a central place in the activities of children
generally, and especially those attending pre-school and primary
schools. Peter Smith provides an extensive overview and analysis of
definitions, processes and effects of play on early schooling. In the
overview he also stresses the importance of relationships between
children and their teachers, and that a considered and structured
approach is necessary for children to receive the range of benefits made
possible through play.
This chapter organizes criteria and categories of play. More usefully,
the diagrammatic presentation of teachers’ and childrens’ roles in
structuring play provides an insightful means to quantify benefits which
may occur in various schemes of play. We are reminded that play is not a
simple or singular occurrence. Effects of play are gradually built up by
the child but are dependent upon the actions in which the child
participates and whether the interpersonal focus of play is on the teacher
or peers. Results from and means of studying play are thoroughly
reviewed. We are made aware that constructive aspects of play are related
to intellectual growth and development. Teachers can tutor children to
use various types of play. Children invariably enjoy themselves when
playing. And, there are advantages in mixing various types of play for
children.
Pertaining centrally to the themes of this volume, we are made
aware that social relationships are necessary for a number of types of
play to take place, and that there are various social benefits of play due
to the type and quality of relationships that children gain with adults
and peers. Thus, while it is important that teachers structure play (either
as initiators or respondents), they must also be aware that outcomes of
these activities may lead to short-term cognitive gains, conformity,
social relational skills and longer-term intellectual development. It is
interesting that the types of benefit derived from play relate to adult and
peer relations respectively. The chapter coincides with themes of the
144
The role of play in the curriculum
type and quality of relationships introduced in earlier chapters, and the

role and structure brought to play by the teacher. Themes from play will
be taken up in the next chapter on friendship (Chapter 10) and seen
later in gender roles (Chapter 11) that may be generated in the
classroom.
Introduction
The aim of this chapter is to examine the role of play in the education of
the young child. Some of the relevant research is from the UK; some is
from the USA where schooling commences at 6 years. I have
incorporated material on older pre-school children, as well as primary
school age children, in this review.
What is play? Types of play seen in young children

It is not really satisfactory to attempt to give a definition, one definitive
statement, of what is or is not ‘play’. For too many behaviours, it is
contentious whether they constitute play, or not. We might perhaps all
agree that children ‘pretending’ to act out roles such as ‘mother’ and
‘baby’ are playing; but what about a child painting a picture, or making a
copy of a model with wooden bricks? We might not agree so well about
whether these behaviours are ‘play’.
Two main ways have been used to cope with this problem. One is to
use ‘play criteria’. The other is to use ‘play categories’.
Some psychologists have suggested that we use certain signs, or
criteria to decide whether behaviour is playful or not. The more such
criteria are present, the more likely it is that we would decide ‘this is
play’. The criteria usually suggested (Krasnor and Pepler 1980; Rubin, et
al. 1983) are that play:
(i) is intrinsically motivated—done for its own sake and not brought
about by basic bodily needs, or by external rules or social demands;
(ii) shows attention to means rather than ends—the child is more
interested in the performance of the behaviour itself than in the
results or outcome of the behaviour;
(iii) is characterized by pretence—is not serious but has an ‘as if’ or
make-believe quality;
(iv) has positive affect—is pleasurable and enjoyable to the child;
(v) shows flexibility—some variation in form or context.
Some other suggested criteria are that play is distinct from exploratory
behaviour, distinct from organized games, and is characterized by active
engagement.
145
Peter K.Smith
In one empirical study (Smith and Vollstedt 1985) it was found that
the criteria (ii), (iii), (iv) and (v) above were all used by observers making
judgements about whether behaviour in children in a nursery school was
‘play’, or not. Criterion (i), intrinsic motivation, was not used. However,
this study was made in a context where a lot of free play was allowed.
Certainly, activities initiated and strongly structured by teachers would
not be considered play, by criterion (i), and would not be so considered
by most observers.
A somewhat different approach is to categorize different types of play;
each category should be defined, or at least illustrated by clear examples of
‘paradigm cases’. Thus Piaget (1951) distinguished between ‘practice
play’, ‘symbolic play’ and ‘games with rules’. Practice play included the
sensorimotor and exploratory play of the young infant—especially 6
months to 2 years; symbolic play, the pretend, fantasy and socio- dramatic
play of the child, from about 2 or 3 to 6 years; while games with roles
characterized the activities of children from 6 or 7 years onwards.
Much of the play of the primary school child will be symbolic in
Piaget’s scheme. Children pretend that some action or object has some
meaning other than it’s usual, real-life meaning; for example if a child
rotates his arms, goes ‘poop-poop’, and gives out pieces of paper, he is
pretending to turn a steering-wheel, sound a horn and give out bus
tickets. When—as in this example—these actions are sufficiently well
integrated, then we can say that the child is in ‘role-play’, in this case,
pretending to be a bus driver. If two or more children are engaged in role
play together, this is ‘socio-dramatic play’. Empirical research suggests
that such forms of play increase with age then decline, through the 3 to 7
year age period (Cole and LaVoie 1985; Peisach and Hardeman 1986).
Smilansky (1968) changed Piaget’s scheme by including also a
category of ‘constructive play’, in which objects are manipulated to
construe or create something. This accords with the fact that many
teachers regard such activities as play. It departs from the analysis of
Piaget, who stated that ‘constructive games…occupy…a position
halfway between play and intelligent work, or between play and
imitation’. The goal-directed nature of constructive activities meant that
Piaget saw it a more ‘accommodative’ (the child adapting its behaviour
to fit reality), whereas symbolic play was more ‘assimilative’ (adapting
reality to fit the child’s own wishes).
Certain kinds of play do not fit well into the schemes of either Piaget
or Smilansky. In particular, physical activity play (running, climbing,
sliding, swinging and other gross muscular play) and rough-and-tumble
play (playful fighting, wrestling and chasing) are neither constructive nor
necessarily symbolic. Even though these forms of play are very
characteristic of the behaviour of children during break-time at school,
they have been rather neglected by psychologists and educationists.
146
Generally, it is constructive play and symbolic play which have had most
interest for discussions of primary school curricula.
Play in relation to the aims of the primary school curriculum

Educators in the kindergarten and primary school period have pointed to
a number of aims in early childhood education (for example, Taylor,
Exon and Holley 1972). These aims include:
(a) Social—to help each child make warm, stable relationships with
other children and adults, encourage responsibility and
consideration for others, and to help build self-confidence,
independence, and self-control, so that he or she has every chance of
leading a full and happy life.
(b) Intellectual—to encourage each child to use his or her intellectual
powers to the full, by fostering the use of language, stimulating
natural curiosity, and developing the ability to learn and form
concepts.
(c) Home/school—to help parents to a fuller understanding of the needs
of young children and of making the child’s educational life
smoother and happier, and easing the transition of the child to formal
schooling.
(d) Aesthetic/creative—to provide opportunities for the child to
experiment with a variety of materials in the fields of art and music,
to encourage creativity and expressiveness, and a growing
awareness and appreciation of beauty.
(e) Physical—to help the child develop motor and manipulatory co-
ordination and skills, and to meet physical needs through the
provision of fresh air, space to play and sleep, good food, training in
personal hygiene, and regular medical inspection.
How does the encouragement of play rate, in relation to these aims? It

can be argued to be relevant to most if not all of them, depending in part
on the forms of play considered:
(a) Much play is social. Socio-dramatic play and rough-and-tumble play

necessarily involve co-ordination of activities with one or more play
partners. Such forms of play can form a primary mode of social
interaction in this age range. This is less true of constructive play,
which can, but need not be, social. Most forms of play occur naturally
between like-aged children, but play can also foster child-adult
relationships if the adult engages in a play activity with the child.
(b) Many theorists claim that play has intellectual benefits. Socio-
dramatic play may foster language and role-taking skills, while
147
Peter K.Smith
constructive play may encourage cognitive development and

concept formation. Such aspects of cognitive development may
overlap with, though are not identical to, school-based criteria of
academic achievement. Rough-and-tumble play is not usually
claimed to have intellectual benefits.
(c) Play experience is probably least relevant to the aim of home/school
transition. If later schooling is to be more formal or structured in
nature, then play experience may not provide direct matching for the
sort of demands to be made on the child at the later time. On the
other hand, play opportunities can be a useful point of contact
between school staff and parents.
(d) Because play is often defined as internally motivated, and flexible,
many theorists believe that it is the optimal way of enhancing
creativity and imagination. Children have the freedom to try out new
ideas in play and can express themselves in their own way,
especially in socio-dramatic and fantasy play.
(e) Much play is physically active. Constructive play may practise fine
motor skills, while gross physical play and rough- and-tumble play
can provide whole-body exercise and motor co-ordination.
Thus, a case can be made that playful activities can further many or most
of the likely aims of early child education. However, many other
activities, such as more structured or didactic activities, organized
games, physical exercises, story-telling, and real-life tasks such as
preparing food, and tidying the classroom, can also further such aims.
Therefore, any benefits of play must be considered against any benefits
of non-playful activities. This issue is further complicated, not only by
the different kinds of play, but also by the different ways in which play
can be used in the early school curriculum.
Ways in which constructive play and fantasy/socio- dramatic play

are (or are not) used in the education of the young child
Depending on the theoretical perspective of staff, and the needs they

perceive in the children, constructive and dramatic play may or may not
have an important role in the curriculum. (I will defer consideration of
physical activity and rough-and-tumble play until a later section.) One
way of conceptualizing this is to use a four-fold classification, as shown
in Figure 9.1:
Teacher initiating/child responding

The teacher structures activities according to perceived needs of the
child, for example, arranging organized activities often in a group
148
format; the child primarily follows the instructions of the teacher, or

imitates others. This very ‘formal’ or ‘structured’ curriculum would have
little if any place for spontaneous play. Rather, adults impart or instruct
children in certain skills.
Teacher initiating/child initiating

The teacher structures activities, but takes account of the kinds of needs
or preferences which the child expresses. Thus, some forms of playful
activity, as well a non-play activities, may be used as the basis for the
teacher to encourage the development of skills such as language, co-
operation, concept development, and the planning of sequenced
activities towards a goal. Both constructive and dramatic play might be
structured in this way by adult intervention. The skill of the teacher lies in
diverting the play to challenging ends without sacrificing too much of the
child’s spontaneity and enjoyment of play.
Teacher responding/child initiating

The teacher does not structure activities, but does respond to the child’s
own needs and activities. Thus, the teacher will encourage the play of the
child by providing suitable play materials, and will often take the
opportunity to comment on the play, make suggestions, and so on.
However, the initiative is left clearly with the child. Here, great value is
placed on play and the spontaneous nature of playful activity.
Teacher responding/child responding

The teacher does not structure activities and only responds to the child in
respect of physical needs, such as avoiding danger, toilet training, meal
times, and conforming to institutional demands. The adult has a custodial
role rather than an educative one. Much of the child’s activity may be
play, but the encouragement or enhancement of play is not seen as of
great importance.
As an example of these four kinds of approaches, consider the aim of
fostering language development:
(i) In an A-type curriculum, children might be led in singing in a group;

or a teacher might engage in language drill exercises with children
individually or in small groups.
(ii) In a B-type curriculum, a teacher might ask a child to talk about
something which happened to them recently, or about what they
may have just made in play or construction; or to make up a song
or story; or might take part in a pretend play episode and try to
extend the child’s language while directing or structuring the play.
149
Peter K.Smith
(iii) In a C-type curriculum, a teacher could see language skills as being

fostered by spontaneous play; enhancing such play would be a
primary means of enhancing language skills, although the teacher
might also comment on or encourage a child to talk about their play.
(iv) In a D-type curriculum, no effort would be made to foster language
development beyond allowing the normal process of social
communication between children themselves, in a safe and healthy
environment.
Figure 9.1 A four-fold scheme for conceptualizing teacher attitudes to play.
The influence of beliefs and theories about play

Historically, play was not seen as educationally valuable in Western
Europe when nursery and infant schools began to be introduced in the
eighteenth and nineteenth centuries. In the main, children were seen as
in need of instruction and, in the case of religious doctrines,
redemption from sinful behaviour. The beginnings of an emphasis on
the value of the child’s own spontaneous growth, the image of the child
was more noticeably ‘good’ and hence of spontaneous play as more
valuable, stems from the Czech writer Comenius, the Swiss writer
150
Rousseau, and early social reformers and educators such as Owen

(UK), Pestalozzi (Switzerland), Froebel (Germany) and Montessori
(Italy). Such writers and educators generally saw the value of active
teacher involvement, while wishing also to respect the child’s own
natural growth.
They would thus tend to be in quadrant B in Figure 9.1, though with
considerable variation in the kinds of activities recommended.
Montessori, for example, favoured constructive play activities using
specifically designed materials, but did not encourage fantasy or socio-
dramatic play.
However, a more positive attitudes to the value of spontaneous play
and dramatic play came about in much of Western Europe in the
twentieth century, especially in the period 1930 to 1970. This play
‘ethos’ has regarded spontaneous play as an important if not essential
component of the child’s social and intellectual development, and of
creative and personal growth. It is epitomized in the phrase ‘play indeed
is the child’s work, and the means whereby he or she grows and
develops’ (Isaacs 1929:9.) It is re-echoed in, for example, the Plowden
Report:
we now know that play—in the sense of ‘messing about’

either with material objects or with other children, and of
creating fantasies—is vital to children’s learning and
therefore vital in school. Adults who criticize teachers for
allowing children to play are unaware that play is the
principal means of learning in early childhood.
(Plowden Report 1967:193).
In part, this play ethos seems to have come about from theoretical
perspectives such as evolutionary biology (suggesting the value of
play throughout mammalian species) and psychoanalysis (suggesting
the role of play in expressing emotions and working through personal
conflicts). In part, it seems to have been influenced by broader socio-
economic changes (such as smaller families, reduced child mortality,
the separation of work from home life, and the growth of the toy
industry) which led to greater concentration on children as in a
separate world with separate needs from adults. For whatever
reasons, the play ethos has been a powerful force in the curriculum
for nursery and primary school. Play has been seen as worthwhile in
its own right, and argued to have great cognitive or social benefits. A
number of theorists, such as Singer, Bruner and Smilansky have
argued along these lines, and the work of other theorists such as
Piaget have to some extent been reinterpreted in this way. Such views
would tend to favour curricula in quadrant C in Figure 9.1. The
151
Peter K.Smith
emphasis is more on respecting the child’s spontaneous play, the

initiative the child makes, than on the teacher structuring or directing
this activity.
Little firm evidence was available to support this very positive view of
spontaneous play through the period in which these views became
prominent. The evidence from recent decades, to be discussed next, is
mixed. Views about the importance of play are divided amongst
educators and child psychologists. The play ‘ethos’ remains influential,
but there has been some shift back to more structured curricula activities
(quadrants B and A in Figure 9.1). The Schools Council Project in
Structuring Play in the Early Years at School (Manning and Sharp 1977)
reflects a belief that play, while motivationally valuable, can be made
more so by some degree of adult involvement, encouragement and
structuring. The work provides some valuable ideas for helping children,
though the effectiveness of the programme has not been evaluated
empirically.
Empirical evidence concerning the role of play in development

and its use in early child education
Does it matter whether we incorporate time for play in the early school
curriculum, and how we do so? The answers will vary depending on the
aims of school provision or education. However some empirical studies
may help in making such decisions. Evidence about the role of play in
development, primarily constructive and dramatic forms of play, will be
considered first; followed by a review of studies comparing school and
pre-school curricula, which do, or do not, incorporate much free play, or
structured play. Some of these studies are correlational in nature; others
are more experimental.
In the correlational studies, observations and measurements are made
on a number of children. It is possible to see then whether those children
who naturally play the most, or in more complex ways, are also more
advanced in skills such as language or intelligence. These studies are
usually more based on real-life situations, but permit multiple
interpretations of the findings. For example, if children who play more
are found to also have more expressive language, this could be because
(i) play helps language development, or (ii) language ability facilitates
play, or (iii) some other factor, such as high intelligence, or secure home
background, facilitates both language development and play.
In experimental studies, some children are selected to have a
different kind of experience (for example, enriched play, or stimulation
of play) compared to other, similar children who do not have this.
Measurements are made of skills such as language or intelligence.
These studies usually allow more definite conclusions, since if the
152
children with more play experience score higher this should be due to
the different experiences, such as the enriched play. However, these
studies may be more artificial in their design. How natural is the play
produced by the interventions, for example? Both kinds of study—
correlational and experimental—have their own advantages and
drawbacks.
Correlational studies of the role of play in development

Quite a large number of investigations have compared children who
naturally engage in a lot of play behaviour, with those who engage in
rather little. Do such children differ in other ways?
(i) In one study of English 3 and 4-years-old (Davie, et al. 1984)

records were made of the behaviour of 165 children in the home
environment. About 12 per cent of their waking time was spent in
some form of fantasy play, though only about 3 per cent was role-
play (the remaining 9 per cent was representational object play,
that is, pushing a toy car along making motor noises). The
investigators also assessed the intellectual level of each child using
an intelligence test. There were no significant correlations (either
positive or negative) between the time spent in any type of fantasy
play, and intelligence test score. This might suggest that such
forms of play are neither beneficial nor harmful to intellectual
development at that age.
(ii) A study in the USA of children aged 4 to 7 years (Peisach and
Hardeman 1986) found the children who engaged in more
imaginative play (as assessed by observation, and interview) scored
less highly on measures of cognitive ability (multiplication of
classes, class inclusion, spatial viewpoint of others), especially at 5
years, but more highly on measures of social perspective taking,
especially at 5 and 6 years.
(iii) In another study of children in the USA aged 3 to 5 years (Connolly
and Doyle 1984), the amount and complexity of social fantasy in
children was correlated with intelligence scores, and also with
measures of social competence. There was again a zero correlation
between fantasy play, and intelligence, but a positive correlation
with the amount of social activity of the child, and a measure of role-
taking skill. In a further study of 2- to 6-year-olds, however, (Cole
and LaVoie 1985), the positive correlation of fantasy, and socio-
dramatic play to role-taking skill was not confirmed.
(iv) Another study of 4-year-olds in the USA (Johnson et al. 1982)
distinguished between socio-dramatic and constructive play. Again,
dramatic play showed no correlation with intelligence scores, but the
amount of time spent in constructive play correlated positively with
153
Peter K.Smith
intelligence, and negatively with the amount of time spent

unoccupied.
Summary:
The results of these and similar studies are not conclusive. The value of a
correlation can be due to some third factor (such as gender, personality)
and need not reflect an actual causal relationship between play and, for
example, intelligence or role-making skills. In so far as we do take this
evidence as important, it tends to suggest that the amount of time spent in
play is neither strongly beneficial nor strongly harmful to other aspects
of development. Possibly, constructive play seems more useful for
intellectual skills than dramatic play.
Experimental studies of the role of play in development
(i) Some investigators have tried to show the benefits of play

experience using small-scale experimental studies. In such studies,
carried out with children aged 3 to 6 years, children are taken
individually to a small room. They are first given either a ‘play’
condition— allowed free play with objects—a ‘training’ condition
(for example, being asked to compare the objects, by an adult) —or
a ‘control’ condition (for example, colouring in a drawing). Then,
the child is given some form of task or problem to solve. For
example, the child might be asked to think of unusual uses for the
objects they have just had; or to use the objects to construct a tool to
retrieve a distant object. The typical design of such studies is shown
in Figure 9.2 (obviously, the details have varied from one study to
another). At first these studies seemed to show that children who
had free play with objects would do better on the problem tasks.
However, subsequent studies suggest that this may not be so. Some
of the earlier results seem to have been due to ‘experimenter effects’
—that is, the experimenter expected the children in the ‘play’
conditions to do better, and, probably unknowingly, gave them some
extra help (perhaps encouraging them more by head nods or smiles
as they made the right steps towards solving the problem). When the
experimental studies are carried out with proper precautions, rather
few differences between the ‘play’, ‘training’ and ‘control’
conditions seem to be found (Simon and Smith 1985). Probably this
is because of the artificial nature of these studies, and their short
duration (usually about 10 to 20 minutes). Although these studies
now seem to fail to show any benefits of play experience, the sorts of
play experience may be so different from those that children have in
school or kindergarten, that the results are not very relevant to those
situations (Smith 1988).
154
Figure 9.2 Typical design of experimental studies of the effects of play experience
on problem solving
(ii) Another series of studies have been on a larger scale. These are the
so-called ‘play tutoring’ studies, which attempt to increase the
amount of fantasy or socio-dramatic play in classrooms. These
studies usually last two or three months, and approach in scope the
experimental studies of curricula, to be discussed in the next
section.
The idea of ‘play tutoring’ was introduced by Smilansky
(1968). She argued that socio-dramatic play was very important
for the development of social, cognitive and language skills in
young children. She thought there was value in both fantasy
(make-believe transformations) and in role-playing. She therefore
gave a high priority to increasing the amount and complexity of
socio-dramatic play in children who seldom engaged in such play
155
Peter K.Smith
in schools and kindergartens. Smilansky tried several methods of

encouraging such play, and found that the most effective method
was for the adults or teachers to initiate role-play with small
groups of children and help them to sustain it and develop it for a
period of time.
Four variants of this method were used by Smilansky and other
researchers:
Modelling.
The teacher participates in the play, joining in, and by acting out a
role demonstrates how it can be performed effectively. For example,
in acting out a ‘doctor’ role, the teacher could pretend a doll is a
baby and pretend a pencil is a thermometer. This demonstrates both
fantasy transformations, and some real parts of the doctor’s role, to
the children. The teacher could help develop an extended sequence
of play, for example, asking a child to go and get some more
‘medicine’ for her.
Verbal Guidance.
The teacher does not join in but makes comments and suggestions to
help the children develop the roles they are in, for example, reminding
a child playing ‘mother’ to wash the baby before bedtime.
Thematic-fantasy training.
Children are helped to act out familiar story-dramas, such as the
‘Three Billy-Goats Gruff’. Since the plot is usually known in advance,
this is a more structured procedure than the previous two.
Imaginative play training.

The teacher trains children in skills which focus on make-believe
activities. For example, they are trained to use finger puppets, or
practising using facial expressions to represent different emotions. Or,
children might sit under a large sheet of grey paper and pretend to be
outside on a rainy day. This kind of training focuses on fantasy and
does not usually involve role-playing.
Smilansky and most other researchers have found that such forms of
adult encouragement and training do increase the amount and
complexity of fantasy and socio-dramatic play in young children,
especially if they do not show much of this play initially.
Furthermore, some increase is maintained even if the play training is
stopped. But how successful is such play training? Do the children
benefit from it? Many researchers have used a design like that
shown in Figure 9.3. After pre- assessment, some children
experience fantasy play training, others a control condition in which
156
an adult is also present. This is followed by a post-assessment. These

studies suggested that there are strong benefits to fantasy and socio-
dramatic play. Children who had had play training improved on
social, cognitive and intellectual skills more than the children in
control groups.
Figure 9.3 Typical design of a study investigating the effects of fantasy play
training
Unfortunately, these studies are not convincing experiments

because of the inadequate nature of the control groups. In these
control groups, an adult was usually present but only in a fairly
passive way. The children in the fantasy play training groups
received a lot of extra adult conversation and encouragement.
Thus, the greater gain they made could be due either to the
increase in fantasy or socio-dramatic play, or to the extra adult
involvement. To choose between these alternatives, a better control
condition is needed. I, and some other investigators, have
compared fantasy play training with skills training. Skills training
involves an equal amount of adult involvement, but not in a fantasy
157
Peter K.Smith
context. Rather, children might be encouraged in shape and colour

matching, jigsaws, picture dominoes, and so on. The play of such a
research study (Smith, et al. 1981) is shown in Figure 9.4. It is an
expansion of the design shown in Figure 9.3; it includes an
additional follow-up assessment.
The results of this and similar studies (Christie 1983; Christie and
Johnsen 1985) show that the fantasy play training and the skills
training conditions are about equally effective. So far as the
Figure 9.4 Plan of a study comparing the effects of play tutoring and skills
tutoring (from Smith, et al. 1981)
158
cognitive and language skills are concerned, children in either

condition made gains, irrespective of which condition they
experienced. This suggests that it is adult involvement which is the
crucial factor, not the encouragement of fantasy. The children in
fantasy play training have sometimes been found to be more
sociable with other children, and more physically active, after the
training is over. This may relate to the kinds of activities encouraged
in the fantasy play and skills conditions in these particular studies
(for example, the skills training activities were often sedentary), but
which need not necessarily be the case.
Summary:
Fantasy and socio-dramatic play training is usually found to be
enjoyable for children and teachers. It is one way of promoting active
adult-child involvement. It may assist many aspects of children’s
development, but probably no more so than other kinds of
involvement which do not involve fantasy or socio-dramatic play. It
is, however, a useful way of increasing adult-child involvement
without detracting from child-child interactions, and it may indeed
enhance the latter.
Correlational studies of early childhood curricula varying in their use

of spontaneous or structured play
These studies have looked at different schools or nurseries, and
correlated the kinds of ways in which they do or do not use the
opportunity for play (as, for example, in Figure 9.1) with the behaviour
of the children and assessments of their development:
(i) In one study (Prescott 1973) 112 children aged 2 to 5 years were
observed in fourteen day-care institutions in the USA. Each child
was watched for about three hours. Of the fourteen pre-schools,
seven were broadly type A in Figure 9.1, and seven were broadly
type C. The children in type C centres were more physically active
and exploratory, but less often challenged in cognitive ways. The
children in type A centres directed more attention to adults, and
received more adult guidance. They were observed to meet
teacher’s expectations more, but to be more often rejected or
frustrated in their attempts to communicate or interact with
teachers or peers.
(ii) In another study, 141 children aged 2 to 5 years were observed in
five different pre-school institutions in the USA (Huston-Stein et al.
1977). These institutions varied in the extent of adult-directed
activities (that is, types A,B versus C,D in Figure 9.1). There was
159
Peter K.Smith
more fantasy play in those institutions with less adult direction.

Children in these institutions tended to be both more helpful, and
more aggressive, to other children. Children in the institutions with
more adult direction showed a higher level of conformity to adult
expectations. They waited more patiently during delays, were more
attentive during group activities, and took more responsibility for
tidying up toys at the end of play sessions.
(iii) In another study of nineteen pre-school institutions in Oxfordshire
for 3- to 4-year-olds (some nursery schools, some parent-
organized playgroups) a distinction was made between ‘structured
programmes’ and ‘free programmes’ (Sylva et al. 1980). All the
institutions allowed some free play but in the ‘structured
programmes’ there were also some compulsory, group-led
activities. Children in the free programmes spent more time in
fantasy play, but less time with structured activities. The observers
felt that the children were presented with more cognitive
challenges in the structured programmes. Children who
experienced some structured sessions, also played in a more
mature and challenging way when free play was allowed. The
conclusion of this study was that some structured sessions which
involved an adult with one or a few children were beneficial, but
that too much structure (for example, largely or entirely adult-led
activities) led to a decline again in cognitive challenge. There was
an optimal amount of time spent in structured activities, preferably
in small group work.
(iv) In a further study by the same investigators (Jowett and Sylva
1986), ninety children were observed when they were just starting
primary school, at 5 years of age. Forty-five had previously
attended nursery schools, usually characterized by more adult-
structured activities (usually types A and B in Figure 9.1), while
forty-five had attended play-groups (usually types C or D in Figure
9.1). The children were matched for age, sex and family structure.
It was found that the children who had attended nursery schools
showed a higher level of cognitive challenge in their free play
activities. They also spent more time on self-initiated writing and
use of work cards, and were more persistent in the face of
difficulties. In general, the children who had attended nursery
school seemed more ready for primary school and more task
oriented. Again, it should be emphasized that the nursery school
curricula studied here were not structured all the time, and some
considerable amount of free play and directed play would still have
occurred.
(v) A study in London (Gardner 1942) examined the achievements of
children in ten infant school classes. In five of these classes
160
considerable time was given over to spontaneous play activities, and

reading, writing and arithmetic were introduced largely through
play or self-initiated activity (for example, writing through helping a
child label a drawing; numbers through playing at shops). Thus,
much of the curriculum would be of type C or type B in Figure 9.1.
In the other five classes, matched for social composition, there was
most emphasis on formal instruction (type A in Figure 9.1). The
children were tested after about one and a half years in the classes.
The children in the play-centred classes scored higher on measures
of creativity and artistic achievement, such as assembling materials
into pictures, and expressing ideas through drawing. They tended to
stay at an activity for longer periods, and to score higher on physical
training and some aspects of language development. They were
more friendly with other children and adults.
There were no differences between the classes with respect to
neatness, memory, vocabulary or reading skills. The children from
the classes with formal instruction were clearly better at writing
skills, and tended to be better at mathematical skills,
(vi) In a study of Canadian 5- to 7-year-olds (Evans 1979), observations
were made in twenty school classrooms. In ten of these, the
curriculum was described as ‘play-based’ or ‘child-centred’ (types C
or B in Figure 9.1). The other ten comparison classes were less child-
centred (type A in Figure 9.1). The children in the play-based
classrooms showed more fantasy and role play, and more exploratory
play. They engaged in more social interaction with other children, and
more independent work which they initiated. In the comparison
classes, children spent more time in teacher-led group experiences
such as reading, word analysis and phonics and printing.
At the end of the school year, the children in the twenty
classrooms were tested on a number of measures. There was little or
no difference between the ‘play-based’ and comparison classes in
language development, perceptual memory, role-taking skills,
understanding of classification, or non-verbal intelligence.
However, the children in the comparison classes did significantly
better in two areas: reading and mathematics. They scored higher on
tests of reading ability, reading comprehension, and understanding
and testing mathematical concepts.
Summary:
These correlational studies, although they vary in detail, suggest that
children in more ‘play-directed’ pre-school programmes engage in more
playful and exploratory activities, and may interact more with other
children. However, their play may be less cognitively complex, or
161
Peter K.Smith
challenging, than children who experience at least some degree or

amount of adult-structured activities. More adult structure leads to
greater conformity to institutional demands, and, at 5 to 7 years, to better
achievement in reading and mathematics.
The conclusions of these studies are limited for two main
methodological reasons. First, children were not assigned randomly to
different pre-school curricula. Thus, differences in curricula may be
partly due to differences in the children as they entered the pre-schools.
Second, the curricula may have varied in a number of ways, not just
whether they were play-based or not. For example, they may have
varied in staff teaching, staff-child ratio, availability of space and
equipment. These other factors may confound some of the results
reported.
A survey by Osborne and Millbank (1987) of the progress of nearly
7,000 children who had experienced different kinds of pre-school
provision (or none) has revealed the complexity of such relationships.
The children were assessed on cognitive and behavioural measures at
both 5 and 10 years of age. The effects of a large number of child
background variables were taken account of statistically, such as social
disadvantage, family size and composition. The main finding (having
allowed for these factors) was that, compared to having no pre-school
experience, attendance at pre-school did seem to help cognitive and
educational development, the effect still being present at ten years. Some
differences were found in the long-term effects of different types of pre-
school, though these were not directly related to curricula effects or the
kind of model shown in Figure 9.1. However, these differences were
small, and the authors state that ‘our findings persistently show that
despite the many and considerable differences we have found between
LEA nursery schools, playgroups, day nurseries, etc., their apparent
effect on children’s educational progress was remarkably similar’
(Osborne and Millbank 1987:241).
Experimental studies of early childhood curricula varying in their use

of spontaneous or structured play
These studies have randomly assigned children to different curricula.
The curricula have planned contents, and vary in the use made of
spontaneous or structured play:
(i) In one study (Thompson 1944) nineteen 4-year-old children were

randomly assigned, eleven to a free play pre-school programme
(type D in Figure 9.1) and eight to a more adult-directed programme
(probably like type B in Figure 9.1). After an eight- month period, it
was found that the children who had the more adult-directed
programme were more constructive when faced with possible
162
failure, and showed more leadership, sociability and social

competence with other children. There were no differences on
intelligence test scores.
(ii) In another study (Smith and Connolly 1980), forty-four children
were randomly assigned, twenty-two to a free play pre-school
programme (type D) and twenty-two to a more adult-directed
programme (mixed type, A, B). This experiment also lasted for an
eight-month period. The children in the free-play programme
engaged in more fantasy play, more physically active play, and
generally interacted more with other children. They seemed to
learn to control their social relationships and especially conflicts
with other children, rather better than the children in the other
programme. The children in the adult-directed programme
engaged in more quiet activities, such as table work, and had more
conversations and contacts with the adults in the pre-school. At
times the adult-led activities did not fully gain children’s attention,
so they were more often observed as being ‘uninvolved in any
activity’. However, the adult-directed activities did seem to
improve the children’s concentration and attention span at
activities. This progressively improved over the eight-month
period. At the end of the period both groups of children
experienced a modified free play curriculum for a few weeks, and
during this time the children who had had adult-directed activities
maintained their greater span of attention. No difference was found
between the two groups of children on tests of cognitive ability, or
language skills;
(iii) In another study (Miller and Dyer 1975), over 200 4-year old children
were randomly assigned to fourteen pre-school classes. Eight of these
were highly adult-directed (type A in Figure 9.1), while six were more
child-centred (types B or C in Figure 9.1). The children were tested
after eight weeks and again after six months. It was found that the
children in the structured, adult-centred classes tended to do better in
tests of cognitive and academic ability. The children in the more child-
centred classes showed more inventiveness and curiosity and greater
social participation with other children. A follow-up was made of the
children four years later, at primary school. The differences found in
areas of cognitive and academic ability at age four were not found at
this later time point, though some differences in the non-cognitive
areas were more stable.
(iv) In a further study (Schweinhart et al. 1986), sixty-eight children
aged about 4 years were assigned to either a very adult-structured
programme (type A in Figure 9.1), a programme which involved
adults helping children plan activities (type B in Figure 9.1), and a
more child-centred programme (type C in Figure 9.1). The children
163
Peter K.Smith
were assessed while at pre-school, and also through their years of

school experience. The children experiencing the adult-structured
(type A) programme showed at first the greatest gains in intellectual
ability, but by the age of 10 years there was no remaining difference
between the groups on this measure. On a subsequent assessment,
when the children were 15 years of age, it was reported that the
children who had experienced the other two programmes (types B
and C) scored higher on good relationships with family participation
in sports, and positive social adjustment.
Summary:
These experimental studies suggest that adult structuring of activities is
beneficial for children’s immediate concentration and attention span.
Highly formal (type A) programmes may lead to the most rapid cognitive
gains, but these may not be long lasting. If some degree of child choice is
allowed, this stimulates curiosity and sociality with other children, and
one study has found that such effects may be more long lasting than any
cognitive effects of pre-school.
Physical activity and rough-and-tumble play

Almost all of the research on children’s play has concentrated on
constructive and dramatic or fantasy play, and it is these forms of play
that teachers have sometimes encouraged or structured in the pre-
school and infant school. However, young children can also spend
quite a lot of time in physical activity play (such as running, climbing,
sliding, swinging) either freely or with large apparatus such as
climbing frames, and in rough-and-tumble play, where they chase and
grapple with each other in an apparently friendly way. These kinds of
play are more likely when plenty of space is provided, and in outdoor
play. Rough-and-tumble play is more likely on soft surfaces. Young
children confined for a while into a fairly restricted indoor space, are
likely to engage in markedly increased physically active play when
more space is available.
Young children are naturally active and need physical exercise for
healthy physical growth and muscle development. Free outdoor play can
provide this, though physical exercise can also be obtained from
organized physical activities such as gymnastics and dance routines.
Rough-and-tumble play can involve large groups of children and may
have some benefits in the social domain. Children generally choose to
rough-and-tumble with friends, and possibly the activity helps to
maintain friendship bonds. Also, in rough-and-tumble children need
specific skills to distinguish play from aggressive intent, and they
164
exercise reciprocal turn-taking skills such as chasing, then fleeing. This

has led Pellegrini (1987) to hypothesize that rough-and-tumble play can
facilitate social skills in children. It is not usually thought to have
educational benefits in the cognitive sense. Some teachers discourage it
or do not allow it because it can be noisy and disruptive. Other teachers
allow it as a natural social behaviour which is very enjoyable for some
children. Rough-and-tumble play seems to be clearly distinct from real
fighting, so far as young children are concerned. For older children
(approaching adolescence) it has been suggested that the distinction
cannot be made so clearly. For a review see Humphreys and Smith (1984)
or Smith (in press).
Conclusions
1. Spontaneous fantasy and socio-dramatic play is enjoyable for most
young children. Some theorists believe it has intrinsic value for
cognitive growth, but the empirical evidence tends to support the
view that a child’s fantasy play reflects its level of cognitive
development rather than actively encouraging it. Such play may
also be valuable for emotional development, but firm evidence is
lacking.
2. Fantasy and socio-dramatic play can be readily encouraged and
developed in younger children by school staff. This provides one
way of facilitating adult-child interaction within the framework of
an approach which allows scope for the child’s own initiative. There
may be benefits for the child’s cognitive and social development,
though probably no more than could be obtained from other forms
of structured activities not involving fantasy.
3. In general, both correlational and experimental studies show that
pre-school and infant school curricula which encourage the child to
use some initiative in the choice of activities (types B and C in
Figure 9.1), enhance children’s curiosity and exploratory activity,
and the quality of social relationships between children.
4. In general, both correlational and experimental studies show that
pre-school and infant school curricula in which the staff play an
active role in initiating and structuring activities (types A and B in
Figure 9.1), enhance children’s concentration span and persistence,
the amount of cognitive challenge with which they are presented,
their level of cognitive development and reading, writing and
mathematical skills. Also, children conform better to immediate
adult expectations and institutional requirements.
5. There is some evidence that the effects of early school experience on
the child’s cognitive development are less long lasting than the
effects on personality and social relationships.
165
Peter K.Smith
6. There may be advantages in an approach (such as type B in Figure

9.1) which tries to combine the cognitive advantages of adult
structuring with the social advantages of allowing child choice. This
can be done, for example, by adults helping children to plan their
own activities, support them, and review their progress. Such
approaches may or may not use ‘playful’ modes of operation; the
‘play-tutoring’ approach uses socio-dramatic play in this way, but
other, less playful, activities can also be used.
7. There may be advantages in mixing different ways of using play.
Some periods of free play allow child choice and (with outdoor
play) physical exercise and free exploration. There is some
evidence that having periods of more adult-structured activities
raises the level of maturity which children then show in free play
periods.
8. Ultimately, any choice must relate sensibly to the needs of the
children and of their society. Decisions about the role of play in the
infant school curriculum must allow for many considerations,
including (a) the previous experience and expectations with which
children come to school; (b) the different types of play shown by
young children; (c) the different ways in which play can be made use
of in school; and (d) the aims and objectives of school institutions,
within the society as a whole.
References
Bruner, J. (1972) ‘The nature and uses of immaturity’, American Psychologist 27:
687–708.
Christie, J.F. (1983) ‘The effects of play tutoring on young children’s cognitive
performance’. Journal of Educational Research 76: 326–30.
Christie, J.F. and Johnsen, E.P. (1985) ‘Questioning the results of play training
research’, Educational Psychologist 20: 7–11.
Cole, D. and LaVoie, J.C. (1985) ‘Fantasy play and related cognitive
developments in 2- to 6-year olds’. Developmental Psychology 21: 233–40.
Connolly, J.A. and Doyle, A.B. (1984) ‘Relation of social fantasy play to social
competence in preschoolers’, Developmental Psychology 20: 797–806.
Davie, C.E., Hutt, S.J., Vincent, E. and Mason, M. (1984) The Young Child at
Home, Windsor: NFER-Nelson.
Evans, M.A. (1979) ‘A comparative study of young children’s classroom
activities and learning outcomes’, British Journal of Educational Psychology
49: 15–26.
Gardner, D. (1942) Testing Results in the Infant School, London: Methuen.
Humphreys, A.P. and Smith, P.K. (1984) ‘Rough-and-tumble in preschool and
playground’, in P.K.Smith (ed.), Play in Animals and Humans, Oxford: Basil
Blackwell.
Huston-Stein, A., Freidrich-Cofer, L. and Susman, E.J. (1977) ‘The relation of
classroom structure to social behaviour, imaginative play, and self-regulation
of economically disadvantaged children’, Child Development 48: 908–16.
166
Isaacs, S. (1929) The Nursery Years, London: Routledge & Kegan Paul.
Johnson, J.E., Ershler, J. and Lawton, J.T. (1982) ‘Intellective correlates of
preschoolers’ spontaneous play’. Journal of Genetic Psychology 106: 115–
22.
Jowett, S. and Sylva, K. (1986) ‘Does kind of pre-school matter?’, Educational
Research 28: 21–31.
Krasnor, L.R. and Pepler, D.J. (1980) ‘The study of children’s play: some
suggested future directions’, in K.Rubin (ed.) Children’s Play, San Francisco:
Jossey-Bass.
Manning, K. and Sharp, A. (1977) Structuring Play in the Early Years at School,
London: Ward Lock Educational.
Miller, L.B. and Dyer, J.L. (1975) ‘Four preschool programs: their dimensions
and effects’, Monographs of the Society for Research in Child Development 40
(162).
Osborne, A.F. and Millbank, J.E. (1987) The Effects of Early Education, Oxford:
Oxford University Press.
Peisach, E. and Hardeman, M. (1986) ‘Imaginative play and logical thinking
in young children’, Journal of Genetic Psychology 146: 233–49.
Pellegrini, A.D. (1987) ‘Rough-and-tumble play: developmental and educational
significance’. Educational Psychologist 22: 23–43.
Piaget, J. (1951) Play, Dreams and Imitation in Childhood, London: Routledge &
Kegan Paul.
Plowden Report (1967) Children and Their Primary Schools, 2 vols, Report of the
Prescott, E. (1973) ‘A comparison of three types of day care and nursery school/
home care’, paper presented at Biennial Meeting of Society for Research in
Child Development, Philadelphia.
Rubin, K.H., Fein, G.G. and Vandenberg, B. (1983) ‘Play’, in P.H.Mussen and
E.M.Hetherington (eds.) Handbook of Child Psychology (4th edn) vol. 4,
Basel: S.Karger.
Schweinhart, L.J., Weikart, D.P. and Larner, M.N. (1986) ‘Consequences of three
preschool curriculum models through age 15’, Early Childhood Research
Quarterly 1: 15–45.
Simon, T. and Smith, P.K. (1985) ‘Play and problem solving: a paradigm
questioned’, Merrill-Palmer Quarterly 31: 265–77.
Singer, J.L. (1973) The Child’s World of Make-believe, New York: Academic
Press.
Smilansky, S. (1968) The Effects of Sociodramatic Play on Disadvantaged
Preschool Children, New York: Wiley.
Smith, P.K. (1988) ‘Children’s play and its role in early development: a re-
evaluation of the “play ethos”’, in A.D.Pellegrini (ed.). Psychological Bases
for Early Education, Chichester: Wiley.
Smith, P.K., (in press) ‘The role of rough-and-tumble play in the
development of social competence: theoretical perspectives and
empirical evidence’, in B. Schneider, G.Attili, J.Nadel and R.Weissberg
(eds) Social Competence in Developmental Perspective, Dordrecht,
Holland: Kluwer.
Smith, P.K. and Connolly, K.J. (1980) The Ecology of Preschool Behaviour,
Cambridge: Cambridge University Press.
Smith, P.K., Dalgleish, M. and Herzmark, G. (1981) ‘A comparison of the effects
of fantasy play tutoring and skills tutoring in nursery classes’, International
Journal of Behavioural Development 4: 421–41.
167
Peter K.Smith
Smith, P.K. and Vollstedt, R. (1985) ‘On defining play: an empirical study of the
relationship between play, and various play criteria’, Child Development, 56:
1042–50.
Sylva, K., Roy, C. and Painter, M. (1980) Childwatching at Playgroup and
Nursery School, London: Grant McIntyre.
Taylor, P.M., Exon, G. and Holley, B. (1972) A Study of Nursery Education,
London: Evans/Methuen Educational.
Thompson, G.G. (1944) ‘The social and emotional development of preschool
children under two types of educational programs’, Psychological
Monographs 56 (5).
168
Chapter ten
The nature of friendship in the

primary school
William Maxwell
William Maxwell’s chapter on friendship combines a number of
personal academic and practical perspectives. The chapter provides a
thorough overview of the development of children’s friendships,
drawing upon earlier arguments in this volume concerning the roles of
adults and peers. It poses that there is no simple interpretation or
convention through which we can understand friendships in the
classroom. And, while we can describe the roles and benefits of
friendships for children, the existence of friendship is based on
‘structured’ access to other children and the social climate within
which friendship activities take place. Maxwell also provides grounds
for key classroom questions to be asked—especially the role and
responsibility assigned to friendship groups for learning. Maxwell
provides this information from the practised eyes of an educational
psychologist
Within the review the reader is made aware of the benefits,
methodologies of studies, and developmental/structural aspects of
friendship. Each section has educational implications, which are stated in
relation to the area of study. The reviews raise many questions for the
teacher and educational researcher: from first acknowledging that
friendship has a vital role in the child’s social (and learning)
development; that not all friendships are the same, and there is evidence
that ‘close’ friendship (as with other relationships) is the most
meaningful for development; that the occurrence of friendship is
strongly affected by the context within which it develops; and that
friendship is not an all-at-once phenomenon but develops over time.
The review leads to a number of fundamental questions to be asked by
the teacher. A key question is whether teachers should rely on children’s
self-provided friendship choices for learning groups. Teachers may find
this strategy useful to initiate an informal atmosphere in the classroom,
but the informality may co-occur with discriminatory choices of gender,
ability and racial preferences. Group work can provide a very effective
169
William Maxwell
learning strategy, but its co-operative occurrence in friendship-orientated

classrooms is questioned by Galton (Chapter 2) and its effective
implementation (Slavin, Chapter 13) negates existing friendships in
preference for those established in heterogeneous groups. The reader is
referred back to considerations in this chapter of the dynamics of group
and friendship formation and to question whether teachers may more
effectively structure activities and groupings to facilitate learning-
orientated friendships. In this consideration the role of personal/social
education plays a major role in the classroom process and will lead to
more effective curricular learning as called for in the national
curriculum.
Introduction
Interaction with peers has long been acknowledged as playing a crucial
role in child development. Damon (1983), like many authors before
him, distinguishes between two social worlds in which children
operate, the world of peer relations in which children seek each other
out for companionship, affection and common amusement, and the
world of adult-child relations based upon protection, care and
instruction.
The fundamental characteristic of peer interaction which determines
its distinctive developmental potential is the characteristic of equality.
Peers, defined as children at similar stages of cognitive, social and
emotional maturity, are at least potentially, equals. In contrast to
interactions involving non-peers, there are no in-built differentials in
terms of authority, knowledge or cognitive ability imposing a
qualitatively different structure on interactions before they even begin.
Whilst the growth of social groupings and hierarchies within any
consistent peer grouping undoubtedly creates structure amongst the
individuals involved, it is negotiated within the group, not imposed
from the start. To use a term from Hartup (1978), peers are social co-
equals.
Of course, peer interaction, from an early age, tends to occur within
the context of regular groupings and within these social groups social
networks of friendships develop. These networks of relationships act to
constrain the nature and extent of interaction available to the members of
the group. To enjoy satisfactory peer interactions children need to be able
to integrate themselves into the pertaining social structure through
developing relationships with their peers, making, keeping and breaking
friendships in an appropriate manner. To a large extent children’s
friendships dominate their activity in the world of peers and their study is
correspondingly a topic of central concern to students of peer interaction
and the social development of children.
170
Friendship in the primary school
In the research literature a wide range of specific developmental

functions have been proposed for interaction between peers and,
particularly, between friends.
Some writers (for example, Hartup 1978) have stressed a socialization
function, especially with regard to the child learning to control
aggressive and sexual impulses in socially acceptable ways. This is said
to be achieved primarily through the negotiation of relations with peers
where the child directly experiences conflict resulting from unacceptable
behaviour. If an ongoing friendship has existed between two peers in
conflict with each other, then clearly the motivation to learn or apply
skills which might resolve the situation is that much greater than if no
such relationship has existed.
Other writers (for example, Rubin 1980) have stressed the role of peer
relationships in the child’s personal identity formation (for example, Mead
1934; Erikson 1968). Children construct images of themselves to a large
extent through the eyes of those others with whom they most closely
identify, that is, their peer group and, in particular, their friends. In the
course of developing a self-image the child relies on feedback from others
and on direct comparison with the attributes and characteristics of others.
The peer group has a primary role in providing the child with this
information on which his view of himself will be based.
Beyond simply providing information about him/herself to the child,
the peer group also plays a crucial role in providing the set of values
within which this information will be interpreted. The value of each
child’s particular set of friends will tend to determine which personal
characteristics are valued positively and which negatively and to what
extent they are accorded salience or importance. In this way, the extent to
which children view their own self-image in positive or negative terms
(self-esteem) is very much a product of the particular social group of
which they are a member.
The acquisition of higher level social skills through interaction with
peers has also been highlighted by some researchers (for example, Asher
and Renshaw 1981). The peer group provides arguably the most efficient
and highly motivating context for the learning and development of the
social skills which will ultimately enable children to live effectively as a
member of adult society. These are the skills of developing and
regulating appropriate relationships with others at various levels, from
professional or workplace acquaintanceships to close personal
friendships.
Further, a crucial role for peer interaction in the process of cognitive
development has been proposed. Piaget (1932) saw the conflict produced
by interacting with similarly ‘egocentric’ peers as providing a crucial
impetus in the process of decentration, a process which itself, according
to Piaget, determines the structure of cognitive development in all
171
William Maxwell
spheres. More specifically, however, researchers in the recently-

burgeoning field of social cognition have ascribed a central role for peer
interaction in the development of cognitive constructs about others
(Brooks-Gunn and Lewis 1978), about moral behaviour (Damon 1981),
about the nature of human relationships (Selman 1981) and about the
workings of society (Furth 1978). Thus, peer interaction is seen as
playing a central role in specifically social areas of cognition as well as in
promoting cognitive development generally.
Given the wide acknowledgement of the importance of children’s
friendships and peer relationships in the research literature, it is perhaps
surprising that more account has not been taken of them as a factor in
educational planning. To be sure, some of the more radical theorists in
education, such as Montessori, Froebel and Steiner, were emphasizing
the importance of children’s social and personal development through
experience with peers many decades ago, and the contemporary ‘trend
setting’ national reports in the UK have also emphasized the need for
more child-centred teaching methods with a greater emphasis on
personal and social development and greater use of autonomous pupil
groups (Scottish Education Department 1965; Plowden Report 1967).
Yet over twenty years on from these reports, which were expected to
revolutionize primary teaching, the radical shift towards pupil autonomy
and group work does not appear to have taken a widespread hold
(Kutnick 1988).
One of the factors underlying the failure to adopt the methods
recommended by Plowden may be a lack of understanding of the social
world of peer relationships and the role it plays in children’s
development. In many respects it is a closed world to adults, but research
is gradually revealing its structure and functions. Through greater
understanding it is to be hoped that teachers can more efficiently harness
the powers of the peer group, both directly to encourage the fuller social
and emotional development of pupils and to manage more effectively
group-based methods of learning in the classroom. The aim of this
chapter is to review the literature on children’s friendships and peer
relations with a view to promoting such understanding and identifying
some practical implications for classroom practice. For convenience of
organization, the review has grouped studies into the following:
sociometric models of development, behaviour with friends, and
individual differences.
Sociometric studies
One of the most prolific and enduring research traditions in the study of
children’s friendships and peer relations has been that based on the
sociometric method.
172
First introduced by Moreno (1934), the sociometric method was soon

widely adopted as a convenient way of exploring children’s peer
relationships. It has been in evidence consistently since then, adapted to
tackle various different research questions. In its basic form the
sociometric method involves asking all the members of a social group to
nominate with which other members of the group they would most like to
carry out a task, go to a social event, or simply be friends. This provides
the researcher with a network of choices between members of the group,
data which can be put to a number of uses.
Hallinan (1981), reviewing the body of sociometric research on children,
identified three different lines of research within it: first, studies of factors
correlating with popularity or sociometric status; second, studies of the
determinants of individual friendship choice; and, third, studies of how
contextual variables affect the structure of relationships within a group.
Popularity studies identify characteristics of children which
determine whether or not they find it easy or difficult to make friends.
Researchers used sociometric data to rank groups of children on the basis
of the number of nominations they received from their fellows. A
measure of peer acceptance was thus produced (for example, Potashin
1946). Later studies also asked children who they disliked amongst their
peers (for example, Moore and Updegraff 1964) and categories of
children grouped by sociometric status were produced. Thus Peery
(1979) discriminated between ‘popular’ children (in receipt of many
nominations, predominantly positive), ‘amiable’ children (few
nominations, predominantly positive), ‘isolated’ children (few
nominations, predominantly negative) and ‘rejected’ children (many
nominations, but predominantly negative). The main aim of this line of
research was to identify distinctive characteristics in popular children in
the hope that knowledge of the bases of popularity would help teachers
devise ways of helping unpopular children become better accepted. A
number of characteristics of popular children have been consistently
identified in this way. Hallinan (1981) lists the following in her review of
the research: physical attractiveness, high intelligence, academic
achievement, athletic ability, and high social class. Even the possession
of a common Christian name has been associated with popularity
(McDavid and Harari 1966). Some studies with a more explicit
theoretical basis also appeared on the scene. Thus Marshall and
McCandless (1957) found that dependence on adults was associated with
poor sociometric status, and Hartup, et al. (1967) found that more
popular children tended to dispense positive or ‘friendly’ acts to their
peers with greater frequency, although popularity was unrelated to the
frequency with which they dispensed negative or hostile acts.
It becomes apparent, however, that much of this research produced
little that could help friendless children establish better social
173
William Maxwell
relationships. Most of the characteristics identified were either very

broad and general, or were unalterable, and the correlations between
them and sociometric status were, in any case, often weak. An easy way
to boost the sociometric status of isolated and rejected children was not
found.
The only line of research which has produced a substantial body of
intervention work has been based on social learning theory (Bandura
1977). Evidence that children of poor sociometric status tend to display
less confidence in general social skills than their more popular age mates
(for example, Gottman, et al. 1975) was interpreted as indicating that
isolated or rejected children were suffering the consequences of
inadequate learning of social skills in their upbringing. It was therefore
concluded that programmes of social-skill training which would teach
these children basic social skills would enable them to then transform
their social status and become popular. A variety of training techniques
were tried. Initial studies using shaping and modelling techniques
(O’Connor 1969, 1972) produced only short-term gains which were not
replicated by later studies (Pullantz and Gottman 1981). More intensive
coaching techniques yielded more promise on average (Gottman, et al.
1976; Oden and Asher 1977; Ladd 1981), but still some replication
studies failed to get results (Hymel and Asher 1977). Some of the skills
which were being taught in these studies were, for example, the skills of
introducing oneself to new friends, maintaining a conversation with a
partner, gaining entry to a group, developing co-operative play or
managing conflict through compromise.
It may be that some of the problems encountered in creating
consistently successful social-skill-coaching programmes may be
caused by a lack of knowledge about the most effective skills to focus on
(Asher, et al. 1979). It also seems likely, however, that it is over-
simplistic to think of children’s friendship patterns as simply reflecting
their ability to perform a repertoire of key social skills. The aims and
goals which children seek to fulfil through participating in peer
relationships will have a major role to play in determining how they
interact with their peers (Renshaw and Asher 1983). Thus, a child who
may show himself to be capable of displaying appropriate social skills in
certain situations may regularly behave in a manner which inhibits
friendship development through the pursuance of maladaptive social
aims (such as that of repeatedly demonstrating dominance over others or
that of repeatedly eliciting evidence of liking from others). In these
circumstances, therefore, the child’s aims and goals have to be addressed
as well as their skills.
Disillusionment with the quest for global determinants of sociometric
status was responsible for the development of Hallinan’s second research
tradition. This was characterized by a search for the factors which
174
influence individuals’ decisions about whom they want to become

friends with, regardless of their degree of overall popularity. As children
of all levels of popularity do form friendships, it was reasoned that a
better understanding of the basis on which individuals’ compat-ibility as
friends is determined would be potentially more helpful than a general
indication of the characteristics of those at the extremes of the
sociometric range. Generally speaking, friendship choices have been
found to become more stable through the primary school years
(Horrocks and Thompson 1946), and so the basis on which they are made
presumably becomes more and more salient.
One individual characteristic which has been shown to have a very
strong influence on friendship choice is gender (Tuma and Hallinan
1979). Children tend largely to choose friends of their own sex.
Similarity of race has also been shown to have a strong effect on
friendship (St John and Lewis 1975, Singleton and Asher 1977),
although to a lesser extent Tuma and Hallinan (1979) further report a
positive relationship between similarity of children’s level of academic
achievement and stability of friendship choice. Clearly, however, these
are broad factors which act only to restrict the range of companions from
which individuals’ friendship choices are most likely to come. Any
notions of identifying factors which determine children’s friendships
have proved to be quite unrealistic.
The third research tradition identified by Hallinan is perhaps the one
with the most obvious direct implications for schools as it focuses on the
relationship between contextual or situational variables and the pattern of
relationships developing within a group. In one study Hallinan (1976)
compared the patterns of sociometric nominations emerging in
classrooms organized along traditional lines, to those emerging in
classrooms of an ‘open’ type where children were organized in groups,
and a much greater degree of interaction was encouraged. She found that
friendship choices were more evenly spread amongst the group in the
open classrooms. Choices were more polarized in the traditional ones,
with greater numbers of very popular and very isolated children being
apparent. Mutual friendship choices were also more stable in open
classrooms. It would seem, therefore, that the way in which a teacher
organizes the classroom has a significant effect on the nature of the peer
relationships experienced by the pupils, with organizing and teaching
along ‘open’ lines creating a more flexible and less exclusive type of
social structure. Teachers should clearly take this into account when
planning their teaching programmes. The methods used have broader
implications than might often be assumed. Hallinan and Tuma (Hallinan
1981) also demonstrated the influence which teachers’ patterns of
classroom organization can have on pupils’ friendships. They found that
assignment to the same small group for instruction had a positive effect
175
William Maxwell
on the formation and stability of friendship choice. They also found that
close friendships were more stable and more likely to develop if pupils
were allowed to determine the membership of their work groups. Again
the implications for teachers are clear.
Gender has been considered as a contextual variable, bearing in mind
the very strong tendency for children to form friendships with peers of
their own sex. Eder and Hallinan (1978) found that girls tend to form
smaller and more exclusive ‘cliques’, whereas boys tend to develop a
more open type of group structure. This accords with the earlier findings
of Waldrop and Halverson (1975), who found that 8-year-old girls had
more ‘intensive’ patterns of friendship in that they reported having fewer
friendships but with a greater contact and commitment.
It is perhaps this third and most recent tradition of sociometric
research which has proved to be the most promising application of the
methodology. New developments of the method are allowing
researchers to identify particular types of social structure and ‘cliques’
within larger groups. If teachers can develop a clear picture of the
social dynamics of their own classes, then social groupings can be
organized to facilitate co-operative work, making sensitive use of
individual preferences. Furthermore, changes in the overall social
structure of the class might be fostered through an alteration in
teaching methods, although the extent to which this can be achieved
remains to be determined.
Models of the development of children’s friendships

Whilst much of the sociometric research could be criticized for being
atheoretical and lacking a developmental perspective, more recently
researchers have increasingly sought to put the consideration of
children’s relationships into a developmental framework.
One fertile theoretical model for the study of how friendships develop
has been Harry Stack Sullivan’s ‘Interpersonal Theory of Psychiatry’.
Sullivan (1953) proposed a theoretical progression for the development
of children’s friendships, based on the central premise that it is through
such close peer relationships that children develop the ability to
empathize and sympathize with other individuals.
Sullivan’s model has four main stages: in the first stage (ages roughly
2 to 5), children are dependent on adults and not able to maintain peer
relationships without adult intervention. The child will simply play with
whoever is next to him or whoever he is directed at.
In the second stage, (ages 4 to 8) children have regular playmates,
independent of any adult support. These relationships are still
essentially self-centred, in that children have no commitment to
maintaining them if they are not serving their immediate needs at that
176
time. Friendship is seen as a transitory phenomenon which has no in-

built stability.
In the third stage (ages c. 7 to 12) intimacy and reciprocity appear in
the relationship and friendships become correspondingly more intense.
This is the stage of ‘pre-adolescent chumship’. At this time, Sullivan
argues, the child is rapidly developing real sensitivity to what matters to
another person. The child becomes concerned to anticipate the friend’s
thoughts and feelings and so act in a way which will please the friend,
thus further reinforcing their relationship. Friendships now have a much
greater commitment from their participants who will seek actively to
make and maintain lasting relationships of this type.
At the adolescent stage Sullivan focuses on the shift in attention
from relationships with same sex peers to heterosexual relationships,
arguing that the new challenge becomes that of translating some of the
characteristics of chumships (loyalty, mutual support, and so on) to the
context of relationships which also involve sexual dynamics.
Friendships also continue to develop, however, with an increasingly
sophisticated sensitivity to the needs, feelings, attitudes and beliefs of
the other and with a greater emphasis on loyalty and emotional self-
disclosure. In many respects the adolescent’s friendships act as
supports for their first dangerous forays into the world of heterosexual
relationships.
Whilst Sullivan’s developmental model is structured around the
emergence of the ability to empathize and sympathize with others,
Selman (1976, 1981) has proposed an alternative developmental stage
model based on the classical Piagetian stages and founded on the premise
that the development of friendships is intimately linked with the
emerging cognitive abilities of the child in non-social contexts.
Selman presents four stages, at stage 0, corresponding to the early pre-
operational stage of Piaget (age c. 3 to 6), friends are said to be merely
‘momentary physicalistic companions’, with there being no real
relationship beyond the here and now. At this stage a close friend is
simply someone with whom one is playing at the time. Friends at this
stage, Selman says, are more accurately described as playmates.
At stage 1 (ages 4 to 9: Piaget’s late pre-operational stage) there is the
beginning of a relationship but it is mainly one way. The child sees the
friend as one who provides assistance but does not appreciate any
obligation to do so in return. Thus the friend is seen as someone who does
things for one, and the notion of ‘close’ friendship is understood by the
child as simply indicating that one has a better knowledge of the friend’s
likes and dislikes.
At stage 2 (ages 6 to 12: Piaget’s stage of concrete operations)
reciprocal obligations emerge but there is only ‘fair weather’ co-
operation which easily breaks down when conflict arises. When
177
William Maxwell
conflicts do arise these tend to lead to a total breakdown of the

relationship with there being no understanding of the possibility of a
relationship continuing in spite of minor differences of opinion on
specific issues.
It is only at stage 3 (ages 9 to 15: Piaget’s stage of formal
operations) that intimate and mutually-shared relationships are
established with each partner recognizing a long-term commitment and
a need to compromise and negotiate to resolve conflict positively.
Friends now share problems and appreciate the importance of
friendship as personal support on a mutual basis. These relationships
can transcend conflicts but there is said to be an over-emphasis on
jealously-guarded two-person cliques arising from anxiety about the
difficulty of forming new friendships of this type.
At stage 4 (ages 12 to adulthood) the balance is redressed; the need for
each partner in a friendship to establish relationships with others is
recognized. There is still dependence on the relationship for support but
there is also a degree of independence and autonomy which makes such
relationships less intensely cliquey.
Both Sullivan’s and Selman’s models lay out a similar course of
developmental progression over a similar time-scale. Whilst focusing on
different aspects of social cognitive development, both also acknowledge
a subtle interplay between developing cognitive constructs and social
behaviour such that friendship plays a key role in providing a context
which facilitates progress.
Two further studies also complement, and in some way support, the
developmental progression described above. These studies are both
based on the analysis of how children responded when asked to describe
their friends. Bigelow and La Gaipa (1980) found that children in the
early school years (c. age 6 to 9) described friends as being people who
do the same things as them or live near to them or simply as people who
were ‘nice to them’. From the ages of about 8 to 12 the emphasis of their
descriptions altered to a focus on character admiration, although this
would tend to be in terms of fairly overt characteristics rather than
personality or psychological traits. From the age of 12 onwards,
however, notions such as loyalty, commitment, genuineness, intimacy
and acceptance dominate the descriptions with a recognition also of the
importance of common interests. Berndt (1981) also describes a similar
progression from definitions of friends in terms of common play,
association and pro-social behaviour in early childhood, towards
definitions in terms of intimacy, trust and loyal support as adolescence
approaches. Berndt differs from the previous authors, however, in that he
sees the development of concepts about friendship as being a gradual,
cumulative elaboration of a construct system, with many of the features
of earlier stages being retained in later stages in some form (rather than a
178
stage-by-stage progression in which each stage transforms the previous

one and renders its concepts redundant).
Overall, with regard to children’s descriptions of friends and their
understanding of friendship at a level which they can articulate out of
context, there seems to be considerable consensus about the general
course of development. This consensus can be broadly summarized in
terms of three stages of progression:
1. At the pre-school and early infant stages most children’s thinking

about friends is predominantly egocentric with only a superficial
awareness of the friend as a person and relatively low commitment
to the relationship.
2. In the middle primary years a greater awareness of the personal
characteristics and viewpoints of others develops, and the child
accordingly becomes more committed to friendships and shows
more selectivity with regard to friends. In parallel with this, more
complex forms of co-operative play develop.
3. By the latter years of primary schooling further levels of intensity
have entered friendships, and concepts of loyalty and commitment
have come to be of central importance in the context of a
relationship which involves much self-disclosure, intimacy and
meshing of views, opinions and values.
Developmental studies of children’s behaviour with friends

The models of friendship development discussed in the last section were
largely based on verbal or written evidence from children of various
ages, acquired through interviews or structured tasks. Some researchers
have attempted to follow this up with studies of how children actually
behave with their friends in an attempt to see whether or not children’s
friendship behaviour, as observed in natural or experimental settings,
supports the validity of these developmental models. The evidence from
studies of this type suggests that children may actually display higher
levels of friendship development in the context of their real-life
behaviour with friends than these developmental models of children’s
thinking about friendship may suggest.
Vandell and Mueller (1980), looking at peer relationships amongst
children of only two years of age, found evidence of lasting friendships,
with the partners tending, mutually, to seek each other out in preference
to other children and with a higher quality of interaction being evident
between the partners. They contrast this finding with Selman’s stage zero
(age range three to six years) which describes friends as merely
momentary, physicalistic companions.
179
William Maxwell
Howes (1983) studied groups of infants (median age 10 months),

toddlers (median age 20 months) and pre-schoolers (median age 42
months). She found stable friendships, defined in terms of mutual
preference, mutual enjoyment, and the ability to engage in skilful,
reciprocal interaction in all these groups, although infant social
exchanges were relatively short and generally object-centred, whilst
older children become increasingly verbal and focused on fantasy and
pretend play.
Rubin and Pepler (1980) report further observations of interactions
between pre-school friends which seem significantly in advance of what
would be expected on the basis of the cognitive developmental models.
They suggest a downward revision of Selman’s age bands is necessary if
they are to reflect children’s actual performance in the context of real
friendships rather than their ability to articulate in the abstract.
What seems to be happening is that children first advance the
sophistication of their relationships with friends in the immediate
context of their day to day social interaction. They do not immediately
acquire the ability to stand back and talk about friendship outside the
immediate context with the same level of insight. Their understanding
of relationships is, at first, context bound, and only later do they
develop more generalized concepts which they can articulate in a more
abstract setting. Furthermore, it may be the case that particular
advances in general cognitive development, established through
interaction with the world of objects in non-social contexts, are
necessary before certain types of friendship behaviour or abstract
concepts about friendship can appear. Thus, progress in the ability to
imagine and co-ordinate views of an object from different physical
perspectives may lay the groundwork for the ability to appreciate the
differing desires or motives of a friend and reconcile them with one’s
own. There is clearly a complex interplay going on between general
cognitive development, behaviour in social contexts and social
cognition.
With regard to older children, Sullivan’s stage of pre-adolescent
chumship has stimulated research, particularly with regard to his
contention that the development of a close relationship of this nature
results in an enhanced tendency to empathize and sympathize with
others. McGuire and Weisz (1982) compare children with close chums to
those without, on a measure of altruistic behaviour. They found, as
predicted, that the group with close chums performed more altruistically.
Children with close chums were also better at imagining the feelings of
another child in a different situation from themselves. Mannarino (1976)
produced similar results using an adaptation of the prisoners’ dilemma
game and a questionnaire. Again pre-adolescents with close friends
performed more altruistically than those without.
180
There does appear, therefore, to be a real relationship between having

a close friend and greater social sensitivity to the needs and feelings of
others, although the direction of causality has not been clearly
established. It is interesting to note, furthermore, that McGuire and
Weisz found no significant association between sociometric popularity
and altruism or affective perspective-taking skill. The importance of
distinguishing between friendship and popularity is again emphasized in
that the benefits of close friendship do not necessarily accrue to children
who are popular but only have relatively superficial relationships with
their associates. For close friendships to develop, with their associated
benefits in the development of empathy and social sensitivity, children
need opportunities to interact frequently and intensively with their
friendship partners, in a context which encourages mutual co-operation
in joint activity. In so far as teachers seek to develop empathy and social
sensitivity, therefore, they should plan opportunities for their pupils to
interact with their friends in this way.
Models of individual differences in children’s friendship patterns

Whilst the sociometric method has produced a significant body of
research, its most appropriate use is at the level of measuring the social
structure of a group, or identifying cliques within a group, rather than at
the level of identifying and comparing the friendships of individual
children.
When used as a basis for the identification or comparison of
individuals’ friendships, sociometric methods have a number of severe
limitations.
First, there is no guarantee that sociometric nominations will be
reflected in real substantive relationships in the playground. Children
may make nominations based on wishful thinking rather than reality, or
on the basis of a very short-term view of who they were playing with that
day rather than whom they tend to relate to in the longer term. There is
much scope for spurious nominations. This is particularly problematic
amongst pre-school and early schoolage children (Hymel 1983). With
children of this age individual nominations are only very weak evidence
that a relationship exists. This problem has led researchers to concentrate
on observational methods for the identification of relationships amongst
pre-schoolers (for example, Clark et al. 1969), and there is some
evidence that high levels of association can be taken as an indication of a
significant relationship existing (Hinde et al. 1985). However, it is also
true that association measures, taken alone, can be highly misleading in
individual cases.
Second, models of popularity or sociometric status fail to take account
of a crucial dimension along which peer relationships differ both
181
William Maxwell
between and within individuals, and that is the dimension of ‘closeness’

or ‘strength’ of the relationship.
Many authors, in recent years, have drawn attention to the importance
of clearly distinguishing between the concepts of friendship and
popularity in considering children’s peer relationships (Foot et al. 1980;
Hayes et al. 1980; Rubin 1980; McGuire and Weisz 1982; Serafica
1982). The two concepts are clearly distinct. To say that a child is more or
less popular tells you little or nothing about the closeness or strength of
the friendships. A child may be generally unpopular whilst still
maintaining a close friendship which is satisfying and rewarding.
Bearing this in mind, it is clearly naive to consider popularity or high
sociometric status to be either a necessary or sufficient condition for
social ‘success’. Rubin (1980) warns against the uncritical promotion of
popularity as a desirable end in itself as this may lead to an emphasis on
superficial relationships and an unhealthy competitive atmosphere
amongst children.
A significant friendship, to adopt a consensus definition derived from
a number of sources, is a dyadic relationship which involves mutual
positive affect and mutual preference such that the two friends prefer
each other as interaction partners. Children who are not popular may
none the less develop deep and rewarding friendships with one or two
others, effectively providing them with high quality peer interaction,
whilst children who are popular could none the less lack any friendships
of such quality. The functions proposed for close, reciprocal friendships
of this nature are usually similar to those proposed for peer relationships
in general, only more so. Thus, the ongoing stability of the friendship is
generally thought to be facilitating higher quality peer interaction which
in turn leads to more efficient acquisition of social skills, enhanced
social-cognitive development, personal identity formation and
socialization.
The validity of distinguishing between close friendships of this type
and more casual acquaintanceship is supported by a study by Hayes, et
al. (1980). They found that pre-schoolers used different concepts when
describing why they liked friends with whom they had reciprocal
friendships from those they used to describe friends with whom they had
only a unilateral relationship (that is, acquaintances with whom they
would like to be friends but currently are not). Cognitively more
advanced concepts seemed to come into play in describing reciprocated
friendships. Howes (1983) also reports some findings which support the
notion that, even at the pre-school age, close best friendships provide a
context which facilitates the development of more sophisticated levels of
social interaction between peers. She found that, over the course of a
year, pre-schoolers in stable friendship pairs showed greater increases in
the complexity of their interaction than those without such relationships.
182
Maxwell (1984) also reported similar findings. He found that

preschoolers with reciprocal best friendships engaged in higher
frequencies of co-operative play with their friends than they did with
other children.
A proper description of a child’s pattern of peer relationships,
therefore, has to take account of a number of factors and draw on a
number of sources of evidence. As well as considering the number of
friends with whom the child interacts and his/her best friend, it is also
necessary to consider the strength of these relationships, and to what
extent they are reciprocated. Maxwell (1984) provides an example of a
system of analysis using a combination of sociometric and
observational data which leads to a classification of pre-schooler’s
friendship patterns into one of three types: (i) children with a
reciprocated best friendship, that is, a strong best friendship which
seems to be of special importance to both partners; (ii) children with an
unreciprocated best friendship, so a strong relationship does not, in
fact, exist; and (iii) children with a pluralistic friendship pattern. These
children seem uninterested in forming intense best friendships but
prefer having a range of regular play companions none of whom is
especially important or significant to them. There is some evidence
from previous research (Waldrop and Halverson 1975) that a tendency
to form either close best friendships on the one hand, or pluralistic
patterns of relationships on the other, are relatively stable individual
characteristics from pre-school to early school years. An increasing
association with gender also seems to become apparent with sociable
girls tending to develop close best friendships whilst the more sociable
boys tend towards pluralistic friendship patterns.
Overall, a complex picture emerges. Whilst there is a clear tendency
for children to develop stronger, more stable friendships through the pre-
school and primary years (Horrocks and Thompson 1946) there are also
individual differences and differences between the sexes, in the extent to
which one special friendship will tend to dominate the children’s pattern
of relationships. These individual differences have to be taken account of
over and above the general developmental trends. Whilst some
developmental advantages seem to accrue in children who develop
stable, reciprocated best friendships with peers, there may be other
advantages associated with maintaining a more pluralistic pattern of
relationships. The least advantaged group, socially, are those who have
neither pattern properly established.
Summary and implications for educational practice

This chapter has reviewed a wide range of studies often coming from
disparate research traditions but, viewed overall, a coherent picture
183
William Maxwell
begins to emerge. A number of general points can be made, each of

which may have implications for educational practice.
A great deal of important learning and development results from
interactions occurring within the context of children’s friendships.
Through their experiences with friends, children learn to manage social
relationships, understand and accommodate the feelings and actions of
others, and relate to social rules and routines. Clearly then, if one of the
main aims of school is to encourage the development of pupils into well-
adjusted, socially-competent adults, then the social life of the classroom
has to be given consideration by teachers. Consideration should include
fostering appropriate relationships within the group and facilitating
constructive group activity between friends, and the rooting of
educational experiences designed to promote social or moral
understanding in the context of children’s day-to-day social experience
with their own friends.
With regard to the first consideration, it is clearly inappropriate to be
too prescriptive about how teachers might foster particular relationships
between pupils. A variety of more or less informal approaches could be
taken depending on the circumstances. In general, however, it is clear
that much might be done by planning class activities based on friendship
pairings or groupings. The important point is that these activities should
encourage co-operation on equal terms towards joint goals, and involve
the exploration of the feelings, views and personality of friends at an
appropriate age level. Much classwork could be adapted to serve these
purposes whilst also serving more conventional academic aims.
With regard to the second consideration, planning for promotion of
social and moral development in pupils should include engaging them in
structured group work focused on their relationships with friends rather
than presenting them with parables or tales with a moral message set in
contexts to which they cannot directly relate. As in other areas of the
curriculum, learning will be more efficient when it is rooted in practical
activity in a real life context with personal relevance to the child. What
better, more personally salient context, than the children’s own
friendships in which they invest so much energy and importance?
Recent sociometric work has demonstrated that teaching style and
classroom management can affect the social structure of a class. Teachers
cannot avoid affecting the friendships of their charges in that decisions
they make about teaching have consequences in that respect Group-
based teaching methods and encouraging children to have greater
autonomy in choosing who they work with seems to have positive pay-
offs in terms of more evenly spread and stable friendships within the
group. A better understanding of the social dynamics of a class will
enable teachers to use group-based teaching methods more effectively.
Although this is an under-researched area, there will be advantages in
184
grouping children on social rather than academic lines, at least for some
purposes. To take an example, co-operative story-writing might be more
productive between two friends of differing individual attainments than
between non-friends of the same academic level. Similarly, practical
problem-solving activities may be accomplished more effectively. (The
increased quality of social interaction evident between friends as
opposed to non-friends suggests that this will be so.)
Many teachers will be concerned about how best to help pupils in
their care who are clearly failing to develop appropriate friendships
within the class. Unfortunately the research reviewed provides no easy
answers in this respect. It does, however, provide a framework within
which to analyse the problem. The teacher should first consider
whether or not the child simply lacks an appropriate repertoire of social
skills with which to introduce him/herself to potential friends, maintain
constructive interaction, and resolve conflicts. If the problem is not
simply lack of knowledge of appropriate strategies which might be
taught and practised, then the teacher will need to address the nature of
the child’s social aims or goals and what the child is seeking to get out
of interacting with peers. In such cases a long-term dialogue will need
to be established with the child to increase awareness of alternative
ways of behaving with friends whilst helping to control and reduce the
types of behaviours which interfere with friendship-making. As an
element in any such programme, the teacher may also wish to harness
the co-operation of parents as the child’s behaviour may be more
clearly understood and more effectively helped with the added insight
and support thus gained.
The importance of distinguishing between popularity and friendship
should also be taken into account. It is important to recognize that a high
degree of popularity is not necessary for a child to have access to a close
friendship with its associated benefits of more sophisticated interaction
and the development of sensitivity to the feelings of others. Teachers
should encourage their pupils to value and respect friendship with its
enhanced developmental potential, rather than promoting the importance
of popularity which can, in itself, be a very superficial achievement.
Whilst there are individual differences in the extent to which children
develop close friendships and with respect to the range of friends with
whom they regularly associate, there are also general developmental
trends towards closer friendships as children proceed through the
primary years. All children can benefit from close friendships at their
own appropriate level.
With regard to the broad developmental progression in children’s
concepts of friendship, there does seem to be considerable consensus.
The insights provided should inform teaching practice at all levels but
they also have less obvious implications for the ways in which teachers
185
William Maxwell
deal with class management. With pre-adolescents, for example, it is

important to recognize the significance of loyalty to friends before
asking one child to inform on another. With younger children loyalty is
not a salient concept in this regard, and with older children it may be
possible for them to appreciate a ‘greater loyalty’ to the social group at
large. At the pre-adolescent stage loyalty between friends is crucial and
absolute, not tempered by higher order considerations of justice. To force
such a child to inform on a friend will consequently have very serious
adverse implications. Applying knowledge of children’s social
development in this way, teachers may be able to deal more sensitively
with potentially difficult situations.
Observational research on children’s behaviour with friends should
encourage teachers to pick up on instances of more advanced friendship
behaviour between pupils, even though the children involved may not be
able to articulate concepts of friendship at an equivalent level of
sophistication. As in so many areas of development the ability to
articulate concepts seems to lag behind their use in real-life situations. It
is important, therefore, not to underestimate children’s friendship
behaviour on the basis of a limited ability to articulate concepts out of
context.
In many respects personal and social development has been the
‘Cinderella’ area of the primary curriculum in that it is only given the
broadest and most unstructured type of consideration in educational
planning. Perhaps, with a greater understanding of the nature of
children’s friendships emerging from the literature, a more central
position will be given to this area of children’s progress.
References
Asher, S.R. and Renshaw, P.D. (1981) ‘Children without friends’, in S.R.Asher
and J.M.Gottman (eds) The Development of Children’s Friendships,
Cambridge: Cambridge University Press.
Asher, S.R., Renshaw, P.D., Geraci, R.L. and Dor. A.K. (1979) ‘Peer acceptance
and social skill training’, paper presented at biennial meeting of the Society
for Research in Child Development, San Francisco, USA.
Bandura, A. (1977) Social Learning Theory, Englewoods Cliffs, NJ: Prentice-
Hall.
Berndt, T.J. (1981) ‘Relations between social cognition, non-social cognition and
social behaviour: the case of friendship’, in J.H.Flavell and L.D.Ross (eds)
Social Cognitive Development, New York: Cambridge University Press.
Bigelow, B.J. and La Gaipa, J.J. (1980) ‘The development of friendship values
and choice’, in H.C.Foot, A.J.Chapman and J.R.Smith (eds) Friendship and
Social Relations in Children, Chichester: Wiley.
Brooks-Gunn, J. and Lewis, M. (1978) ‘Early social knowledge: the development
of knowledge about others’, in H.McGurk (ed.) Issues in Childhood Social
Development, London: Methuen.
186
Clark, A.H., Wyon, S.M. and Richards. M.P.M. (1969) ‘Free play in nursery
school children’. Journal of Child Psychology and Psychiatry 10: 205–16.
Damon, W. (1981) ‘Exploring children’s social cognition on two fronts’, in J.H.
Flavell and L.Ross (eds) Social Cognitive Development, Cambridge:
Cambridge University Press.
Damon, W. (1983) Social and Personality Development, New York: W.W. Norton.
Eder, G. and Hallinan, M. (1978) ‘Sex differences in children’s friendships’,
American Sociological Review 43: 237–50.
Erikson, E.H. (1968) Identity, Youth and Crisis, New York: W.W.Norton.
Foot, H.C., Chapman, A.J. and Smith, J.R. (1980) ‘Patterns of interaction in
children’s friendships’, in H.C.Foot, A.J.Chapman and J.R.Smith (eds)
Friendship and Social Relations in Children, Chichester: Wiley.
Furth, H.G. (1978) ‘Young children’s understanding of society’, in H.McGurk
(ed.) Issues in Childhood Social Development, London: Methuen.
Gottman, J.M., Gonso, J. and Rasmussen, B. (1975) ‘Social interaction, social
competence and friendship in children’. Child Development 46: 709–18.
Gottman, J.M., Gonso, J. and Schuler, P. (1976) ‘Teaching social skills to isolated
children’, Journal of Abnormal and Social Psychology 4: 179–97.
Hallinan, M.T. (1976) ‘Friendship patterns in open and traditional classrooms’,
Sociology of Education 49: 254–65.
Hallinan, M.T. (1981) ‘Recent advances in sociometry’, in S.R.Asher and J.M.
Gottman (eds) The Development of Children’s Friendships, Cambridge:
Hartup, W.W. (1978) ‘Children and their friends’, in H.McGurk (ed.) Issues in
Childhood Social Development, London: Methuen.
Hartup, W.W., Glazer, J.A. and Charlesworth, R. (1967) ‘Peer reinforcement and
sociometric status’, Child Development 38: 1017–24.
Hayes, D.S., Gershman, E. and Bolin, L.J. (1980) ‘Friends and enemies:
cognitive bases for preschool children’s unilateral and reciprocal
relationships’, Child Development 51: 1276–9.
Hinde, R.A., Titmus, G., Easton, D. and Tamplin, A. (1985) ‘Incidence of
“friendship” and behaviour towards strong associates versus nonassociates in
preschoolers’, Child Development 56: 234–45.
Horrocks, J.E. and Thompson, G.G. (1946) ‘A study of the friendship
fluctuations of rural boys and girls’, Journal of Genetic Psychology 69: 189–
98.
Howes, C. (1983) ‘Patterns of friendship’, Child Development 54: 1041–53.
Hymel, S. (1983) ‘Preschool children’s peer relations: issues in sociometric
assessment’, Merrill-Palmer Quarterly 29: 237–60.
Hymel, D. and Asher, S.R. (1977) ‘Assessment and Training of Isolated
Children’s Social Skills’, paper presented at the biennial meeting of the
Society for Research in Child Development, New Orleans, USA.
Kutnick, P.J. (1988) Relationships in the Primary School Classroom, London:
Chapman.
Ladd, G.W. (1981) ‘Effectiveness of a social learning method for enhancing
children’s social interaction and peer acceptance’. Child Development 52:
171–8.
McDavid, J.W. and Harari, H. (1966) ‘Stereotyping of names and popularity in
grade school children’, Child Development 37: 453–9.
McGuire, K.D. and Weisz. J.R. (1982) ‘Social cognition and behaviour
correlates of pre-adolescent chumship’, Child Development 53: 1478–84.
187
William Maxwell
Mannarino, A.P. (1976) ‘Friendship patterns and altruistic behaviour in

preadolescent males’, Developmental Psychology 12: 555–6.
Marshall, H.R. and McCandless, B.R. (1957) ‘Relationships between
dependence on adults and social acceptance by peers’. Child Development 28:
413–19.
Maxwell, W.S. (1984) ‘Patterns of friendship and interaction style in young
children’, unpublished PhD thesis, Edinburgh University.
Mead, G.H. (1934) Mind, Self and Society, Chicago: University of Chicago Press.
Moore, S.G. and Updegraff, R. (1964) ‘Sociometric status of preschool children
as related to age, sex, nurturance giving and dependency’. Child Development
35: 519–24.
Moreno, J.L. (1934) Who Shall Survive?, Washington: Nervous and Mental
Disease Publishing Company.
O’Connor, R.D. (1969) ‘Modification of social withdrawal through symbolic
modelling’. Journal of Applied Behaviour Analysis 2: 15–22.
O’Connor, R.D. (1972) ‘Relative efficacy of modelling, shaping and the
combined procedures for modification of social withdrawal’, Journal of
Abnormal and Social Psychology 79: 327–34.
Oden, S. and Asher, S.R. (1977) ‘Coaching children in social skills for friendship
making’, Child Development 48: 495–506.
Peery, J.C. (1979) ‘Popular, amiable, isolated, rejected: a reconceptualisation of
sociometric status in preschool children’. Child Development 50: 1231–4.
Piaget, J. (1932) The Moral Judgement of the Child, New York: Free Press.
Plowden Report (1967) Children and their Primary Schools, 2 vols, Report of the
Central Advisory Council for Education in England: HMSO.
Potashin, R. (1946) ‘A sociometric study of children’s friendships’, Sociometry
9: 48–70.
Pullantz, M. and Gottman, J.M. (1981) ‘Social skills and group acceptance’, in
S.R.Asher and J.M.Gottman (eds) The Development of Children’s
Friendships, Cambridge: Cambridge University Press.
Renshaw, P.D. and Asher, S.R. (1983) ‘Children’s goals and strategies for social
interaction’, Merrill-Palmer Quarterly 29: 353–74.
Rubin, Z. (1980) Children’s Friendships, London: Fontana.
Rubin, K.H. and Pepler, D.J. (1980) ‘The relationship of child’s play to
socialcognitive growth and development’, in H.C.Foot, A.J.Chapman and
J.R.Smith (eds) Friendship and Social Relations in Children, Chichester:
Wiley.
Scottish Education Department (1965) Primary Education in Scotland (The
Primary Memorandum), Edinburgh: HMSO.
Selman, R.L. (1976) ‘Towards a structural analysis of developing interpersonal
relations concepts: research with normal and disturbed preadolescents’, in
A.D.Pick (ed.) Minnesota Symposium on Child Psychology, vol. 10,
Minneapolis: University of Minnesota Press.
Selman, R.L. (1981) ‘The child as a friendship philosopher’, in S.R.Asher and
J.M.Gottman (eds) The Development of Children’s Friendships, Cambridge:
Serafica, F.C. (1982) ‘Conceptions of friendship and interaction between friends:
an organismic-developmental perspective’, in F.C.Serafica (ed.) Social-
Cognitive Development in Context, London: Methuen.
Singleton, L.C. and Asher, S.R. (1977) ‘Peer preference and social interaction
amongst third grade children in an integrated school district’. Journal of
188
St John, N. and Lewis, R. (1975) ‘Race and the social structure of the elementary
classroom’, Sociology of Education 48: 346–68.
Sullivan, H.S. (1953) The Interpersonal Theory of Psychiatry, New York: Norton.
Tuma, N. and Hallinan, M. (1979) ‘The effects of sex, race and achievement in
school on children’s friendships’, Social Forces 57: 1265–85.
Vandell, D.L. and Mueller, E.C. (1980) ‘Peer play and friendships during the first
two years’, in H.C.Foot, A.J.Chapman and J.R.Smith (eds) Friendship and
Social Relations in Children, Chichester: Wiley.
Waldrop, M.F. and Halverson, C.F. (1975) ‘Intensive and extensive peer
behaviour: longitudinal and cross-sectional analysis’, Child Development 46:
19–26.
189
Chapter eleven
Sex roles in the primary classroom
Paul Croll and Diana Moses
The last decade or so has seen a notable increase in the amount of
research attention directed at the issue of gender differentiation within
the school. As Croll and Moses point out here, it is clear that within
society generally one’s gender has a critical impact upon one’s life
chances. It is natural to assume that school reflects, and is, at least,
partially responsible for these differences.
At the start of the decade clear differences existed in the 16+ age
range, and while these differences are subject to change they are
clearly still significant. Croll and Moses’ task here is to assess
available evidence with a view to establishing the role of the primary
school in this. As will be seen, their conclusion is that while
differences between boys and girls clearly do exist at primary school,
the evidence suggests that girls are not educationally disadvantaged by
their experiences there.
Croll and Moses suggest that researchers need to pay closer attention
to an examination of the links between experiences in the primary school
and later developments at secondary level. No adequate account of these
links currently exists, but factors related to self-esteem and motivation
are likely to prove significant. The reader is therefore asked to consider
the case made out here together with the issues discussed by Schunk and
Rogers (Chapters 5 and 6).
Croll and Moses also argue that studies picking out gender differences
may be influenced by the different levels of special educational needs
and disruptive behaviour found in boys and girls. The reader is also,
therefore, directed specifically to Chapters 12 and 13 on special needs
and co-operative group work.
Introduction
The focus of attention of this chapter is an exploration of the differences
between the educational experiences of boys and girls in their primary
190
schools. Are girls and boys treated differently from one another and do
they have different educational experiences in the early years of
schooling? If so, what significance and consequences does such
differentiation have? Do boys and girls develop different skills and
attitudes, for example, and do any differences between classroom
experiences and educational attitudes and achievements which emerge in
the primary school years have implications for later educational
performance and in the adult world of work?
As we shall see later in this chapter, there is evidence that girls flourish
in the primary school. They at least hold their own on measures of
academic attainment compared with boys and they are much less likely
to experience problems, either in connection with their academic
performance or with their behavioural and emotional adjustment to
school. On the other hand we know that, later in life, females are most
unlikely to enjoy the same degree of occupational success and financial
rewards as their male peers. In Britain in the 1980s women still do not
enjoy the same occupational opportunities as men. They earn
considerably less on average and, even within the same occupational
groups, tend to have lower incomes. Much of the work done exclusively
or virtually exclusively by women is particularly poorly paid. Women are
underrepresented at the top of all occupations, even those such as nursing
and primary school teaching which have traditionally been the preserve
of women and in which they still predominate numerically (see Equal
Opportunities Commission 1987, for a summary of some of the relevant
statistical information on the economic situation of women).
The question naturally arises of whether, despite the apparently
satisfactory performance of girls at primary level, they are, nevertheless,
in some way being prepared to be second best to boys. A number of
writers have suggested that the educational system is, in various ways,
implicated in the disadvantages females appear to suffer in later life (for
example, Acker 1984, 1988). Related to such arguments is the suggestion
put forward by Sara Delamont that schools do not simply reflect the
different sex roles and gender differentiation of the wider society but
exaggerate and amplify such distinctions (Delamont 1980, 1983).
Delamont argues that schools operate as a conservative force in which
traditional female roles are emphasized and exaggerated and in which
females suffer discrimination greater than that prevailing elsewhere.
In this chapter we shall attempt to summarize the available
information on various aspects of the different school experiences,
attainments and responses of girls and boys in British primary schools.
First we shall consider the extent of gender differentiation in the
everyday experience of the primary classroom. This analysis will be
extended to look at possible disadvantaging features of classroom life,
especially those related to teacher-pupil interaction. We shall then
191
examine the learning outcomes for pupils, comparing the attainments in

academic subjects and the extent of academic and related difficulties
experienced by girls and boys. Finally we shall consider evidence on
differences between female and male pupils’ attitudes to and perceptions
of school and differences in the ways that teachers perceive male and
female pupils. Following the presentation of the empirical evidence on
gender differences in the experience of primary education, we shall
return to the question of the link between early educational experience
and later occupational and financial disadvantage, and also consider
whether schools exaggerate gender differentiation compared with that
existing in the wider society.
Gender differentiation in the classroom

Gender differentiation is a prominent feature of this and other societies
and it is not surprising that differentiation between boys and girls should
be a feature of schools. Schools cannot exist in a vacuum and by the time
they come to school, and then alongside their school experience, children
have learnt and continue to learn a great deal about the world from their
parents, their peers, books, comics, television and other sources. When
they come to school at the age of four or five the child will have a clear
image of her/himself as a boy or girl. The great majority of given names
are specific to males or females, similarly, the clothes children wear are
often gender specific. Boys and girls are likely to have been treated rather
differently from the time that they were born and, for instance, given
different toys to play with. Evidence on sex differences in pre-school
play is given by Serbin (1983).
Studies such as those of King (1978) have documented the routine
pattern of gender differentiation in a variety of school routines and
activities. Teachers also automatically used gender as an organizing
category within the classrooms in his study. Boys’ and girls’ names were
listed separately on the register, coats were hanged up separately, record
cards were in different colours for boys and for girls, histograms
containing data from the children were done separately for girls and boys
and so on. King notes that a significant feature of these practices was that
they were ‘taken for granted’ and not regarded as in any way problematic
by the teachers who did not think they needed explaining, still less
justifying. In a study of a Scottish primary school Hartley (1985)
describes the predominance of organizational practices based on gender
and also notes that children’s friendships tend to be same-sex specific.
Another feature of the classrooms described by King was the use of
competition between boys and girls both as a strategy for control and as
an organizational feature of the classroom. Contrasts between the
behaviour of the groups were made by teachers and ‘gender-
192
inappropriate’ behaviour commented on. Task were sometimes divided

between boys and girls with a competitive element introduced.
However, gender differentiation is not always a feature of descriptions
of classrooms. As Delamont (1980) points out, the British observational
studies of classrooms conducted before that of King made no reference at
all to gender. Later studies have not generally seen gender as central
although studies such as those of Hartley (1985) consider gender
differences. Pollard (1985) considers gender as an issue but concludes
that other differentiating characteristics of children are more relevant for
his analysis of the social experience of schooling. He writes of ‘…the
close similarity in the perspectives of boys’ and girls’ friendship groups
regarding school which led me to focus primarily on the goody, joker and
gang distinction as having more analytic power than that of gender’
(Pollard 1985:195). King, who is one of the researchers to put most
emphasis on gender differentiation, also stresses that this is only one of
the differentiating factors in the classroom and not the most important of
these (King 1978).
What emerges from these studies is that primary classrooms are, in a
commonsense fashion, heavily gender differentiated. Primary classes
are made up of little boys and little girls rather than little children and
reference to this and the use of it for organization, control and class
management is a routine feature of teaching in the primary school.
Such differentiation is probably the most obvious single feature of
primary classrooms and the lack of reference to it in many discussions
of teaching is an indication of its ‘taken-for-granted’ nature. This is
different, however, from saying that gender differentiation is the most
important feature of primary classrooms or, in particular, that it
operates in a way that disadvantages one or other of the sexes. Some
possible ways in which the different experiences of the primary
classrooms may in fact be disadvantaging, in particular to girls, will be
considered below.
Teacher-pupil interaction in the classroom

One possible difference in the classroom experience of boys and girls
which has received recent attention is that girls may be disadvantaged in
the classroom by receiving less attention from the teacher than is given to
boys. Some fairly extreme claims about this gender imbalance have been
made, especially by writers approaching the issue from a feminist
perspective. Perhaps the best-known work in this tradition is that of
Spender who writes of boys receiving ‘…so much more attention from
teachers than do girls…’ (Spender 1982:54). Spender also claims that
gender imbalances are so routinized and expected in classrooms that
even when teachers are trying to equalize attention girls get only just over
193
a third of the teacher’s time (Spender 1982). However, no details are

given of the basis of these assertions.
Similarly, Stanworth (1981), reporting on a study of pupil perceptions
of classroom interactions, shows pupils reporting boys as twice as likely
to seek teacher attention and four times as likely to offer contributions to
discussion. Buswell (1981:196) writes of lessons in which female pupils
took no part at all and that there were ‘…many more classes [in which]
girls received only minimal attention compared with boys’.
Some of these assertions come from studies of classrooms using a
qualitative observational methodology. Other studies of classrooms
using such an approach do not report such differences. As was shown
above, Pollard (1985), in an extensive observational study of primary
schools, made little use of gender as an explanatory category. Similarly,
gender differences do not emerge from the observations in infant
classrooms reported by Sharp and Green (1975). King (1978) and
Hartley (1985) both focus on gender differences in their studies of
primary classrooms but do not report differences in levels of teacher
interaction.
The obvious way to establish whether boys get more attention than
girls in the primary classroom is to observe in a substantial sample of
classrooms and to count or time in a systematic fashion the number of
teacher interactions which are directed to boys and girls and see if they
differ. If we wish to go beyond statements about the relative number of
interactions and say something about the type of interaction it will also
be necessary to distinguish between different types of interaction in the
counting or timing procedure. Data relevant to this issue come from the
many studies which have conducted systematic observation in
classrooms.
Systematic observation provides a method of getting precise
quantitative data on aspects of interaction and behaviour in classrooms.
Observations are made by using a predetermined set of categories and a
precise timing system. For example, a very simple system relevant to the
present discussion would be one where an observer focused observation
on a teacher and every ten seconds noted whether or not the teacher was
involved in a one-to-one interaction with a pupil and, if so, whether this
was with a boy or a girl. Such an observation system allows a precise
estimate to be given of the amount of teacher interaction with male and
female pupils. Clearly such a system could be made more sophisticated
and could also allow the categorization of type of interaction such as
curriculum-related or non-curriculum-related, or whether the teacher or
the pupil had initiated the interaction. A full account of this kind of
classroom research can be found in Croll (1986).
A large number of studies using systematic observation have been
conducted in school classrooms, most of them in the United States, but
194
also in British classrooms. Although the issue of differential attention to

boys and girls has not usually been the main focus of these studies, many
of them provide data on such differences. Alison Kelly has recently
published a meta-analysis of 81 research studies which provide
quantitative data on the relative amount of teacher interaction received
by boys and girls in classrooms (Kelly 1988). Meta-analysis is a
technique for summarizing the results of different studies to give an
overall estimate of an effect (Kelly 1986). In addition to data on overall
differences in teacher interactions with boys and girls, Kelly also
retrieves information about different aspects of the interactions from
many of the studies.
The analysis of overall differences in the amount of teacher
interaction with boys and with girls derived from the meta-analysis
shows that girls receive 44 per cent of all classroom interactions and boys
receive 56 per cent. The type of interactions in which girls were least well
represented were those involving criticism by the teacher. Girls received
only 35 per cent of criticisms and only 32 per cent of criticisms directed
at behaviour. However, the underrepresentation of girls cannot be
accounted for simply by the boys being more heavily criticized: girls
received 44 per cent of questions, 44 per cent of response opportunities,
and 48 per cent of praise. The analysis also shows that the
underrepresentation of girls in classroom interactions does not arise from
an unwillingness to participate as girls were more likely than boys to
volunteer or put their hands up (52 per cent). They were, however, less
likely than boys to call out answers (41 per cent). Another result to
emerge from the study was that when children were divided into those
having desirable characteristics (high ability, good behaviour, and so on)
and those having undesirable characteristics (low ability, poor
behaviour), the sex bias was more pronounced among the ‘desirable’
group than among the ‘undesirable’ group.
The studies analysed by Kelly covered a variety of age ranges and not
just the primary years. Relatively few of the studies were conducted in
Britain and over 80 per cent of them were American. A breakdown of the
estimates from different groups of studies shows that in those conducted
by British authors the sex bias was very slightly less (45 per cent).
Studies conducted on 6- to 11- year-olds also show a slightly lower sex
bias (46 per cent) although the sex bias in studies of children under 6 is
much greater (41 per cent).
This sort of meta-analysis provides a very valuable overall estimate
from a wide range of studies, but it is not possible to derive from it specific
estimates for British primary schools. Moreover, a number of recent major
studies of British primary schools were not included in the analysis. For
these reasons, and also because it is interesting to locate estimates of
interaction differences in specific identifiable studies, a number of
195
researches using systematic observation in British primary school

classrooms will be discussed below. There are five studies published within
the last ten years which give quantifiable data on teacher-pupil interactions
in a large sample of English primary school classrooms, and which allow
distinctions to be made between interactions with male and female pupils.
These are: the ORACLE research which provides observational data on
120 classrooms across the junior age range (Galton et al. 1980), the One in
Five study (Croll and Moses 1985) which provides data on 34 second year
junior classes, the NFER study (National Foundation for Educational
Research 1987) giving data on 59 third and fourth year junior classrooms,
the Inner London Education Authority (ILEA) study (Mortimore et al.
1988) giving data on 50 Inner London junior classrooms, and the Thomas
Coram study (Tizard et al. 1988) based on observation on 30 infants’
classrooms. Of these studies only the ORACLE research was included in
the meta-analysis described above (Kelly 1988, Appendix 1), probably
because the others were published too recently for the information to be
available.
The ORACLE research (Galton et al. 1980) was the first large-sale
observational study to be conducted in British classrooms. Gender
differences were not a feature of the main analysis which concentrated
on the impact of different teaching approaches and the consequences for
classroom interaction of a predominantly individualized approach to
teaching. It showed how, even when teachers interacted mainly with
individual pupils, the main contact with the teacher of pupils in the class
was as a member of the whole class audience. This clearly has
implications for differences between boys and girls in the amount of
teacher attention they received. In the data derived from observations of
individual pupils and describing the amount of teacher interaction they
received, boys were observed to receive more individual attention than
girls. In the terms used by Kelly, boys received 54 per cent of the
teachers’ attention to individuals while the girls received 46 per cent. But
when all interactions, including whole class interactions, are included in
the analysis, girls are involved in over 49 per cent of interactions
compared with 46 per cent of individual interactions.
The research reported in One in Five (Croll and Moses 1985), and in a
subsequent analysis of these data (Croll 1986), was mainly concerned
with children with special educational needs, but also included
observational data on the classroom interactions of a control sample of
other children in second year junior classrooms. These data provide a
very similar picture of sex differences in pupil-teacher interaction to
those of the ORACLE study. Girls receive 46 per cent of individual
teacher interactions and boys 54 per cent. As in the ORACLE study, the
great majority of teacher-pupil interactions were whole class interactions
where there were no sex differences.
196
The ILEA junior school project followed the progress and school
experiences of 2,000 children as they went through 50 Inner London
junior schools and observed children in second and third year junior
classes. This research found that ‘Teachers contributed more at an
individual level with boys than with girls…’ (Mortimore et al. 1988:
167). Expressed in terms comparable with those used in the Kelly
analysis, girls received 46 per cent of the teachers’ individual attention in
the second year classroom and 42 per cent in the third year classrooms.
The major difference between boys and girls was that boys received more
comments, both critical and neutral, about their behaviour, but this did
not completely account for the gender imbalance.
These three projects all found sex differences in levels of teacher
interaction with boys and girls. The final two studies, however, found no
such differences. A project conducted by the National Foundation for
Educational Research used the ORACLE observation system to observe
language and mathematics lessons in 59 3rd and 4th year junior classes
(NFER 1987). This study found no differences in the amount of teacher
attention to boys and to girls in either mathematics lessons or language
lessons.
Finally, research conducted by the Thomas Coram Research Centre
observed pupils in 30 classes in ethnically mixed Inner London infant
schools (Tizard et al. 1988). This study also reported no differences in
overall levels of teacher interactions with boys and girls.
A consideration of Kelly’s meta-analysis and of the five British studies
discussed above shows that there is a consistent tendency for girls, on
average, to receive slightly less individual teacher attention than boys
although this is not invariably so in either individual classrooms or in all
studies. The studies reviewed by Kelly range from those which show no
differences in interaction to those showing girls receiving less than 40 per
cent of interactions. Averaged across all the studies girls received 44 per
cent of interactions. Of the five recent large-scale British studies, three
showed lower levels of interaction with girls of an extent broadly
comparable to Kelly’s results while two showed no differences. No study
has shown girls receiving more individual teacher attention than boys. The
studies which show boys getting more attention almost invariably show
that this is in part due to boys receiving substantially more behavioural
criticism, but the studies also show that this does not account for the overall
differences in levels of interaction. Compared with the claims reported
above about the discrepancies in teacher attention across the sexes, the
differences reported here are not large but, as Kelly (1988) points out, over
a school career they amount to a considerably lower level of direct contact
with the teacher for individual girls.
Various explanations have been put forward for this pattern of
inequality in teacher interaction. Some authors have suggested that sexist
197
bias is so prevalent in society that male dominance is ‘natural’ and

effectively invisible (for example Megarry 1981; Spender 1982). Kelly
(1988) suggests that the emphasis that teachers put on working with
children as individuals and responding to their individual needs may
mean that lack of equality between groups such as males and females is
not apparent or not important to them. An observation originally made by
French and French (1984) is that there is not a general pattern of higher
levels of interaction with all boys but rather a tendency for a few boys to
get more attention than other pupils generally. This explanation was
tested by Croll (1986) in an analysis of classroom interaction in 34 2nd
year junior classes. The hypothesis that the increased attention to boys
occurs because teachers regard a much higher proportion of boys than
girls as having special educational needs was also investigated. The
results of this analysis suggest that to some extent the increased attention
to boys is a result of a higher proportion of boys having special
educational needs. Both boys and girls, identified as having special needs
received very much above average levels of teacher attention which were
equal across the sexes. This matches the result reported in Kelly’s meta-
analysis that sex imbalances were greatest among pupils having
‘desirable’ characteristics and least among pupils having ‘undesirable’
characteristics. However, these differences did not entirely account for
the imbalance of attention to boys and girls and the suggestion put
forward by French and French was also supported. In these classrooms a
few boys, who were not identified as having special needs, nevertheless
were involved in very high levels of teacher interaction in a way that was
not true of girls.
If it is the case that there is not a general pattern whereby all or most
boys get slightly more attention than all or most girls, but rather a
pattern whereby a few boys get very much more attention than all other
pupils, then the gender imbalance is best seen as a problem of
classroom management rather than necessarily a problem of sexist
bias. Such a pattern would also explain both the overall similarity
between studies and the pattern of variation in results whereby in some
classrooms there is no sex bias and in others there is a bias towards
boys, but no studies report a bias towards girls. If the imbalance reflects
poor classroom management strategies or poor class control then this is
something which we should expect to vary considerably between
teachers and possibly between schools and types of schools. Such a
view also offers an explanation of the way that some qualitatively-
based studies claim very much greater levels of inequality than are
reported in any of the systematic observation studies. The qualitative
studies by their nature are focused on very few classrooms and may
chance upon classrooms where problems of management are greatest.
It is also likely that a few incidents involving very high levels of
198
attention to certain boys can come to dominate the researcher’s

perception of the classroom and disguise the similar levels of attention
given to most boys and most girls.
Other aspects of classroom behaviour

Teacher interaction is one of the most contentious areas for looking at
differences in the classroom experiences of boys and girls, but there are
other areas where differences may arise. The ORACLE study gives the
most detailed account of classroom behaviour and interactions of
primary-aged pupils and, as in the case of teacher interaction, differences
in other types of behaviour are small, although girls have slightly higher
total work involvement than boys (Galton et al. 1980). The study of
infant classrooms by Tizard et al. (1988) found very similar patterns of
classroom experience and, for example, the same level of time-on-task
for boys and girls. One difference which did emerge in this study was that
white boys in the sample spent more time engaged on mathematics tasks
and this seemed to reflect teacher allocation rather than pupil choice
(Tizard et al. 1988).
One aspect of classroom interaction which is apparent from the
ORACLE research and from other studies is the marked tendency for
girls to interact with other girls and boys to interact with other boys. In
the ORACLE study same-sex seating pairs were observed to interact
twice as frequently as opposite sex pairs. In mixed sex groups interaction
is twice as likely to be with a pupil of the same sex and the grouping
pattern with the highest levels of pupil-pupil interaction were same sex
groups. In all, over 80 per cent of pupil-pupil interactions were with
members of the same sex and this clearly arose, for the most part, from
pupil choice rather than from teacher grouping arrangements (Galton et
al. 1980). In respect of pupil interactions, gender differentiation in the
primary classroom is imposed by the pupils themselves through their
social behaviour. These results are in accord with those of researchers
such as Hartley (1985) who reports a strong tendency to same-sex
friendship choices in the primary school.
Sex differences in academic achievement

Perhaps the central issue in any consideration of the consequences of sex
differences and gender differentiation in the primary school is in the area
of educational achievement. Academic achievement, especially in the
areas of reading, language and mathematics, is central to the aims of
primary schoolteachers (Ashton et al. 1975) and predominates in the
content of the primary curriculum (Galton et al. 1980). If there are sex
role differences in primary schools and, in particular, if differentiation
199
works to the disadvantage of girls, differences in patterns of

achievement, if they occur, would be the most serious manifestation of
these differences.
An important source of evidence for levels of achievement in the
primary school is the work of the Assessment of Performance Unit
(APU). Surveys of achievement among 11-year-olds in language,
mathematics and science were conducted by the APU in the late 1970s
and early 1980s. The 1979 survey of language (APU 1981a) showed no
significant differences between the performance in reading of boys and
girls but found that girls were ahead in writing skills. A survey a year
later found that girls were ahead in both reading and writing. These
differences were statistically significant but in absolute terms were
small (APU 1982a). Surveys of mathematics achievement were made
in 1978, 1979 and 1980 (APU 1980,1981b, 1982b). In all of these
studies there were sex differences between the various subcategories of
mathematics achievement tested, with girls scoring higher in some and
boys in others. However, in all three surveys boys received higher
scores in the majority of the subcategories. The survey of achievement
in science made in 1981 showed no overall differences in performance
between girls and boys. However, boys performed better at measures of
science concepts while girls performed better at measures concerned
with representing information. The report suggests that boys were
performing better at the specifically scientific aspects of the
assessments (APU 1981c).
The research conducted in infant classrooms in London by Professor
Tizard and her colleagues at the Thomas Coram Research Centre
provides evidence on the achievements of a group of children at an early
stage in their primary school careers (Tizard et al. 1988). At the end of
the children’s time at nursery school, through the infant school, and in
the first year of the juniors, a variety of tests measuring reading, writing
and number skills were administered. The results at the end of the
nursery school show that at this stage there were few sex differences in
performance although the girls had better writing skills. In reading girls
and boys made about the same progress up to the middle infants but girls
made slightly more progress in their year in the top infants. In the first
year junior girls were ahead on reading. In writing girls maintained but
did not increase their lead over boys through the infant school. Writing
was not tested in the junior schools. In mathematics boys had made more
progress than girls by the top infants and were ahead in the first year
juniors.
Recent studies of the attainments of junior-aged children include the
ORACLE research (Galton and Croll 1980), the ILEA study
(Mortimore et al. 1988), and the One in Five study (Croll and Moses
1985). The ORACLE research showed an overall identical level of
200
performance by girls and boys. Within this overall figure was a slight
advantage to boys in the mathematics test and a slight advantage to
girls in the language test. Performance on the reading test was identical
for boys and girls (Galton and Croll 1980). The ILEA study followed
2000 children through their four years in the junior school. The test
results showed a general pattern of somewhat higher attainments for
girls. Girls had higher levels of attainment at each of the assessment
points for both reading and writing. However, the higher attainments in
each of these subjects at the end of the junior school years were no
more than would have been predicted from the girls’ higher level of
attainments at the beginning of junior school: progress, as opposed to
level of attainment, was the same for boys and girls. In the case of
mathematics there was little difference in attainment or progress
between boys and girls, although girls made more progress during the
third year of the junior school. No sex differences were found in
performance on tests of oral skills (Mortimore et al. 1988). The One in
Five study assessed both second year and fourth year juniors on tests of
reading and non-verbal reasoning. Girls were found to be slightly
ahead on both these measures (Croll and Moses 1988).
The APU surveys and the primary school research projects described
above have broadly comparable results to those of a much earlier study
of children’s attainments, the Medical Research Council longitudinal
study reported in Home and School (Douglas 1964). This research
replicated earlier results in indicating that girls are more successful than
boys at primary school. In this study girls performed better on tests of
verbal and non-verbal intelligence, tests of sentence completion and tests
of arithmetic than boys. The girl’s advantage was noticeably greater in
the sentence completion tests than in the arithmetic tests. The study
conducted during the time of selection for secondary education at 11+,
also showed that girls were more likely than boys to be selected for
grammar school education (Douglas 1964).
All of these studies are in broad agreement that girls perform at least
as well and generally slightly better than boys during the primary school
years. They also agree that the relative performance of girls and boys
differs across subjects and, to some extent across different aspects of a
subject. Girls consistently do better on language-related aspects and
relatively less well on mathematical aspects of assessments, although in
some of the studies they still do as well as or better than boys at
mathematics.
At the time of Douglas’s research, although girls were outperforming
boys at primary school and were more likely to be selected for grammar
school, far fewer girls than boys were successfully completing 16+ and
18+ examinations and continuing to higher education. The possibility
needs to be considered that even though the recent studies show girls
201
doing as well or better than boys at primary school they are, nevertheless,
in some way being prepared for failure later in their school careers and
their equality or advantage will disappear in the secondary school. The
most recent data relevant to secondary school performance of boys and
girls comes from the 1987 edition of the Department of Education and
Science statistics (DES 1987). The figures given here relate, for the most
part, to the academic year 1985–6 when children in the ORACLE study
and in some of the APU survey were entering post-compulsory education
These figures suggest that there is no longer an increasing
underachievement of girls as they pass through the school system. As an
indication of performance at 16+ the DES statistics give the proportion of
boys and girls leaving school with ‘O’ levels at grade C or above or
Grade 1 in GCSE (irrespective of any ‘A’ levels obtained). Of the girls,
58 per cent had at least one such grade compared with 52 per cent of
boys. This performance is also reflected in the way that a higher
proportion of girls than boys in the 16–18 age group remain in post-
compulsory education, 34.9 per cent compared with 30.5 per cent of
boys. Of those remaining in the school sector of post-compulsory
education there was again a slightly higher proportion of girls, 18.1 per
cent, compared with 17.2 per cent of boys (DES 1987).
A comparison of the ‘A’ level performances of boys and girls for the
1985–6 school year shows a very similar pattern in overall
performance. An identical proportion of girls and boys, 18.5 per cent,
left school with at least one ‘A’ level. Fractionally more boys than girls,
14.9 per cent compared with 14.3 per cent, left school with two or more
‘A’ levels. This equality is not sustained into higher education,
however. Among the 19 to 20-year-old age group, for example, 13.3
per cent of males were in higher education compared with 11.8 per cent
of females. And in 1985, 56 per cent of those receiving first degrees
were male (DES 1987).
Although the overall level of secondary school achievement is similar
for boys and girls, the achievements in different subjects differ
considerably. The direction of this difference reflects in an exaggerated
form differences which had begun to be apparent at primary school. At
16+ 46 per cent of girls received an ‘O’ level pass or CSE Grade 1 in
English compared with 35 per cent of boys. In contrast, 35 per cent of
boys achieved this result in mathematics compared with 30 per cent of
girls. In post-compulsory education this tendency has become much
stronger: 55 per cent of boys with two or more ‘A’ levels included
mathematics in their results compared with 32 per cent of girls, while 51
per cent of girls with two or more ‘A’ levels passed ‘A’ level English
compared with 27 per cent of boys (DES 1987).
As has been apparent from the measures of academic achievement
presented above, there is no straightforward way in which girls can be
202
regarded as academically disadvantaged in the primary school. Their

overall performance is at least as high as that of boys and in most studies
is rather higher. Measures of achievement and participation rates in
secondary school also show that girls achieve equally with boys. It is not
until entry into higher education that the disadvantages women suffer in
occupations and in the wider society appear in their educational
experiences. This makes it unlikely that the primary school can be
regarded as in some way preparing females to fail despite their apparent
equal or higher achievement compared with males.
The question of subject bias in achievement is more complex. It is
plain that girls’ higher levels of achievement manifest themselves most
clearly in aspects of achievement related to language and reading. In
mathematics and science girls are generally either not ahead of boys or
performing slightly less well. This polarization of achievement is only
very marginal at primary level, especially with regard to girls’
underachievement. At secondary school, however, the polarization of
achievement and subject choice between linguistic/humanities-
orientated subjects and mathematical/scientific subjects is very
apparent. The consequences of this polarization are more difficult to
establish. At primary level reading achievement is the single most
important academic aim for teachers (Ashton et al. 1975). Reading is
central to primary classrooms as a medium for instruction and a criteria
for assessment, and dominates the judgements teachers make of
children’s academic ability (Croll and Moses 1985). At this level it is
certainly not a less highly-valued skill than scientific and mathematical
achievements.
Special educational needs

Although the difference between the average attainment of boys and
girls in the primary school is relatively small, it is boys who are far
more likely than girls to have special educational needs. Several
studies have found that, over the whole range of special needs, boys
outnumber girls by approximately two to one. This was the conclusion
reached by the epidemiological studies of Rutter et al. (1970) and
Pringle et al. 1966).
More recently, in a study of teachers’ views of special educational
needs, Croll and Moses (1985) found that twice as many boys as girls
were regarded as having special educational needs by their class
teachers. The research showed that, overall, teachers of junior-aged
pupils thought that 24.4 per cent of boys in their classes had special
educational needs but only 13.2 per cent of the girls. When various
kinds of educational needs are considered interesting differences
emerge. Learning difficulties were attributed to 19.5 per cent of the
203
boys and 11.1 per cent of the girls and behavioural problems to 10.9 per
cent of boys and 4.5 per cent of girls. Although, in total, relatively few
children were regarded as being discipline problems, boys
outnumbered the girls by nearly four to one (boys 6.4 per cent and girls
1.7 per cent). These boys were not simply indulging in boisterous
‘boylike’ behaviour that was a nuisance to the teacher but expected and
accepted. Virtually all the boys who were regarded as discipline
problems were also thought to have emotional and behavioural
difficulties which were creating problems for the child as well as being
a nuisance to the teacher. Behaviour problems were also associated
with learning difficulties, and many more boys than girls experienced
this complex of special needs.
As we have seen, there are only small differences between the
average scores of boys and girls on reading tests but there are many
more boys than girls who experience problems with reading in the
primary school. Croll and Moses (1985) compared the reading ages of
boys and girls in 34 second year junior classrooms and found that
nearly twice as many boys as girls had a reading age two years or more
behind their chronological age (6.7 per cent of boys and 3.6 per cent of
girls). The proportion of boys scoring between one and two years
behind their real ages was half as great again as the proportion of girls
(17.6 per cent of boys and 11.8 per cent of girls). These figures are a
clear indication that more boys have reading difficulties, and it is not
surprising that the 428 junior class teachers interviewed by Croll and
Moses described 9.5 per cent of the girls in their classes as ‘poor
readers’ compared with 17.5 per cent of the boys. However, a
comparison of test scores with teacher assessments suggested that,
when performance is taken into account, teachers tended to
underestimate the extent to which girls had reading difficulties
compared with their estimation of the reading difficulties of boys.
Previous discussions of the relationship between teachers’ views of
their pupils and the pupils’ actual performance have tended to
concentrate on the possible advantages accruing to children who are
favourably regarded by their teachers and the disadvantages of being less
well regarded. Concern with the effect of teacher expectations and with
‘self-fulfilling prophecies’ has led researchers to concentrate on the
positive aspects of being favourably regarded by teachers and the
possible unfairness this involves to the other children in the class. If we
take this view then the advantage is overwhelmingly in favour of the
girls. Nevertheless, there is little solid evidence for such processes (see
Boydell 1978 and Croll 1981 for a review of some of the relevant
evidence). It is, at least on the face of it, at least as plausible that a child
may be disadvantaged by not having her or his difficulties recognized as
that such a recognition will become self-fulfilling.
204
Attitudes to school among girls and boys

There is evidence that girls express more positive attitudes towards
school than do boys. The ORACLE study found that girls scored higher
than boys on a measure of contentment at school and on intrinsic
motivation to do well at school (Croll and Willcocks 1980) and the ILEA
research reported that girls had higher scores on a scale of attitude
towards school and also had more positive self-concepts in relation to
school (Mortimore et al. 1988). However, there is also evidence that girls
have higher levels of anxiety (Croll and Willcocks 1980). Tizard et al.
(1988) report that among infant school children boys are more likely than
girls to rate themselves as above average on academic tasks although
their performance did not justify this. Evidence on aspects of pupil
attitudes with regard to particular curriculum areas can be found in the
Assessment of Performance Unit (APU) surveys. In mathematics there
was no difference in expressed levels of enjoyment between boys and
girls, a result also reported in the ILEA study (ILEA 1988). There were
also no differences in their perceptions of the usefulness of mathematics.
However, boys had more confidence in their own mathematical ability
than did girls (APU 1980). In the case of science boys were more positive
than girls about the experimental aspects of science and rated these more
highly than the non-experimental aspects in a way that was not true of
girls (APU 1981c).
It seems that in their overall view of school girls are rather more
positive and motivated than boys. However, there is some tendency for
girls to be more anxious and for boys to be more self-confident about
school work.
Teacher perceptions of girls and of boys

There is ample evidence that teachers perceive boys’ classroom
behaviour as being more problematic than that of girls. When
considering special educational needs we saw that teachers were much
more likely to describe boys as having behaviour problems and as posing
discipline problems in the classroom. Earlier studies such as that of
Douglas (1964) also report teachers as being much more critical of boys
than of girls. The ILEA research and the London infant school survey
also showed that boys were more likely than girls to be described as
having behavioural difficulties (Mortimore et al. 1988; Tizard et al.
1988). In his detailed case study of a primary school Hartley (1985)
reports that teachers have more positive views of girls and regard boys as
noisy and disruptive. A similar study in infant classrooms by King (1978)
reports that boys were reproved more and that teachers had more
favourable assessments of girls’ behaviour.
205
It is less clear if these perceptions have any wider consequences for

the classroom experiences of boys and girls. Boys are more likely to be
seen by teachers as having learning difficulties than girls (see above), but
this still only applies to a minority of boys. However, Douglas shows that
at the time of the 11+ examination teachers were more likely to think that
girls rather than boys should be given grammar school places, even
though there was little difference in their test scores (Douglas, 1964).
The ILEA research reports that primary records are more favourable to
girls than to boys (Mortimore et al. 1988), and King (1978) found that in
the infant classrooms he studied boys were typified by teachers as
making less progress than girls.
For the most part these differences in teacher perceptions almost
certainly reflect real differences in pupil behaviour and attainment. The
poorer behaviour of boys and their greater likelihood of having learning
difficulties are mirrored in their teachers’ views. It should also be noted
that, as King (1978) points out, the most important typification teachers
make of pupils is as individuals. Although King describes the different
perceptions teachers had of boys and of girls, he emphasizes that such
generalizations were of limited importance compared with the teachers’
stress on each child’s individuality.
Discussion
The studies discussed above have pointed to aspects of both difference
and similarity in the educational experiences of girls and boys at primary
level. In many respects sex roles are highly differentiated in the primary
classroom. Children are differentiated by their names, their clothes and
some of the activities they engage in. Gender groupings are also an
important aspect of the way classrooms are managed and most of the
children’s own social and work interactions are with pupils of the same
sex. However, in other respects boys’ and girls’ educational experiences
are very similar: they have broadly similar levels of academic
achievement, broadly similar attitudes towards school and broadly
similar levels of interaction with the teacher. Where differences emerge
in these areas the extent to which they can be regarded as disadvantaging
either girls or boys varies. Boys get rather more individual attention from
the teacher, are slightly more confident and sometimes perform better at
mathematics; girls like school more and are more motivated, perform
slightly better, especially in the area of language, and are considerably
more favourably regarded by their teachers. However, in all these
respects differences within gender groups are very much more important
than differences between them.
The research reviewed here cannot be regarded as supporting the
thesis that gender differentiation in the primary school contributes in any
206
substantial fashion to the disadvantages females suffer in the outside

world and, in particular, in the world of employment. Not only do girls
succeed in the primary school, they continue to perform equally with
boys throughout secondary schooling. If the primary school is preparing
them for failure, it takes a long time for this effect to make itself felt. It
has sometimes been argued that it is the failure of girls to choose and to
do well in mathematical and scientific subjects which in part accounts for
their disadvantaged occupational position (for example, Deem 1978).
The beginnings of this process can be identified in some of the studies of
primary schools although the underachievement of girls, even where it
occurs, is small. What is ignored in this argument however, is that boys
who study the subjects in which girls excel are not disadvantaged in
employment.1
A further issue to be considered is whether schools go beyond gender
differentiation in society as a whole and impose a more rigid division into
traditional sex roles. Sara Delamont argues that schools are essentially
conservative institutions in which prevailing sex stereotypes are not only
not challenged but are exaggerated (Delamont 1980, 1983). This is an
important and carefully stated argument, but while we accept that
schools do not in general challenge prevailing societal assumptions about
sex roles and gender divisions (Acker 1988), the extent to which they can
be seen as exaggerating them is far less clear. Examples of exaggerated
gender divisions include rules about clothing, separate registers, separate
games, and so on. Other examples, however, could include equal levels
of achievement, equal (or higher) representation in high achieving
streams, equal access at the primary level to virtually all of the
curriculum and all its central aspects, and equal regard for their
achievements by teachers. Although these features seem so obviously
right that they are not worth commenting on, it should be noted that such
equality does not generally prevail in society. These are also the aspects
of school which are central to their purpose. It is in these areas where
schools are most distinctive that they are also more equal than society at
large. Although girls are differentiated at primary school the evidence
suggests that they are not disadvantaged.
Note
1 It is commonly suggested that the disadvantages women experience
in employment can be attributed to the fact that they are less likely
than men to have studied mathematical and scientific subjects at
school and in higher education. Deem writes:
Once these patterns of subject choice are established, public

examination entries and pass rates reflect them, and may serve
207
to stop girls from ever going beyond the narrow occupational

confines of arts disciplines whether they enter a job
immediately after leaving or go on to higher or further
education first.
(Deem 1978:71, emphasis added)
As we have seen above, a third of girls taking ‘A’ levels pass

maths compared with just over half of the boys, so the subject
differences are not so great as the above quotation suggests.
More importantly, however, there is no reason to suppose that
the large number of males taking arts ‘A’ levels and graduating
in arts disciplines are occupationally disadvantaged compared
with those from science and mathematical backgrounds. Elite
positions in British society are largely dominated by men from
non-scientific and technical backgrounds. This is true of
government, both Parliament and the civil service, financial
institutions, boardrooms and the media. Even in industry there
are frequent complaints that senior positions are not held by
technical experts. If scientific and mathematical training is not
an advantage to men, why should its lack be disadvantaging to
women?
References
Acker, S. (1984) ‘Sociology, gender and education’ in S.Acker et al. (eds) World
Yearbook of Education: Women and Education, London: Kogan Page.
Acker, S. (1988) ‘Teachers, gender and resistance’, British Journal of Sociology
of Education 9 (3): 307–22.
Ashton, P., Kneen, P., Davis, F. and Holley, B.J. (1975) The Aims of Primary
Education, London: Macmillan.
Assessment of Performance Unit (APU) (1980) Mathematical Development:
Primary Survey Report 1, London: HMSO.
Assessment of Performance Unit (APU) (1981a) Language Performance in
Schools: Primary Survey Report 1, London: HMSO.
Assessment of Performance Unit (APU) (1981b) Mathematical Development:
Assessment of Performance Unit (APU) (1981c) Science in School Age 11,
London: HMSO.
Assessment of Performance Unit (APU) (1982a) Language Performance in
School: Primary Survey Report 2, London: HMSO.
Assessment of Performance Unit (APU) (1982b) Mathematical Development:
Boydell, D. (1978) The Primary Teacher in Action, London: Open Books.
Buswell, C. (1981) ‘Sexism in school routines and classroom practices’, Durham
and Newcastle Research Review, 9: 195–200.
Croll, P. (1981) ‘Social class, pupil achievement and classroom interaction’, in
B.Simon and J.Willcocks (eds) Research and Practice in the Primary
Classroom, London: Routledge & Kegan Paul.
208
Croll, P. (1986) Systematic Classroom Observation, Lewes: Falmer Press.

Croll, P. and Moses, D. (1985) One in Five: The Assessment and Incidence of
Special Educational Needs, London: Routledge & Kegan Paul.
Croll, P. and Willcocks, J. (1980) ‘Personality and classroom behaviour’, in M.
Galton and B.Simon (eds) Progress and Performance in the Primary
Classroom, London: Routledge & Kegan Paul.
Deem, R. (1978) Women and Schooling, London: Routledge & Kegan Paul.
Delamont, S. (1980) Sex Roles and the School, London: Methuen.
Delamont, S. (1983) ‘The conservative school? Sex roles at home, at work and at
school’, in S.Walker and L.Barton Gender Class and Education, Lewes:
Falmer Press.
Department of Education and Science (DES) (1987) Statistics of Education 1987,
London: HMSO.
Douglas, J.W.B. (1964) The Home and the School, London: MacGibbon &
Kee.
Equal Opportunities Commission (1987) Women and Men in Britain: A Statistical
Profile, London: HMSO.
French, J. and French, P. (1984) ‘Gender imbalances in the primary classroom: an
interactional account’, Educational Research, 26 (2): 127–36.
Galton, M. and Croll, P. (1980) ‘Pupil achievement and progress’, in M.Galton
and B.Simon (eds) Progress and Performance in the Primary Classroom,
Galton, M., Simon, B. and Croll, P. (1980) Inside the Primary Classroom,
Inner London Education Authority (ILEA) (1988) The Junior School Project, vols
1–3, London: ILEA.
Hartley, D. (1985) Understanding the Primary School, London: Croom Helm.
Kelly, A. (1986) ‘A method to the madness? Quantitative research reviewing’,
Research in Education, 35: 25–41.
Kelly, A. (1988) ‘Gender differences in teacher-pupil interactions: a metaanalytic
review’, Research in Education, 39: 1–23.
King, R. (1978) All Things Bright and Beautiful? Chichester: Wiley.
Megarry, J. (1981) Sex, Gender and Education, Glasgow: Jordanhill College of
Education.
National Foundation of Educational Research (NFER) (1987) Teaching Styles
and Pupil Performance at the Primary Level, NFER mimeo.
Rinehart & Winston.
Pringle, M.L.K., Butler, N. and Davie, R. (1966) Eleven Thousand Seven Year
Olds, London: Longman.
Rutter, M., Tizard, J. and Whitmore, K. (1970) Education, Health and Behaviour,
London: Longman.
Serbin, L. (1983) ‘The hidden curriculum: academic consequences of teacher
expectations’, in M.Marland (ed.) Sex Differentiation and Schooling, London:
Heineman.
Sharp, R. and Green, A. (1975) Education and Social Control, London:
Spender, D. (1982) Invisible Women: the Schooling Scandal, London: Writers and
Readers Publishing Cooperative.
209
Stanworth, M. (1981) Gender and Schooling: a Study of Sexual Divisions in the

Classroom, London: Women’s Research and Resources Centre.
Tizard, B., Blatchford, P., Burke, J., Farquhar, C. and Plewis, I. (1988) Young
Children at School in the Inner City, Hove and London: Lawrence Erlbaum
Associates.
210
Chapter twelve
Interaction with children with special

educational needs
David Galloway
Pupils with special needs in the primary classroom are a major concern
of teachers and the subject draws upon many of the themes of this
volume. Estimates range from 20 to 40 per cent of pupils presenting
some form of special need during their schooling career. A very
sizeable minority of school children, most of whom have special
needs, will be identified and given attention in the classroom.
Galloway, who has undertaken much research in this area, places
special needs in a decidedly social context. Teachers make judgements
of special needs through social comparisons with ‘normal’ pupils. A
vast majority of pupils labelled as having special needs are found in
ordinary classrooms and will form an understanding of themselves
through the social context in which they interact and the quality of
learning (or curricular) treatment that they are offered. Teachers are
central to the development of all children in their charge though the
socio-emotional climate created in the classroom, the relationships
that are structured for learning, and the interpretation of curriculum
assigned to each child.
Themes that have arisen in this volume and which are discussed in
greater detail here include: the creation of individual identities and
self-efficacy through the interactions and social comparisons in
which the special needs child participates; the understanding that the
presentation of the curriculum to the child provides an orientation to
bring together or divide the class; friendships based on perceptions of
similarity (and legitimated by the teacher) may not allow for the
integration of special needs children unless active efforts are made to
enhance children’s social skills and ability to work with others; the
central role of teachers in structuring the social and curricular
experiences of their charges.
There is much ‘food for thought’ in this chapter as it relates to special
needs and ‘normal’ classroom experience. The chapter is strongly based
in past and current legislation. It offers terms for practical consideration
211
David Galloway
that includes moral climate, co-operative groupings and individual

differences. The chapter describes happenings in classrooms where
children with special needs are integrated without their social needs
being catered for, and directs the reader to a number of alternative
processes through which the teacher can make their integration more
successful.
Introduction
According to the Warnock Committee, children with learning difficulties
would need either special means of access to the curriculum, a special or
modified curriculum, or ‘special attention to the social structure or
emotional climate in which learning takes place’ (DES 1978a). With the
dubious wisdom of hindsight, the problem here is not that this distinction
is inappropriate, but rather that it failed to address some of the most
pressing issues in provision for special educational needs. The Warnock
Committee is widely believed by teachers to have encouraged integration
of children with special educational needs into the mainstream
classroom. The arguments for integration are at least as concerned with
opportunities for children with special needs to interact socially with
their peers in the mainstream, as with opportunities for them to
experience the mainstream curriculum.
Historically, provision for children with special needs had been based
on special or modified curricula. This was provided either in full-time
special schools or classes, or in the withdrawal groups and remedial
classes of mainstream schools. More than ten years after the Warnock
Report it is not clear how far this has changed. In a majority of local
education authorities (LEAs) withdrawal groups for extra help with
reading and numeracy are the most prevalent means of helping primary
children with learning difficulties. These frequently consist of activities
which have only an indirect relationship to work the children have been
doing in their regular class. Thus backward readers are more likely to
find themselves doing new and different activities than to receive help
which gives them a feeling of competence in classroom tasks on which
they are currently failing.
This chapter is concerned with social interactions among primary
children for whom access to social and curricular experiences of the
ordinary class is a realistic goal. In practice, this refers to a vast majority
of the 20 per cent of children considered by Warnock to require some
form of special educational treatment at some stage in their school
careers. This group includes children with physical or sensory
disabilities not associated with severe learning difficulties. It also
includes children currently placed in schools for pupils with moderate
learning difficulties, for whom the educational benefits of integration
212
Children with special needs
have been extensively documented (for example, the review in Galloway

and Goodwin 1987). Not least it also includes those children defined as
disruptive or having emotional or behavioural difficulties. In passing it is
tempting to ask whether as many as 20 per cent of school leavers would
not have benefitted at some stage in their school careers from ‘special
attention to the social structure or emotional climate’ in which learning
took place. It is hard to see how the Warnock Committee could have
regarded this as other than a thinly-veiled argument for the removal of
children whose behaviour teachers found difficult or disturbing. Yet the
evidence from school effectiveness research suggests clearly and
consistently, that interactions within the school play the major part in the
origin of such problems (see Galloway et al. 1989).
My argument, then, will be that many of the social problems created
and experienced by children with special educational needs derive from
lack of effective access to the mainstream curriculum. I shall start by
considering the purpose of integration as regards children’s social
development and will then identify some of the conflicting pressures
which may reduce the quality of social interaction for children with
special needs. Finally, I will argue that teachers’ understanding of social
interaction in their classes is illuminated by their understanding of
individual differences.
The purpose of integration

Internationally, the movement for integration, or mainstreaming, seems
to have acquired its own momentum in which ideology and personal
values are as important as empirical evidence. The problem with
movements that acquire their own momentum is that their goal comes to
be seen as an end in itself rather than a means to an end. There would be
no point in integration if the quality of social and learning experiences
was not developmentally more appropriate for the children concerned
than alternative provision in a special school. Hence we must consider
what social experiences we are trying to provide. A starting point is to ask
how learning in an ordinary class can benefit the social development of
children with special needs.
In most classes social development probably depends largely on
vicarious learning, by enhancing opportunities for modelling age-
appropriate behaviour, or by increasing the frequency of chance
interactions. The quality and range of a child’s social skills depend
largely on the stimulation provided (Strain 1984). The less stimulation
provided, the more limited will be the children’s social skills. I recently
saw an example of this in a special school for children with very severe
physical and intellectual impairments. The children were seated on three
sides of a large table. The teacher and two welfare assistants showed each
213
David Galloway
child individually the various stages of mixing the ingredients for bacon
omelette. When talked to directly each child responded to the adult,
albeit in a limited way. Yet for over 85 per cent of the lesson each child
was sitting on his or her own. The adults were occupied with other
children and no attempt was made to facilitate social interactions
between the children, if only at the simple level of ‘look what Simon is
doing’. For most of the lesson the children’s opportunities for social
interaction were virtually non-existent.
This incident took place in a special school class that had previously
been held up as an example of good practice by the head teacher and by
a college tutor. It illustrated vividly the importance of planning social
interactions for children with special needs. Yet while the importance of
planning is perhaps most evident in special school settings, there can be
no justification for assuming that similar problems do not exist for
children with special needs in the mainstream or, for that matter, for all
other children.
The ‘opportunities’ for social interaction with non-handicapped peers
may be illusory if the ethos of the mainstream school and classroom does
not facilitate co-operative activities. As the Joint Council for the
Education of Handicapped Children had argued before the Warnock
Report was published, isolation in a large community can be far worse
than the loneliness of the small community. When special schools
operate as small self-contained communities children can easily feel cut
off from their mainstream peers, but the effects of this loneliness can be
mitigated by the support provided within the community. If adequate
attention is not paid to the curriculum and social climate, children in the
mainstream may feel equally cut off, without the mitigating support. This
may explain why some, though not all, evaluation studies have found that
children with moderate learning difficulties appear to be socially better
adjusted in special schools than in the mainstream.
Leaving to chance or osmosis the social development of children with
special needs, may not only be disastrous for the children concerned but
also poor social education for their peers in the mainstream. Hence, if
modelling is to facilitate the social development of children with special
needs, much will depend on the kind of model to which they are exposed.
In other words, there may be a need not only for a more actively
interventionist approach towards the children themselves, but also for
attention, as Warnock put it, to the underlying social structure and
emotional climate of the (mainstream) classroom. Indeed, the presence
of children with special needs may actually enable teachers to identify
tensions in this area. If every child’s progress and social adjustment
appeared satisfactory, there would be little incentive to review current
practice. Thus, if the jerk-jerk chatter inanities of Dick and Dora had
captured all children’s imagination, providing the background to a self-
214
reinforcing love of reading, teachers would have had little incentive to

seek more imaginative and stimulating materials.
Even when integration appears to be facilitating the social
adjustment of a child with special needs, a closer look at the child’s
interactions with peers may be revealing. Guralnick (1981) has argued
that when allowed a free choice children tend to choose friends from
other children who have similar cognitive ability to themselves, are
high in social status in their own peer group, and adopt a control-and-
demand orientation, behaving towards their developmentally-delayed
peers as though they were adults and their peers with special needs
were children. This may foster leadership qualities in the children with
high social status, and may also have some educational value for
children with special needs. It may be unlikely, though, to encourage
them to develop the skills needed for active participation in their peer
groups. By implication, a structured programme may be needed if
children with special needs are to acquire the social skills needed for
participation with mainstream peers.
Whether this kind of relationship develops between high status
children and peers with emotional and behavioural difficulties or with
mild learning difficulties is unclear. It seems improbable that they will
interact as equals. Whether the relationship is characterized by control-
and-demand orientation or by a helping, co-operative one may depend
less on the children themselves than on the ‘moral climate’, or ethos
which the teacher has established in the classroom. Nevertheless, it is
clear that children’s consciousness of social status increases with age,
and is likely to have an increasingly powerful influence on how children
with special needs perceive their own position in their peer group.
On an INSET course, a group of 70 teachers took less that five
minutes to produce a list of more than 70 derogatory labels in current use
in their own schools when pupils were referring to peers regarded as slow
learners (for example, ‘divvy’, ‘remmo’, ‘spas’). This problem may be
less acute and less pervasive in primary schools, but interactions at this
stage may well provide the basis for more destructive labelling later.
Even in the infant school children are sensitive to their teacher’s attempts
to group them by ability, recognizing promotion or demotion to a higher,
or lower table. Many parents can recall their infant school children
talking about being moved ‘up’ or ‘down’ a table.
While it is clear that the social development of children with special
needs may at times be the object of direct intervention in the mainstream
classroom, the limitations of such intervention should be acknowledged.
Strain (1984) has demonstrated the value of social skills training with
nursery school children with autistic symptoms, but his conclusion may
not hold true for older children nor for the less handicapped children with
whom this chapter is concerned. The teacher’s task is to create a climate
215
David Galloway
which accepts and respects individual differences. At times this will

require explicit attention to the social problems of individual children,
but the impact of any individualized programme is likely to be limited.
There are four problems:
1. Much of the rhetoric of special needs has focused on identifying and

meeting individual needs. In the United States, for example, Public
Law 94–142 required Individual Educational Programmes for
handicapped children as well as introducing into Federal Legislation
the concept of the ‘least restrictive alternative’. In Britain, the 1981
Education Act provides an elaborate machinery for assessing
individual needs. This individualized focus may have done teachers
and children a disservice. Teachers may legitimately feel that
‘experts’ have failed to recognized the practical difficulties in
implementing individual programmes in a class consisting of thirty
individuals. Both on practical and theoretical grounds, the problem
lies in finding ways to meet individual needs by means of group
activities.
2. Quite apart from the practical problems in implementing
individualized programmes in a mainstream classroom, there is the
theoretical point that children’s—and adult’s—sense of personal
identity develops in a social context (Hargreaves 1983). After the
family, the principle influence is that of the school. Programmes
which focus on the individual as an individual rather than as a
contributing member of a group may inadvertently restrict the
child’s personal development. This is particularly true in the
mainstream class where children with special needs are more viable
and more vulnerable than in a separate special class.
3. Developing the last point, the hidden curriculum is likely to exert a
more profound as well as a more subtle influence on children’s
social development than any teacher-directed programme. If the
hidden curriculum fosters a climate in which all children expect, and
are expected, to contribute to the work of the class, then the need for
programmes focusing explicitly on the social development of
individual children will be greatly reduced. If the hidden curriculum
does not foster such a climate, then an individualized programme
will be operating in opposition to tensions in the social structure of
the school or classroom.
4. Even at the pre-school stage, children with moderate learning
difficulties benefit from opportunities for observational learning in
mainstream classes. As they progress through primary school, they
become increasingly sensitive to the negative implications inherent
in individual programmes, namely that the aim is to remedy their
defects. Their peers also become sensitive to the negative
216
implications, and the effect is seen in the derogatory labels referred

to earlier.
Conflicting pressures on teachers and children
National developments
Until recently, education in Britain was seen as a three-way partnership
between the Department of Education and Science (DES), LEAs and
schools. The 1988 Act, however, accelerated a trend evident in the earlier
Acts of 1980 and 1986, shifting power away from teachers and LEAs to
central government and, to a lesser extent, parents. These changes seem
likely to exert a far-reaching influence on the social climate in schools as
well as on the educational climate. The 1988 Act attracted opposition
from virtually all professional groups involved in education (Haviland
1988), and particularly from special educators. It is not, however, clear
that their fears are justified. In this section I shall address one of the most
controversial aspects of the Act, the National Curriculum and its
associated attainment targets for particular age groups.
The desirability of a National Curriculum was seriously contested
only by professional educationalists. Fears were expressed that it would
lead to a narrowing of the curriculum for children with special needs (for
example, Brennan 1987). There were also fears that the imposition of
attainment targets would encourage not only a narrowly academic
pedagogy, but also attempts by head teachers to gain exemption from the
National Curriculum’s requirements for children whose learning or
behavioural difficulties threatened the progress of stability of the rest of
the class. Further, publication of the results of national testing would, it
was alleged, damage the self-esteem of lower attaining pupils.
These are not groundless fears but they do conveniently overlook
the damaging consequences of the curricular arrangements replaced by
the 1988 Act. In particular, the lack of any agreed National Curriculum
and of associated attainment targets placed enormous responsibility on
teachers’ professional judgement. It led to the widespread adoption of
alternative, low-status curricula, for ‘non-examination’ classes in
secondary schools. The social consequences were at least as severe as
the educational. Individually, most teachers were catering as
effectively as they could for their academically less able pupils, but
they were constrained by a system which promoted the needs of an
academic elite. The 1988 Act does at least require that all pupils should
have access to the National Curriculum except in specified, and
restricted, circumstances. Potentially, this could encourage head
teachers and governors to take the needs of their ‘special’ pupils more
217
David Galloway
seriously than before. Yet it may also encourage them to refer pupils for
formal assessment under the 1981 Act, not in order to obtain a clearer
picture of their needs, but in order to gain their exemption from the
requirements of the National Curriculum. More seriously, there are
already signs that the National Curriculum itself may legitimize
differential curricula based on perceived pupil ability. This concern
comes from the Science Working Party’s decision to accept pressure
from the Secretary of State to recommend a two-tier curriculum in
which an elite of pupils will receive more science teaching than the rest
(DES 1988a). The fact that this results from the national shortage of
science teachers should not obscure the likelihood that it will
perpetuate a second-class curriculum for second-class pupils, at least in
the eyes of the pupils themselves, their parents and many of the
teachers who teach both groups.
Whether the National Curriculum will, as some critics maintain,
encourage a formal teaching style with a narrow focus on attainment
targets and minimal interaction between pupils, will depend on how the
various subject working groups define their task. They are, however,
under pressure for a narrow and formal curriculum from the Secretary of
State, as illustrated by his demand for more grammar following the
report of the English Working Group (DES 1988b). Nevertheless, the
trend throughout the 1980s has consistently been towards a more
interactive relationship between pupils and between pupils and teachers.
In the primary sector this was recommended as far back as the Plowden
Report (1967). In the secondary sector it is evident in the group work
required in many GCSE curricula, pupil involvement in records of
achievement, and the range of developments in 14–19 education.
In the primary sector the impact of the policy changes has not always
been clear. Superficially, the insistence on teachers developing subject
specialities, first raised in the Her Majesty’s Inspectors (HMI) Primary
Survey (DES 1978b), maintained in successive DES circulars (for
example, DES 1984) and reflected in the subject divisions of the
National Curriculum, strikes at the heart of the post-Plowden primary
ethos with its emphasis on teaching the ‘whole child’ through theme,
topic and project work. Research, however, suggests that this ethos has
never been reflected in what actually happens in the majority of primary
school classrooms. Rather, as HMI have repeatedly insisted, the evidence
suggests that class teachers have great difficulty in covering the whole
range of the curriculum effectively. Moreover, if primary teachers are to
have full-time responsibility for a class, then it is essential for the school
to contain a member of staff with expertise in each area. The focus on
subject specialisms may well lead to a defensive retreat into formal
subject teaching in some schools, but this is not inevitable. One could
equally well argue that it should facilitate rather than impede the
218
development of cross-curricular approaches and teaching methods that

encourage pupils’ active participation. These are presumably the trends
that HMI is seeking to encourage, since lack of pupils’ involvement in
their own learning is one of their most frequent criticisms of primary
schools, as of secondary. A more realistic assessment, though, may be
that the National Curriculum and associated pressures to develop subject
specialisms will polarize the profession. A minority of schools will
develop further their existing interactive approaches, while others,
perhaps a majority, will feel justified in resorting to ever-increasing
formality. The children who suffer most from the passivity induced by
increasing formality will be those with special needs. Schools which
resort to this kind of strategy may find themselves caught up in a vicious
circle in which pupil behaviour deteriorates as the curriculum and
pedagogy becomes more restrictive.
Influences on classroom behaviour

It is important to see the impact of recent developments dictated by
central government and/or by HMI in the context of other research on
children’s learning. Doyle (1985) identified two principal tasks for class
teachers: keeping order and facilitating learning. In practice these are
interlinked. An unprepared lesson encourages a range of disruptive
behaviours, and careful preparation is of little avail if the teacher lacks
basic classroom management skills.
The interlinked tasks of keeping order and facilitating learning are
themselves subject to developmental influences which deserve more
attention than they have so far received. It seems probable that pupils’
perception of teachers shifts as they progress through the stages of
compulsory education from that of facilitator to that of disciplinarian.
Kutnick (1983) observed that junior school teachers were much more
controlling than infant teachers, and this was associated with a concom-
itant extension of children’s understanding of teachers. Yet although
children become increasingly likely to see their teachers in disciplinary
roles as they get older, it does not follow that they themselves become
less amenable or co-operative. In most secondary schools a minority of
teachers rely on excessive note-taking, ‘chalk and talk’ or ‘death by a
thousand work-sheets’. Most secondary pupils acquiesce in the tedium
induced by these approaches for most of the time. In junior schools they
would lead almost invariably to open disruption and in infant schools to
pandemonium. In other words, the infant teacher’s survival depends on
active teacher-pupil and pupil-pupil interaction to an extent which,
perhaps regrettably, is not true in secondary schools.
Primary and secondary school teachers are, nevertheless, subject to
many of the same pressures and many of them adopt similarly self-
defeating strategies. In concluding that up to 20 per cent of children
219
David Galloway
would require some form of special educational treatment at some stage

in their school career, the Warnock Committee drew on research on the
epidemiology of learning and behavioural difficulties. The research on
learning difficulties used tests of cognitive and educational attainment
which were designed to ensure that responses from a large sample would
fit the normal distribution. The research on behavioural difficulties used
screening instruments in which items were selected at least partly on the
basis that they would be familiar to teachers, thus ensuring that a
substantial minority could be identified as ‘deviant or ‘maladjusted’. The
problem does not lie in the research itself which provided a much-needed
overview of the epidemiology of learning and behavioural difficulties,
but rather in the way it has been used to encourage too narrow a focus on
the alleged needs of individual children.
There is no great difficulty in constructing clinical instruments which
provide pseudo-scientific ‘proof’ of what a teacher already knows, for
example, that a hostile child is high on the hostility scale of Stott’s (1971)
Bristol Social Adjustment Guide. The objection lies in the implication
that the problem is in the child rather than in interactions which occur in
a particular social context. There is no great difficulty in finding research
evidence in support of this view. If the problem lay principally in the
child one would not expect to find major differences in pupil behaviour
between schools, provided one controlled for pupil intake, or between
classes in one school. Yet recent school effectiveness research
demonstrates that schools vary widely in the behaviour of their pupils,
and there is a similarly wide variation between classes within a school
(Mortimore et al. 1988). Further, the differences are not attributable only
to variations in the pupils’ social and educational backgrounds. Within
an entirely different research tradition, Berger (1982) has drawn
attention to the problem of transfer in classroom behaviour modification
studies. The evidence suggests that behaviours acquired in one context
will not transfer to other contexts unless transfer is explicitly built into
the programme. In other words, classroom behaviour is context specific.
This brings us back to Hargreaves’ (1983) point that personal identity
is established in a social context. The study of social interaction in the
classroom reveals ways in which children obtain recognition, acceptance
and friendship. One can do this within a behaviourist paradigm, though
many behaviourists might regard these as unacceptably nebulous
contacts. Thus analysis of the behaviour of a child regarded as the ‘class
clown’ by his teacher may demonstrate that the child consistently
receives reinforcement from the teacher and/or his peers for the
unwanted behaviours. Other paradigms lead to equally plausible and not
dissimilar explanations. Sociologists and social psychologists might
invoke the concept of secondary deviance to describe the same situation
(for example, Hargreaves et al. 1975), arguing that the clown’s status in
220
the classroom had changed as a result of the teacher’s irritation early in

the school year.
It is one thing for social psychologists to study interaction at the
‘micro’ level of the classroom, but quite another to show how interaction
at this level is influenced by changes at ‘macro’ level in the state
education system. Yet this is arguably the more interesting and more
important question. The nature and quality of social interaction will
reflect the ethos or moral climate of the school and of the class (Pring
1984). The moral climate refers to the climate of social expectations that
influence relationships between pupils, between a teacher and pupils
and, outside the classroom, between teachers. It is also reflected in the
teacher’s order of priorities and in the value ascribed to the achievements
of particular pupils or groups of pupils.
The ethos or moral climate of a school is, however, as much a
political construct as an educational one. One person’s ideal ‘moral
climate’ will emphasize attributes such as obedience, conformity and
competition between pupils. Another person’s will emphasize the
development of pupil autonomy, co-operation and independent critical
questioning of school and societal policies (the range is from
Summerhill to Eton). Successive Education Acts since 1980 have
extended parents’ rights to information about their schools, as well as
restricting LEA freedom to limit entry to popular schools. For better or
worse, schools are now in the market-place and survival will depend on
their ability to attract pupils, or rather to attract pupils’ parents. In this
context, any heads who respected the traditional classroom autonomy
of primary teachers could be charged with having an irresponsibly
laissez-faire attitude to their school’s future. With the introduction of
local management of schools, class teachers will increasingly find
themselves held accountable to the head, and through the head to
governors for the progress and motivation of their class. How this will
affect responses to pupils with special needs in unclear. The pessimistic
view is that teachers will increasingly seek individualized explanations
for problems of group interactions. This would result in more frequent
use of formal assessment under the 1981 Act in order to gain exemption
from the requirements of the National Curriculum. The more optimistic
view is that a developing sense of mutual accountability within a
school’s staff, as well as of partnership with parents and governors, will
encourage attempts to identify special needs in terms of problems in
pedagogy and in group interaction rather than in terms of supposed
deficits in individual children. What is not in dispute is that attempts to
find individual solutions to all of Warnock’s 20 per cent with special
needs, let alone Sir Keith Joseph’s under-achieving 40 per cent, are
doomed to failure. Yet this does not mean we can ignore individual
differences.
221
David Galloway
Social interaction and individual differences

Psychologists often find it easier to urge teachers to encourage co-
operative learning in their pupils than to learn from each other. Just as
clinical and educational psychologists sometimes underestimate the
relevance of contextual influences in the school or classroom when
assessing the needs of individual children, social psychologists
sometimes seem oblivious to the relevance of individual differences. In
theory, the 1981 Education Act requires educational psychologists to
assess children’s needs without reference to how or where they should be
met, since the decisions on placement and resources are a matter for the
Chief Education Officer (DES 1983). The distinction may make sense to
a tidy bureaucratic mind, but fails to recognize, not only that children fail
in a particular setting, but also that professional and parental responses to
their failure will be influenced by a variety of factors, including
perceptions about the availability of resources.
Trying to assess special needs without reference to how and where
they should be met is like trying to assess the ability to drive a car by
giving applicants a test on the Highway Code. Whether candidates know
the Highway Code is probably important, but their performance on the
road is a great deal more important. Similarly, psychometric information
may reveal some useful information on a child’s special needs, but
without additional information about the classroom context and about
the child’s responses, it will be, to say the least, of limited use.
How a child performs in a particular classroom, then, is the product of
a complex interaction between the child, the teacher and other pupils.
Two children may respond to the same experience in totally different
ways. Hargreaves (1983) describes the development of antiauthority
subculture in secondary schools in terms of the pupils’ attempt to
maintain their self-respect in the face of low status dictated by the
school’s official policies. From a different perspective Willis (1977) had
earlier noted the pride with which the ‘lads’ in his research valued their
ability to flout school rules. This kind of response is not confined to
secondary schools, but has origins in the pressure which much younger
children exert on each other in matters such as dress, speech and
behaviour in the classroom or in the playground. The question is not
whether some children adopt strategies of which teachers and parents
disapprove in order to maintain their sense of self-worth. Clearly some
do. The more interesting question concerns the range of strategies
adopted by children with special needs in response to feelings of
educational and/or social inadequacy.
Craske (1988) has reviewed studies indicating that motivation to
defend a sense of self-worth by avoiding effort, as proposed in different
ways by Hargreaves and Willis, is more prevalent amongst boys than
222
girls. The reverse is true of the ‘learned helplessness’ response in which

pupils perform increasingly poorly following failure. Girls, in other
words, are more likely to attribute their educational failure to lack of
ability, whilst boys are more likely to attribute it to a ‘boring’
curriculum.
Craske’s own work suggested that learned helpless children can be
innoculated against the negative effects of failure by a training
programme which encourages them to attribute success and failure to
effort or lack of effort respectively, rather than to lack of ability. In
contrast, training did not have the same potentially beneficial effects with
the self-worth group. These children continued to attribute failure to lack
of ability in spite of the attribution retraining programme. Consequently
it seemed likely that they would continue to rationalize their lack of
success in the curriculum by making little effort. How far teachers can
innoculate children against learned helplessness as part of their everyday
classroom work, without resorting to a structured programme, is not
clear. Nor is there adequate information on whether improvements
observed in structured programmes generalize to the classroom, or if so,
for how long. Similarly, there is a dearth of evidence not only on the
origins of self-worth motivation, but also on ways in which teachers may
encourage children in this group to develop a sense of self-worth based
on effort as well as ability (Beery 1975).
Conclusions
Contrary to popular belief, the commitment to integration in the 1981
Act was decidedly lukewarm (see Galloway and Goodwin 1987). The
Act managed to give a sympathetic nod to the groundswell of
international opinion in favour of integration, while acknowledging the
many practical problems that such a policy would create. The Act has
nevertheless encouraged greater attention than ever before to pupils
with special needs, especially those remaining in ordinary schools.
This attention has coincided with the government’s often-repeated
concern about the alleged under-achievement of a large minority of
pupils. Education is essentially a social process, and the learning
difficulties of an arbitrally defined minority of pupils are a product of
that process.
At the earliest stage of identification teachers define a child as having
special needs because they find the child’s behaviour, relationships with
other children or educational progress disturbing. This is neither a
criticism nor a value judgement, merely an example of how a child’s
perceived need results from a process of interaction. The pressure on
teachers, and hence the frequency of identification, may increase as a
result of recent legislation, though this is less self-evident than some
223
David Galloway
critics would have us believe. In contrast, there can be little doubt that
pressures will increase to integrate children who would previously have
been placed in special schools, and that the progress and adjustment of
academically less able pupils will continue to exercise the interest of
politicians.
Given this scenario, it would be tempting for teachers and
educational psychologists to seek ‘special’ teaching and learning
strategies for use in mainstream lessons. This would probably be a
futile search. The reason is that the teaching and learning strategies
which benefit ‘special’ pupils are precisely those which also benefit the
majority. Teachers vary in their effectiveness, but the evidence suggests
that they vary in their effectiveness with all pupils; those who are
effective in teaching able pupils are also effective in teaching pupils
with learning and adjustment difficulties and vice versa. The task for
teachers and educational psychologists, then, is to identify strategies
for utilizing available resources in ways which promote the teacher’s
interaction with all pupils, not just those whose behaviour or progress
is at one or other end of the continuum. The strategies in question are
neither new nor particularly controversial. Research on school and
teacher effectiveness indicates the importance of ‘open’ questioning
techniques, matching the task to the pupil’s ability, a rigorous
awareness of the effects of negative labelling, encouraging children to
monitor their own performance, involving them in formulating and
maintaining the school’s disciplinary system, and insisting that all
children contribute, and are seen to contribute, to the life and work of
the school. What all this underlines, though, is that children with
special needs benefit from the same teaching strategies as all other
children. We should emphasize what they have in common with their
peers, not the relatively minor differences.
References
Beery, R. (1975) ‘Fear of failure in the student experience’ Personnel and
Guidance Journal 54: 190–203.
Berger, M.M. (1982) ‘Applied behaviour analysis in education: a critical
assessment and some implications for training teachers’. Educational
Brennan, W. (1987) ‘Once more into the core’, Special Children XIV (October):
14–15.
Craske, M.L. (1988) ‘Learned helplessness, self-worth motivation and attribution
retraining for primary school children’, British Journal of Education
Department of Education and Science (DES) (1978a) Special Educational Needs
(The Warnock Report), London: HMSO.
Department of Education and Science (DES) (1978b) Primary Education in
England: a survey by HM Inspectors of Schools, London: HMSO.
224
Department of Education and Science (DES) (1983) Assessments and Statements

of Special Educational Needs, (Circular 3/83), London: DES.
Department of Education and Science (DES) (1984) Initial Teacher Training:
Approval of Courses (Circular 3/84), London: DES.
Department of Education and Science (DES) (1988a) Science for Ages 5–16,
London: DES.
Department of Education and Science (DES) (1988b) English for Ages 5–11:
Proposals of the Secretary of State for Education and Science and the
Secretary of State for Wales (The Cox Report), London: DES.
Doyle, W. (1985) ‘Classroom organisation and management’, in M.C.Wittrock
(ed.) Handbook of Research on Teaching, 3rd edn, New York: Macmillan.
Galloway, D. and Goodwin, C. (1987) The Education of Disturbing Children:
Pupils with Learning and Adjustment Difficulties, London: Longman.
Galloway, D., Mortimore, P. and Tutt, N. (1989) ‘Enquiry into discipline in
schools: evidence from the Departments of Educational Research and Social
Administration, University of Lancaster’, in N. Jones (ed.) School Curriculum
and Pupil Behaviour, Lewes: Falmer (in press).
Guralnick, M.J. (1981) ‘The social behaviour of pre-school children at different
developmental levels: effects of group composition’. Journal of Experimental
Child Psychology 31: 115–30.
Hargreaves, D.H. (1983) The Challenge for the Comprehensive School: Culture,
Curriculum and Community, London: Routledge & Kegan Paul.
Hargreaves, D.H., Hestor, S.K. and Mellor, F.J. (1975) Deviance in Classrooms,
Haviland, D. (1988) Take Care, Mr Baker, London: Fourth Estate.
Kutnick, P. (1983) Relating to Learning, London: Allen & Unwin.
Lemert, E.M. (1967) Human Deviance: Social Problems and Social Control,
Mortimore, P., Sammons, P., Stoll, L., Lewis, D. and Ecob, R.C. (1988) School
Plowden Report (1967) Children and their Primary School, 2 vols. Report of the
Pring, R. (1984) Personal and Social Education in the Curriculum, London:
Hodder & Stoughton.
Stott, D.H. (1971) The British Social Adjustment Guide, London: University of
London Press.
Strain, P.S. (1984) ‘Social interactions of handicapped pre-schoolers in
developmentally integrated and segregated settings: a study of generalization
effects’, in T.Field, J.L.Roopnarine and M.Segal (eds) Friendship in Normal
and Handicapped Children, Norwood, NJ: Ablex.
Willis, P. (1977) Learning to Labour: How Working Class Kids Get Working
Class Jobs, London: Saxon House.
225
Chapter thirteen
Co-operative learning
Robert E.Slavin
The final substantive chapter of the book provides an account by Robert
Slavin of the results of a considerable amount of research into co-
operative learning. The chapter has been written very much with the
school-based practitioner in mind. Slavin discusses the broad issues
involved, examines the results of evaluation research and then provides
practical guidelines for implementation. These serve as a clear invit-ation
to teachers to try it out in their own classrooms.
Slavin’s research has identified three key features of successful co-
operative learning programmes. Pupils must be operating under a regime
which rewards team results, but where each individual remains
accountable for their own contribution and where each individual has an
equal opportunity of being successful. In order to ensure this, Slavin’s
work involves the use of highly-structured programmes. Readers will
need to consider carefully the ways in which these might fit into their
own teaching situations.
The chapter serves a useful purpose in drawing together the various
issues explored throughout this volume. Co-operative learning brings
together considerations of social structures, personal identity and
autonomy, the development and importance of friendship, the
enhancement of motivation and the essentially social nature of the
activity of learning. Slavin’s chapter clearly implies that to maximize the
positive effects of all of this, teachers need to be prepared to undertake
some form of social engineering. Laissez-faire groupings or highly
individual approaches are unlikely to suffice. While Slavin’s approach
has limited applic-ability (it can only be strictly implemented in
situations where pupils’ work can be assessed in a relatively
straightforward right or wrong manner) it provides an interesting and
challenging example of the application of social psychological principles
and theory.
226
Introduction
‘Class,’ said Ms James, ‘Who remembers what part of
speech are words such as it, you, and he?’.
Twenty hands shoot up in Ms James’s top end junior class.
Another ten pupils try to make themselves small in hopes
that Ms James won’t call on them. She calls on Eddie.
‘Proverb?’
The class laughs. Ms James says, ‘No, that’s not quite right’.
The pupils (other than Eddie, who is trying to sink into the
floor in embarrassment) raise their hands again. Some of
them are halfway out of their seats, calling ‘me! me!’ in their
eagerness. Finally Ms James calls on another child
‘Elizabeth, can you help Eddie?’.
Think about the scene being played out in Ms James’s class, a common
sequence of events at every level of schooling, in every subject, in all
sorts of schools. Whether or not she is conscious of it, Ms James has set
up a competition between the children. The children want to earn her
approval, and they can only do this at the expense of their classmates.
When Eddie fails, most of the class is glad; pupils who know the answer
now have a second chance to show it, while others know that they are not
alone in their ignorance. The most ironic part of the vignette is when Ms
James asks if Elizabeth can ‘help’ Eddie. Does Eddie perceive
Elizabeth’s correct answer as help? Does Elizabeth? Of course not.
What is wrong with the competitive situation Ms James has
established? If properly structured, competition can be a healthy,
effective means of motivating individuals to perform. However,
competition in the classroom is typically of a less positive nature.
Consider what is going on below the surface of Ms James’s class. Most
of the class is hoping that Eddie (and also Elizabeth) will fail. Their
failure makes their classmates look good. Because of this, over time
pupils begin to express norms or values opposed to doing too well
academically. Pupils who try too hard are ‘teacher’s pets’, ‘nerds’,
‘grinds’, and so on. Pupils are put in a bind; their teachers reward high
achievement, but their peer group rewards mediocrity. As children enter
adolescence, the peer group becomes all-important, and except for a few
very talented individuals, most pupils accept their peers’ beliefs that
doing more than what is needed to get by is for suckers. Research on
secondary schools clearly shows that academic success is not what gets
children accepted by their peers.
Typical classroom competition can also be detrimental for another
reason. Recall the ten pupils who tried to make themselves invisible
227
Robert E.Slavin
when Ms James asked her question. For low achievers, a competitive

situation is both a poor motivator and, for some, almost constant
psychological torture. Children enter any class with widely divergent
skills and knowledge. Low-achieving pupils may lack the prerequisites
to learn new material. For example, children may have difficulty
learning long division because they never learned to subtract well. For
this and other reasons, success is difficult for may pupils, while it
comes easily for others. Success is defined on a relative basis in the
competitive classroom. Even if low achievers learn a great deal, they
are still at the bottom of the class if their classmates learn even more.
Day in, day out, low achievers get negative feedback on their academic
efforts. After a while they learn that academic success is not in their
grasp, so they choose other avenues in which they may develop a
positive self-image. Many of these avenues lead to anti-social,
delinquent behaviour.
How can teachers avoid the problems associated with classroom
competition? How can pupils help one another learn and encourage one
another to succeed academically?
Think back to Ms James’s class. What if Eddie and Elizabeth and two
other pupils had been asked to work together as a team to learn parts of
speech, and the teams were rewarded on the basis of the learning of all
team members? Now the only way for Eddie and Elizabeth to succeed is
if they make certain that they have learned the material and that their
team-mates have done so. Eddie and Elizabeth are now motivated to help
one another and to encourage one another to learn. Perhaps most
importantly, they are rooting for one another to succeed, not to fail.
This is the essence of co-operative learning (Slavin 1983 a,b). In co-
operative learning methods pupils work together in four member teams
to master material initially presented by the teacher. For example, in a
method called Student Teams-Achievement Divisions or STAD (Slavin
1986b), a teacher might present a lesson on map-reading, and then give
children time to work with maps and answer questions about them in
their teams. The teams are heterogeneous, made up of high, average and
low achievers, boys and girls, and students of any ethnic groups
represented in the class. After having a chance to study in their teams,
pupils take individual quizzes on map-reading. The individual quiz
scores are added up. All teams whose average scores meet a high
criterion receive special recognition, such as fancy certificates or having
their team picture posted in the classroom.
The idea behind this form of co-operative learning is that if pupils
want to succeed as a team, they will encourage their team-mates to excel
and will help them to do so. Often, pupils can do an outstanding job of
explaining difficult ideas to one another by translating the teacher’s
language into their own.
228
Of course, co-operative learning methods are not new. Teachers have

used them for many years in the form of laboratory groups, project
groups, discussion groups, and so on. However, recent international
research has created systematic and practical co-operative learning
methods intended for use as the main element of classroom organization,
has documented the effects of these methods, and has applied them to the
teaching of a broad range of curricula. These methods are now being
used extensively in every conceivable subject, from infant level to
college level, and in all kinds of schools throughout the world.
Co-operative learning methods

While social psychological research on co-operation dates back to the
1920s (see Slavin 1977), research on specific applications of co-
operative learning in the classroom did not begin until the early 1970s. At
that time four independent groups of researchers began to develop and
research co-operative learning methods in classroom settings. At present
there are researchers all over the world studying practical applications of
co-operative learning principles, and there are many co-operative
learning methods available. This chapter focuses on the methods that
have been most extensively researched and used in primary and
secondary classrooms.
Student team learning

Student team learning methods are co-operative learning techniques
developed and researched at Johns Hopkins University, USA. More than
half of all studies of practical co-operative learning methods involve
student team learning methods.
All co-operative learning methods share the idea that pupils work
together to learn and are responsible for one another’s learning as well as
their own. In addition to the idea of co-operative work, student team
learning methods emphasize the use of team goals and team success
which can only be achieved if all members of the team learn the
objectives being taught. That is, in student team learning the students’
tasks are not to do something as a team but to learn something as a team.
Three concepts are central to all student team learning methods: team
rewards, individual accountability and equal opportunities for success. In
these techniques teams may earn certificates or other team rewards if
they achieve above a designated criterion. The teams are not in
competition to earn scarce rewards; all (or none) of the teams may
achieve the criterion in a given week. Individual accountability means
that the team’s success depends on the individual learning of all team
members. This focuses the activity of the team members on tutoring one
229
Robert E.Slavin
another and making sure that everyone on the team is ready for a quiz or
other assessment which they will take without team-mate help. Equal
opportunities for success means that pupils contribute to their teams by
improving over their own past performance. This ensures that high,
average and low achievers are equally challenged to do their best, and the
contributions of all team members will be valued.
Research on co-operative learning methods (summarized below) has
indicated that team rewards and individual accountability are essential
elements for producing basic skills achievement (Slavin 1983a, b, 1988).
It is not enough simply to tell pupils to work together. They must have a
reason to take one another’s achievement seriously. Further, research
indicates that if pupils are rewarded for doing better than they have in the
past, they will be more motivated to achieve than if they are rewarded on
the basis of their performance in comparison to others, because rewards
for improvement make success neither too difficult nor too easy for
students to achieve (Slavin 1980).
Four principal student team learning methods have been extensively
developed and researched. Two are general co-operative learning
methods adaptable to most subjects and grade levels: Student Teams-
Achievement Divisions, or STAD, and Teams-Games-Tournament, or
TGT. The remaining two are comprehensive curricula designed for use in
particular subjects at particular grade levels: Team Assisted Individual
ization (TAI) for mathematics in grades 3 to 6 (approximately 8- to 12-
year-olds) and Co-operative Integrated Reading and Composition
(CIRC) for reading and writing instruction in grades 3 to 5
(approximately 8- to 11-year-olds). These four methods all incorporate
team rewards, individual accountability, and equal opportunities for
success, but in different ways.
Student Teams-Achievement Divisions (STAD)
In STAD (Slavin 1978, 1986b) children are assigned to four-member

learning teams that are mixed in performance level, sex, and, where
appropriate, ethnicity. The teacher presents a lesson, and then pupils
work within their teams to make sure that all team members have
mastered the lesson. Finally, all pupils take individual quizzes on the
material, at which time they may not help one another.
Children’s quiz scores are compared to their own past averages, and
points are awarded based on the degree to which they can meet or exceed
their own earlier performance. These points are then summed to form
team scores, and teams which meet certain criteria may earn certificates
or other rewards. The whole cycle of activities, from teacher
presentation, to team practice, to quiz usually takes three to five class
periods.
230
STAD has been used in every imaginable subject, from mathematics

to language arts, to social studies, and has been used from age 7 to
college students. It is most appropriate for teaching well-defined
objectives with single right answers, such as mathematical computations
and applications, language usage and mechanics, geography and map
skills, and science facts and concepts.
The main idea behind Student Teams-Achievement Divisions is to
motivate children to encourage and help each other to master skills
presented by the teacher. If pupils want their team to earn team rewards,
they must help their team-mates to learn the material. They must
encourage their team-mates to do their best, expressing norms that
learning is important, valuable, and fun. Pupils work together after the
teacher’s lesson. They may work in pairs and compare answers, discuss
any discrepancies, and help each other with any learning problems. They
may discuss approaches to solving problems, or they may quiz each
other on the content they are studying. They teach their team-mates and
assess their strengths and weaknesses to help them succeed on the
quizzes.
Although children study together, they may not help each other with
quizzes. Every pupil must know the material. This individual
accountability motivates children to do a good job tutoring and
explaining to each other, as the only way for a team to succeed is if all
team members have mastered the information or skills being taught.
Because team scores are based on children’s improvement over their own
past records (equal opportunities for success), all of them have the
chance to be the team ‘star’ in a given week, either by scoring well above
their past record or by getting a perfect paper, which always produces a
maximum score regardless of past averages.
A detailed guide to the use of STAD appears at the end of this
chapter.
Teams-Games-Tournament (TGT)
Teams-Games-Tournament (DeVries and Slavin 1978; Slavin 1986b)
was the first of the Johns Hopkins co-operative learning methods. It uses
the same teacher presentations and team work as in STAD, but replaces
the quizzes with weekly tournaments in which pupils compete with
members of other teams to contribute points to their team scores.
Children compete at three-person ‘tournament tables’ against others with
similar past records in mathematics. A ‘bumping’ procedure keeps the
competition fair, that is, the winner at each tournament table brings six
points to his or her team, regardless of which table it is; this means that
low achievers (competing with other low achievers) and high achievers
(competing with other high achievers) have equal opportunities for
231
Robert E.Slavin
success. As in STAD, high performing teams earn certificates or other

forms of team rewards.
TGT has many of the same dynamics as STAD, but adds a
dimension of excitement contributed by the use of games. Team-
mates help one another prepare for the games by studying worksheets
and explaining problems to one another, but when pupils are
competing their team-mates cannot help them, ensuring individual
accountability.
Team-Assisted Individualization (TAI)

Team-Assisted Individualization (Slavin, Leavey and Madden 1986)
shares with STAD and TGT the use of four-member mixed ability
learning teams and certificates for high-performing teams. But where
STAD and TGT use a single pace of instruction for the class, TAI
combines co-operative learning with individualized instruction. Also,
where STAD and TGT apply to most subjects and age levels, TAI is
specifically designed to teach mathematics to pupils in grades 3 to 6
(approximately 8- to 12-year-olds) or older pupils not ready for a full
alegbra course.
In TAI, children enter an individualized sequence according to a
placement test and then proceed at their own rates. In general, team
members work on different units. Team-mates check each others’
work against answer sheets and help one another with any problems.
Final unit tests are taken without team-mate help and are scored by
pupil monitors. Each week teachers total the number of units
completed by all team members and give certificates or other team
rewards to teams which exceed a criterion score based on the number
of final tests passed, with extra points for perfect papers and
completed homework.
Because pupils take responsibility for checking each others’ work
and managing the flow of materials, the teach can spend most classtime
presenting lessons to small groups of children drawn from the various
teams who are working at the same point in the mathematics sequence.
For example, the teacher might call up a decimals group, present a
lesson on decimals, and then send the children back to their teams to
work on decimal problems. Then the teacher might call the fractions
group, and so on.
TAI has many of the motivational dynamics of STAD and TGT.
Children encourage and help one another to succeed because they want
their teams to succeed. Individual accountability is assured because the
only score that counts is the final test score, and pupils take final tests
without team-mate help. Children have equal opportunities for success
because all have been placed according to their level of prior knowledge;
232
it is as easy (or difficult) for a low achiever to complete three subtraction

units in a week as it is for a higher-achieving classmate to complete three
long-division units.
Unlike STAD and TGT, TAI depends on a specific set of instructional
materials. These materials cover concepts from addition to an
introduction to algebra. Although designed for use in grades 3 to 6
(approximately 8- to 12-year-olds), they have been used for primary
instruction in grades 2 to 8 (approximately 7- to 14-year-olds), and as
remedial instruction in secondary schools. The TAI materials include
specific concept lesson guides that suggest methods of introducing
mathematical ideas by using demonstrations, manipulatives, and
examples. The curricular emphasis in TAI is on rapid, firm mastery of
algorithms in the context of conceptual understanding and on
applications of mathematical ideas to solution of real-life problems.
Co-operative Integrated Reading and Composition (CIRC)

The newest of the student team learning methods is a comprehensive
programme for teaching reading and writing in the upper elementary
grades called Co-operative Integrated Reading and Composition or
CIRC (Madden, Stevens and Slavin 1986; Stevens, et al. 1987). In
CIRC teachers use basal readers and reading groups, much as in
traditional reading programmes. However, pupils are assigned to teams
composed of pairs of pupils from two different reading groups. While
the teacher is working with one reading group, pupils in the other
groups are working in their pairs on a series of cognitively engaging
activities, including reading to one another, making predictions about
how narrative stories will come out, summarizing stories to one
another, writing responses to stories, and practising spelling, decoding
and vocabulary. Children work in teams to master main ideas and other
comprehension skills. During language arts periods children engage in
writing drafts, revising and editing one another’s work and preparing
for ‘publication’ of team books.
In most CIRC activities, pupils follow a sequence of teacher
instruction, team practice, team pre-assessments and quiz. That is, they
do not take the quiz until their team-mates have determined that they
are ready. Team rewards are certificates given to teams based on the
average performance of all team members on all reading and writing
activities. Because pupils work on materials appropriate to their
reading levels, they have equal opportunities for success. Pupils’
contributions to their teams are based on their quiz scores and final
independently written compositions, which ensures individual
accountability.
233
Robert E.Slavin
Other co-operative learning methods

Jigsaw
Jigsaw was originally designed by Elliot Aronson and his colleagues
(Aronson, et al. 1978). In Aronson’s Jigsaw method, children are
assigned to six-member teams to work on academic material which has
been broken down into sections. For example, a biography might be later
divided into early life, first accomplishments, major setbacks, later life
and impact on history. Each team member reads his or her section. Next,
members of different teams who have studied the same sections meet in
‘expert groups’ to discuss their sections. Then pupils return to their teams
and take turns teaching their team-mates about their sections. Since the
only way pupils can learn sections other than their own is to listen
carefully to their team-mates, they are motivated to support and show
interest in one another’s work.
A modification of Jigsaw developed at Johns Hopkins University as
part of the student team learning programme is called Jigsaw II (Slavin
1986b). In this method, children work in four- or five- member teams
as in TGT and STAD. Instead of each child being assigned a unique
section, they all read a common narrative, such as a book chapter, a
short story or a biography. However, each individual receives a topic on
which to become an expert. Those with the same topics meet in expert
groups to discuss them, after which they return to their teams to teach
what they have learned to their team-mates. Then pupils take individual
quizzes, which result in team scores based on the improvement score
system of STAD. Teams which meet preset standards may earn
certificates.
Learning together
David and Roger Johnson at the University of Minnesota developed the
Learning Together model of co-operative learning (Johnson and
Johnson 1987). The methods they have researched involve pupils
working in four- or five- member heterogeneous groups on assignment
sheets. The groups hand in a single sheet, and receive praise and
rewards based on the group product.
Group Investigation
Group Investigation, developed by Shlomo Sharan at the University of
Tel-Aviv (Sharan and Sharan 1976), is a general classroom organization
plan in which pupils work in small groups using co-operative inquiry,
group discussion, and co-operative planning and projects (Sharan and
Sharan 1976). In this method pupils form their own two- to six- member
groups. After choosing subtopics from a unit being studied by the entire
class, the groups further break their subtopics into individual tasks, and
234
carry out the activities necessary to prepare group reports. Each group
then makes a presentation or display to communicate its findings to the
entire class.
Research on co-operative learning

Co-operative learning methods are among the most extensively
evaluated alternatives to traditional instruction in use in schools today.
More than seventy high-quality studies have evaluated various co-
operative learning methods over periods of at least four weeks in regular
primary and secondary schools; sixty-three of these have measured
effects on pupil achievement (see Slavin 1988). These studies all
compared effects of co-operative learning to those of traditionally-taught
control groups on measures of the same objectives pursued in all classes.
Teachers and classes were either randomly assigned to co-operative or
control conditions, or they were matched on pre-test achievement level
and other factors.
Academic achievement
Overall, of sixty-three studies of the achievement effects of co-
operative learning, thirty-six (57 per cent) have found significantly
greater achievement in co-operative than in control classes. Twenty-six
(41 per cent) found no differences, and in only one study did a control
group out-perform the experimental group. However, the effects of co-
operative learning vary considerably according to the particular
methods used. As noted earlier, two elements must be present if co-
operative learning is to be effective: group goals and individual
accountability (Slavin 1983a, b, 1988). That is, groups must be
working to achieve some goal or earn rewards or recognition, and the
success of the group must depend on the individual learning of every
group member. In studies of methods of this kind (for example, STAD,
TGT, TAI, CIRC), effects on achievement have been very consistently
positive; thirty-four out of forty-one such studies (83 per cent) found
significantly positive achievement effects. In contrast, only four of
twenty-two studies (18 per cent) lacking group goals and individual
accountability found positive effects on student achievement. Two of
these positive effects were found in a study which compared four co-
operative learning methods to a competitive control group in a Nigerian
junior high science classes (Okebukola 1985). This study found greater
learning in Learning Together and Jigsaw classes (co-operative
learning methods lacking group goals and individual accountability)
than in competitive control classes, but the STAD and TGT classes
learned considerably more than did the other co-operative methods (or
the control method). The other two successful studies of co-operative
235
Robert E.Slavin
learning methods lacking group goals evaluated Group Investigation in

Israel (Sharan et al. 1984; Sharan and Shachar 1986). In Group
Investigation, pupils in each group are responsible for one unique part
of the group’s overall task, ensuring individual accountability. Then,
the group’s overall performance is evaluated. Even though there are no
specific group rewards (as are used in STAD, TGT, TAI and CIRC), the
group evaluation probably serves the same purpose.
Why are group goals and individual accountability so important? To
understand this, consider the alternatives. In many forms of co-operative
learning pupils work together to complete a single worksheet or to solve
one problem together. In such methods there is little reason for more able
pupils to take time to explain what is going on to their less able group-
mates or to task their opinions. When the group task is to do something,
rather than to learn something, the participation of less able pupils may
be seen as interference rather than help.
In contrast, when the group’s task is to ensure that every group
member has learned something, it is in the interests of every group
member to spend time explaining concepts to their group-mates.
Research on pupils’ behaviour within co-operative groups has
consistently found that those who gain most from co-operative work
are those who give and receive elaborated explanations (Webb 1985).
What group goals and individual accountability do is to motivate
children to give such explanations, to take one another’s achievement
seriously.
Co-operative learning almost never has negative effects on
achievement, so teachers can reap the social benefits of co-operation
by simply allowing pupils to work together or by giving them
problems to solve as a group. However, if they wish to use co-
operative methods to accelerate pupil achievement, the research
evidence is clear that they must set up co-operative activity so that
their groups are rewarded (for example, with certificates, recognition,
a small part of their grades) based on the individual achievement of
every group member. Usually this means that pupils study together in
their groups and then have their scores on individual quizzes
averaged to form a team score.
Do co-operative learning methods work equally well for all types of
children? In general, the answer is yes. While occasional studies find
particular advantages for high or low achievers, boys or girls, and so
on, the great majority find equal benefits for all types of children.
Sometimes a concern is expressed that co-operative learning will hold
back high achievers. The research provides absolutely no support for
this claim; high achievers gain from co-operative learning (relative to
high achievers in traditional classes) just as much as do low and
average achievers.
236
Intergroup relations
Classroom studies of co-operative learning have found quite consistently
that children express greater liking for their classmates in general as a
result of participating in a co-operative learning method (see Slavin
1983a). This is important in itself. Liking among pupils is especially
important, however, when they come from different ethnic backgrounds.
Anyone who has spent much time in a racially-mixed secondary school
in the USA knows that white children associate mostly with white
children, black children with black children, and so on. There is
substantial evidence that, left alone, ethnic separateness in schools does
not naturally diminish over time (Gerard and Miller 1975).
Social scientists have long advocated inter-ethnic co-operation as a
means of ensuring positive intergroup relations in a desegregated setting,
emphasizing, however, that positive intergroup relations would arise
from school desegregation if, and only if, pupils were involved in co-
operative, equal-status interaction sanctioned by the school (see Allport
1954; Slavin 1985a).
The research on co-operative learning methods has borne out the
expectations. Co-operative learning methods embody the requirements
of co-operative, equal-status interaction between children of different
ethnic backgrounds sanctioned by the school (Slavin 1985a). In most
of the research on intergroup relations, children were asked to list their
best friends at the beginning of the study and again at the end. The
number of friendship choices children made outside their own ethnic
groups was the measure of intergroup relations. Positive effects on
intergroup relations have been found for STAD, TGT, TAI, Jigsaw,
Learning Together, and Group Investigation models (Slavin 1985b).
Two of these studies, one on STAD (Slavin 1979), and one on Jigsaw II
(Ziegler 1981) included follow-ups of intergroup friendships several
months after the end of the studies. Both found that children who had
been in co-operative learning classes still named significantly more
friends outside their own ethnic groups than did children who had been
in control classes.
Integration and special needs

Although ethnicity is a major barrier to friendship, it is not so large as
the one between physically or mentally handicapped children and their
normal-progress peers. Since 1975 the integration movement in the
United States has created an unprecedented opportunity for
handicapped children to take their place in the mainstream of society
by integrating such children in regular classrooms. It has also created
enormous practical problems for classroom teachers, however, and it
often leads to social rejection of the handicapped children. The
237
Robert E.Slavin
situation is similar in the UK (Hegarty, et al. 1981). Because co-

operative learning methods have been successful in improving
relationships across the ethnicity barrier—which somewhat resembles
the barrier between integrated and normal-progress pupils—these
methods have also been applied to increase the acceptance of the
newly-integrated child.
The research on co-operative learning and integration has focused
on the academically-handicapped child. In one study STAD was used
to attempt to integrate children performing two years or more below
the level of their peers into the social structure of the classroom. The
use of STAD significantly reduced the degree to which the normal-
progress children rejected their integrated classmates, and increased
the academic achievement and self-esteem of all children, integrated
as well as normal-progress (Madden and Slavin 1983). Similar effects
have been found for TAI (Slavin, Madden and Leavey 1984), and
other research using co-operative teams has also shown significant
improvements in relationships between integrated academically-
handicapped pupils and their normal-progress peers (Ballard, et al.
1977; Cooper, et al. 1980). In addition, one study in a self-contained
school for emotionally disturbed adolescents found that the use of
TGT increased positive interactions and friendships among
participants (Slavin 1977). Five months after the study ended, these
positive interactions were still found more often in the former TGT
classes than in the control classes. In a study in a similar setting,
Janke found that the emotionally-disturbed children were more on-
task, better behaved, and had better attendance in TGT classes than in
control classes (Janke 1978).
Perhaps the most important fact about co-operative learning methods
in the integrated classroom is that these techniques are not only good for
the handicapped children, but they are among the very few methods for
helping these children that also have a clear benefit for all pupils in terms
of academic achievement.
Self-esteem
One of the most important aspects of a child’s personality is his or her
self-esteem. Many people have assumed that self-esteem is a relatively
stable attribute of a person that schools have little ability to change.
Several researchers working on co-operative learning techniques have
found, however, that teams do increase childrens’ self-esteem.
Children in co-operative learning classes have been found to have more
positive feelings about themselves than do those in traditional classes.
These improvements in self-esteem have been found for TGT and
STAD (Slavin 1983a), for Jigsaw (Blaney, et al. 1977), and for the three
methods combined (Slavin and Karweit 1981). Improvements in pupil
238
self-concepts have also been found for TAI (Madden, Slavin and
Leavey 1984). Why does this occur? First, it has been consistently
found that TGT and STAD pupils report that they like others and feel
liked by others more than control pupils do (Slavin 1983a). Liking of
others and feeling liked by others are obvious components of feeling
worthwhile. Second, it seems probable that pupils feel (and are) more
successful in their school work when they work in teams than when
they work independently. This can also lead to an increase in self-
esteem. Whatever the reason, the effect of co-operative learning
methods on self-esteem may be particularly important with regard to
the long-term effects on mental health. A child who has had a co-
operative, mutually supportive experience in school may be less likely
to be antisocial, withdrawn or depressed in later life. In fact, a
remarkable study in the Kansas City (Missouri) schools found that
lower socio-economic-status children at risk of becoming delinquent,
who worked in co-operative groups in the sixth grade (approximately
12 year-olds), had better attendance, fewer contacts with the police,
and higher behavioural rating by teachers in the seventh through
eleventh grades (approximately 13- to 16-year-olds) than did control
children (Hartley 1976).
Other outcomes
In addition to effects on achievement, positive intergroup relations,
greater acceptance of integrated children, and self-esteem, effects of co-
operative learning have been found on a variety of other important
educational outcomes. These include liking of school, development of
peer norms in favour of doing well academically, feelings of individual
control over the child’s own fate in school, and co-operativeness and
altruism (see Slavin 1983a). TGT (DeVries and Slavin 1978) and STAD
(Janke 1978; Slavin 1978) have been found to have positive effects on
pupils’ time on-task.
Conclusion
The positive effects of co-operative learning methods on a variety of
outcomes are not found in every study or for every method, but the
overall conclusion to be drawn from this research is that when the
classroom is structured in a way that allows pupils to work co-
operatively on learning tasks, they benefit academically as well as
socially. The greatest strength of co-operative learning methods is the
wide range of positive outcomes that has been found for them in the
research. Although there may be many ways to improve relationships
between children of different ethnic backgrounds or between
integrated and normal-progress pupils, few can also help improve
239
Robert E.Slavin
pupil achievement. And although there are certainly many ways to

accelerate pupil learning in one or more subjects or age levels, few
apply equally well in almost all subjects and age levels; and fewer
still can document improvements in learning and also show
improvements in pupils’ social relationships, self-esteem, liking of
school, and other outcomes.
Other special features of all the co-operative learning methods are
their inexpensiveness and their ease of use. In their simplest form, all
these methods require is that the teacher assign pupils to small teams,
give them material to study together, assess pupil learning, and give the
teams some kind of recognition or reward based on the average of the
team members’ scores. Teachers need minimal training to use these
techniques. Detailed teacher’s manuals are available for TGT, STAD and
Jigsaw II (Slavin 1986b), TAI (Slavin, Leavey and Madden 1986), and
CIRC (Madden, Stevens and Slavin 1986). Books describing the original
Jigsaw (Aronson, et al. 1978), the Learning Together model (Johnson
and Johnson 1987), and Group Investigation (Sharan and Sharan 1976)
are also available. Thousands of teachers have successfully used these
methods, especially STAD and TGT, with nothing more than the manuals
and books, and thousands more have done so after a one-day workshop.
Once teachers know how to use them, the methods require little or no
additional preparation time.
Because of their effectiveness, their practicality, and perhaps most
importantly, the fact that teachers and pupils simply enjoy using them,
co-operative learning methods are being used more and more widely
throughout the United States and in several other countries.
In sum, the research on co-operative learning methods supports the
usefulness of these strategies for improving such diverse outcomes as
pupil achievement at a variety of age levels and in many subjects,
intergroup relations, relationships between integrated and normal-
progress pupils and pupil self-esteem. Their widespread and growing use
demonstrates that in addition to their effectiveness, co-operative learning
methods are practical and attractive to teachers. The history of the
development, evaluation, and dissemination of co-operative learning is
an outstanding example of educational research resulting in directly
useful programmes that have improved the educational experience of
many children and will continue to affect many more.
Student team-achievement divisions
A practical guide
Student Teams-Achievement Divisions or STAD, is among the simplest
and most adaptable of the co-operative learning methods. It can be used
240
at any primary school level, and in any subject area in which there are
single right answers, such as mathematics, language, science and much
of social studies. STAD can be used all year in a given subject, but it is
most often used for six to eight week units at various points during the
school year.
Assigning students to teams

Teams in STAD have four or five members. Four is preferable; make five-
member teams only if the class is not divisible by four. To assign pupils to
teams, rank them from top to bottom on some measure of academic
performance (for example, past grades, test scores) and divide the ranked
list into quarters, with any extra pupils in the middle quarters. Then put
one pupil from each quarter on each team, making sure that the teams are
well balanced in sex and ethnicity. Extra (middle) pupils may become
fifth members of teams.
Determining initial base scores

In STAD, the points children contribute to their teams are based on the
degree to which their quiz scores exceed their past performance. This
makes it equally difficult for all children to contribute to their team
scores. Each child gets a base score which represents his or her past
average. If you are starting STAD after you have given at least three
quizzes or tests, average these scores to determine base scores.
Otherwise, use other suitable scores.
Preparing materials
Make a worksheet and a short quiz for each unit you plan to teach.
Schedule of activities
STAD consists of a regular cycle of instructional activities, as follows:
Teach: Present the lesson.
Team study: Pupils work on worksheets in their teams to
master the material.
Test: Pupils take individual quizzes.
Team recognition: Team scores are computed on the basis of team
members’ improvement scores, and a class
newsletter or bulletin board recognizes high-
scoring teams.
Descriptions of these activities follow.
Teach
Each lesson in STAD begins with a class presentation. The lesson should
take one to two class periods.
241
Robert E.Slavin
Team Study
During team study (one or two class periods) the team members’ tasks
are to master the material themselves and to make sure that their team-
mates have done so. Pupils have worksheets and answer sheets they can
use to practise the skill being taught and to assess themselves and their
team-mates. Only two copies of the worksheets and answer sheets are
given to each team, to force team-mates to work together. After
teaching a lesson, introduce team study to your class using the
following steps:
Read off team assignments.
Have team-mates move their desks together or move to team tables,
and allow pupils about ten minutes to decide on a team name.
Hand out worksheets and answer sheets (two of each per team).
Tell pupils on each team to work in pairs or threes. If they are
working on problems (as in mathematics), each pupil in a pair or
three should work on the problem, and then check with his or her
partner(s). If anyone missed a question, that pupil’s team-mates
have a responsibility to explain it. If pupils are working on short-
answer questions, they may quiz each other, with partners taking
turns holding the answer sheet or attempting to answer the
questions.
Emphasize to pupils that they are not finished studying until they are
sure all their team-mates will score full marks on the quiz.
Make sure that pupils understand that the worksheets are for studying
—not for filling out and handing in. That is why it is important for
pupils to have the answer sheets to check themselves and their team-
mates as they study.
Have pupils explain answers to one another instead of just scoring
each other against the answer sheet
When children have questions, have them ask a team-mate before
asking you.
While pupils are working in teams, circulate through the class,
praising teams that are working well, sitting in with each team to hear
how they are doing, and so on.
Test
After pupils have had adequate time to study as a team (one period is
usually sufficient), distribute the quiz. Do not let pupils work together on
the quiz; at this point they must show what they have learned as
242
individuals. Have them move their desks apart if this is possible. Either
allow pupils to exchange papers with members of the other teams, or
collect the quizzes to score after class. Test scores should be expressed in
terms of percentage correct.
Calculating individual improvement scores

As soon as possible after each quiz you should calculate individual
improvement scores and team scores and award certificates or other
rewards to high-scoring teams. If at all possible, the announcement of
team scores should be made in the first period after the quiz.
Pupils earn points for their teams based on the degree to which their
quiz scores exceed their base scores, as follows:
Calculating team scores

Calculate team scores by adding up the improvement points earned by
the team members and dividing the sum by the number of team members
present on the day of the quiz.
Recognizing team accomplishments

There are two levels of awards given based on average team scores, as
follows:
Note that all teams can achieve the awards; teams are not in
competition with one another. The criteria are set to make success
difficult but not impossible for all teams. If the criteria turn out to be too
easy or too difficult for your pupils, you may change them.
243
Robert E.Slavin
Provide some sort of recognition or reward for achieving at the

GREATTEAM level and a larger or fancier award for SUPERTEAMS.
For example, you might give small certificates to GREATTEAMS and
larger ones for SUPERTEAMS; let SUPERTEAMS line up for lunch and
break first, GREATTEAMS second; post photographs of
SUPERTEAMS and GREATTEAMS on notice boards, and so on. Your
own enthusiasm about team scores and a communication that you value
co-operation and success as a team are as important as any other factor in
the success of STAD.
Changing teams and revising base scores

After five or six weeks of STAD, reassign pupils to new teams. This gives
pupils who were on low-scoring teams a new chance, allows them to
work with other classmates, and keeps the programme fresh. Also, revise
pupils’ base scores at this time, computing new averages using pupils’
quiz scores from the previous five to six weeks.
Notes
1 This chapter was written under a grant from the Office of Educational
Research and Improvement, US Department of Education (no. OERI-G-86–
0006). However, any opinions expressed are mine and do not represent OERI
policy. Sections of this chapter are adapted from earlier articles by Slavin
(1986a, 1987).
2 For a more detailed description of STAD and other Student Team-Learning
methods, a teacher’s manual (Slavin 1986b) can be obtained from the Johns
Hopkins Team Learning Project, 3505 N.Charles Street, Baltimore,
Maryland, 21218, USA.
References
Allport, G. (1954) The Nature of Prejudice, Cambridge, Mass.: Addison-
Wesley.
Aronson, E., Blaney, N., Stephan, C., Sikes, J. and Snapp, M. (1978) The Jigsaw
Classroom, Beverly Hills, Cal: Sage.
Ballard, M., Corman, L., Gottlieb, J. and Kauffman, M. (1977) ‘Improving the
social status of mainstreamed retarded children’, Journal of Educational
Blaney, N.T., Stephan, S., Rosenfeld, D., Aronson, E. and Sikes, J. (1977)
‘Interdependence in the classroom: a field study’, Journal of Educational
Cooper, L., Johnson, D.W., Johnson, R. and Wilderson, F. (1980) ‘Effects on
cooperative, competitive and individualisticexperiences in interpersonal
attraction among heterogeneous peers’, Journal of Social Psychology 111:
243–52.
DeVries, D.L. and Slavin, R.E. (1978) ‘Teams-Games-Tournament (TGT):
review of ten classroom experiments’, Journal of Research and Development
in Education 12: 28–38.
244
Gerard, H.B. and Miller, N. (1975) School Desegregation: A Long-range Study,

New York: Plenum.
Hartley, W. (1976) ‘Prevention outcomes of small group education with school
children: an epidemiologic follow up of the Kansas City School Behavior
Project’, unpublished manuscript. University of Kansas Medical Center.
Hegarty, S., Pocklington, K. and Lucas, D. (1981). Educating Pupils with Special
Needs in the Ordinary School, Windsor: NFER-Nelson.
Janke, R. (1978) The Teams-Games-Tournament (TGT) Method and the
Behavioral Adjustment and Academic Achievement of Emotionally Impaired
Adolescents, paper presented at the annual convention of the American
Educational Research Association, Toronto, (April).
Johnson, D.W. and Johnson, R.T. (1987) Learning Together and Alone, (2nd edn),
Madden, N.A. and Slavin, R.E. (1983) ‘Cooperative learning and social
acceptance of mainstreamed academically handicapped students’, Journal of
Special Education 17: 171–82.
Madden, N.A., Slavin, R.E. and Leavey, M. (1984) ‘Combining cooperative
learning and individualized instruction: effects on student mathematics
achievement attitudes and behaviors’, Elementary School Journal 84: 409–22.
Madden, N.A., Stevens, R.J. and Slavin, R.E. (1986) A Comprehensive
Cooperative Learning Approach to Elementary Reading and Writing: Effects
on Student Achievement, Baltimore, Md: Center for Research on Elementary
and Middle Schools, The Johns Hopkins University, report no. 2.
Okebukola, P.A. (1985) ‘The relative effectiveness of cooperative and competitive
interaction techniques in strengthening students’ performance in science
classes’. Science Education 69: 501–9.
Sharan, S., Kussell, P., Hertz-Lazarowitz, R., Bejarano, Y., Raviv, S. and Sharan,
Y. (1984) Cooperative Learning in the Classroom: Research in Desegregated
Schools, Hillsdale, NJ: Erlbaum.
Sharan, S. and Shachar, C. (1986) ‘Cooperative learning effects on students’
academic achievement and verbal behavior in multi-ethnic junior high school
classrooms in Israel, unpublished paper. University of Tel Aviv, Israel.
Sharan, S. and Sharan, Y. (1976) Small-group Teaching, Englewood Cliffs, NJ:
Educational Technology Publications.
Slavin, R.E. (1977) ‘A student team approach to teaching adolescents with special
emotional and behavioral needs’, Psychology in the Schools 14 (1): 77–84.
Slavin, R.E. (1978) ‘Student teams and achievement divisions’, Journal of
Research and Development in Education 12: 39–49.
Slavin, R.E. (1979) ‘Effects of biracial learning teams on cross-racial
friendships’, Journal of Educational Psychology 71: 381–7.
Slavin, R.E. (1980) ‘Effects of individual learning expectations on student
achievement’, Journal of Educational Psychology 72: 520–4.
Slavin, R.E. (1983a) Cooperative Learning, New York: Longman.
Slavin, R.E. (1983b) ‘When does cooperative learning increase student
achievement? Psychological Bulletin 94: 429–45.
Slavin, R.E. (1985a) ‘Cooperative learning: applying contact theory in
desegregated schools’, Journal of Social Issues 41: 45–62.
Slavin, R.E. (1985b) ‘Team-assisted individualization: a cooperative learning
solution for adaptive instruction in mathematics’, in M.Wang and H.Walberg
(eds) Adapting Instruction to Individual Differences, Berkeley, Calif:
McCutchan, pp. 236–53.
Slavin, R.E. (1986a) ‘Learning together’, American Educator: 6–13.
245
Robert E.Slavin
Slavin, R.E. (1986b) Using Student Team Learning, (3rd edn). Baltimore, Md:
Center for Research on Elementary and Middle Schools, The Johns Hopkins
University.
Slavin, R.E. (1987) Cooperative learning: Student Teams, (2nd edn). Washington,
DC: National Education Association.
Slavin, R.E. (1988) ‘Cooperative learning and student achievement’, in
R.E.Slavin (ed.) School and Classroom Organization, Hillsdale, NJ:
Erlbaum.
Slavin, R.E. and Karweit, N. (1981) ‘Cognitive and affective outcomes of an
intensive student team learning experience’, Journal of Experimental
Education 50: 29–35.
Slavin, R.E., Leavey, M.B. and Madden, N.A. (1986) Team-Accelerated
Instruction—Mathematics, Watertown, Mass.: Mastery Education
Corporation.
Slavin, R.E., Madden, N.A. and Leavey, M.B. (1984) ‘Effects of Team-Assisted
Individualization on the mathematics of academically handicapped and
nonhandicapped students’. Journal of Educational Psychology 76: 813–19.
Stevens, R.J., Madden, N.A., Slavin, R.E. and Farnish, A.M. (1987) ‘Cooperative
Integrated Reading and Composition: two field experiments’, Reading
Research Quarterly 22: 433–54.
Webb, N. (1985) ‘Student interaction and learning in small groups: a research
summary’, in R.E.Slavin, S.Sharan, S.Kagan, R.Hertz-Lazaraowitz, C.Webb
and R.Schmuck (eds) Learning to Cooperate, Cooperating to Learn, New
York: Plenum, 148–72.
Ziegler, S. (1981) ‘The effectiveness of cooperative learning teams for increasing
cross-ethnic friendship: additional evidence’. Human Organization 40: 264–8.
246
Chapter fourteen
Individuals, groups and interventions
Colin Rogers and Peter Kutnick
Readers who have taken the unusual step of reading the book from start
to finish in the order that we, as editors, have judged to be appropriate,
will have been rewarded by concluding with the reading of the chapters
by Galloway and Slavin, both of which make strong calls for collective
and interventionist approaches. While they each get to these conclusions
from rather different starting points, and while their conclusions differ in
respect to a number of points of detail, each of them has drawn attention
to matters of fundamental importance for a social psychology of
education and its future.
Written from a British perspective in 1989, it is not surprising to find
Galloway concerned with the implications of the 1988 Act, the National
Curriculum and all that. It is important, though, to notice that Galloway’s
main concerns, writing from the vantage point of a special needs
specialist, are not with the National Curriculum itself, or even with its
associated testing, but more with the introduction of ‘market forces’ into
the education system. The shift away from a centrally funded, locally
planned system, to one that is dominated by central regulation subject to
local forces of supply and demand, has considerable potential
significance. It is, at present, far too early to assess just what the impact
of the Local Management of Schools (LMS), the right to ‘opt out’ of
Local Education Authority control, open enrolment and the increased
power of governing bodies will turn out to be. Galloway’s discussion of
these issues serves to remind us, however, of two important features of
these changes.
First, it is clear that teachers are to be placed in a position of greater
but perhaps, more importantly, different forms of accountability. While
the changes in the pressures that this will bring to bear upon teachers will
be important in their own right, it is the recognition of the existence of
pressures and constraints per se that has the greater significance for the
further development of the social psychology of education. Several of the
contributors to this volume have addressed themselves to the need to
247
develop the autonomy of the individual pupil, so as to create independent

learners. However, autonomous learners are less likely to be produced
within a system that places external constraints and pressures upon their
teachers. Galloway addresses this within the particular context of those
pupils with identified special needs, but his point is one that has a general
application. Social psychologists need to attend more closely to the links
between educational policy and classroom practice, both in terms of
drawing out policy implications from social psychological theory, but
also in terms of assessing the implications of policy within a framework
provided by such theory.
Second, from the outset it appears clear that any consideration of
educational issues cannot be conducted exclusively at the level of the
individual. Again Galloway makes the point most clearly with respect
to the pupil with special needs arguing that the nature of the
relationship between pupils in a class will be determined more by the
‘moral climate’ or ethos of the class than by the characteristics of the
children themselves. This is a point that social psychologists would
readily concede, recognizing as it does the essentially social nature of
human life.
Slavin’s work also represents a recognition of the importance of a
collective or group approach to education rather than one that is based
solely upon the activities of the individual. For present purposes,
however, the more important aspect of Slavin’s work is his advocacy of a
strongly interventionalist approach. For Slavin, it is clear that the benefits
of group-based work in school classrooms will not simply come about by
teachers asking pupils to work together. Co-operative teaching and
learning techniques are presented as a highly systematic, carefully
controlled and monitored strategy.
The teacher’s interventions, however, are aimed not at the
individual, nor directly at the activities of the group. Rather the aim of
the enterprise is to establish sets of conditions that will then generate
activity on the part of the pupil that will further development. As such,
Slavin can be seen to be unashamedly advocating a form of social
engineering.
Slavin’s approach can also be seen as a very pragmatic one.
Systematic research has revealed techniques which work, and
techniques which appear to be less successful. This systematic
approach has also enabled him to identify critical elements within the
process such as equal opportunity scoring, individual accountability
and team rewards. The point that we would wish to emphasize here is
that these clearly implicate the importance of the individual pupil
developing strategies to deal with the conditions thus established, and
thereby coming to behave in a manner which appears to be more
conducive to a higher level of learning.
248
Individuals within a social context

The idea of pupil strategies is not a new one (Woods 1980) and a number of
the chapters within this volume have discussed this aspect of classroom
life. Galton draws attention to the demands imposed upon pupils by
various forms of group work. Pupils find themselves in situations in which
they are required to establish strategies that enable them to obtain a sense
of ownership over the work they are undertaking, while at the same time
not wishing to be seen to be accepting responsibility for work that might be
judged to be substandard by their peers. A group, or collectively-based
classroom, is one in which a high proportion of the pupils’ activities
become public. Added to the appraisal of the teacher is the appraisal of the
peer group. While it is generally accepted that peer appraisal is likely to be
of greater significance for the older, secondary school pupil, it is clear that
it is already an issue for the primary child.
In developing classroom practices which place a greater emphasis
upon the work carried out by groups, teachers need to be mindful that
they are sensitive to the needs of the individuals that comprise them. A
group focus does not imply a neglect of the individual, but rather a focus
upon the individual as a group member. Group strategies will have their
effects via the influence they exert upon individual members. Just as the
individual pupil needs to interpret teacher actions (and bases his or her
response upon the interpretation rather than the action itself) so too must
the pupil interpret, and thereby respond to, the actions of the group.
Both Rogers and Schunk have discussed some of the burgeoning
literature dealing with the cognitions of individuals in social settings and
the effects that these cognitions will have upon the motivations and self-
esteem of the pupil. While this work does not lead directly to
instructional techniques in the way that Slavin’s work does, it does
provide a broad theoretical perspective which has a potential for
application in a wide variety of settings. If teachers are to be able usefully
to anticipate the strategies adopted by pupils in a variety of different
educational contexts, then they need to have a clear perspective on the
nature of those strategies and the educational consequences of the
adoption of any one of the range available. The concerns addressed by
Galton provide one clear example of potential application.
A consistent theme of the work written for this volume has been the
requirement for teachers to adopt a clearly interventionist role in the way
in which they approach their work with pupils. A child-centred,
‘progressive’ education is not seen as one in which the child is left to
discover themselves in whatever context that happens to be available.
Rather it is seen to be one in which the teacher takes a proactive role
designed to foster the skills of the child that make pupil progress more
likely to take place. The group is utilized as a device in this procedure,
249
partly out of recognition of the near impossibility of dealing directly with

the needs of each individual child and partly from recognition of the
group’s central role in pupil learning. As demands for high educational
standards grow, a positive approach to a child-centred (or, as the
emphasis is on groups, children-centred) education becomes more of a
necessity.
The call for a group focus, however, is not merely a pragmatic one.
The work discussed here by French, Edwards, Kutnick, Wright, Smith
and Maxwell illustrates in a variety of ways the degree to which the
primary curriculum and the child’s experience of that curriculum is by
necessity social in nature. From the centrality of play, through the
complexities of the development of pupil autonomy, to the social nature
of the knowledge to be taught and learnt itself, the social experience of
the child appears as fundamental. In their various ways each of these
authors is pointing us towards the conclusion that the social aspects of
primary schooling cannot be simply regarded as an optional extra to be
bolted on to the National Curriculum as and when time, resources and
teacher patience allow. The contents of the National Curriculum itself are
to be seen as the results of social interaction, and the transmission of the
same by further social process as necessary as well as desirable.
Dimensions of action
The work reported here can be considered to represent points on a
number of dimensions. The remainder of this concluding chapter will
seek, briefly, to illustrate the nature of these and then go on to set out
some of the future requirements for a social psychology of education.
One clear view that emerges in virtually all of the writing here is that
of the pupil as an active participant in the process of schooling. Pupils
have their own concerns with the maintenance of peer group relations,
self-identity, gender roles and so on, and each of these concerns interacts
with the demands made upon the pupil by the teacher. Education is often
characterized as a game in which the different participants play out their
roles and seek to obtain their own objectives within an implicitly agreed
and evolving set of rules. As long as it is not seen to trivialize the issues,
such an analogy is a useful one. (To paraphrase the late Bill Shankly,
‘Education is not a matter of life and death, it is much more important
than that’.) In order for teachers to be able to keep at least one step ahead
in the game plan it is necessary for a clear understanding of the nature of
the pupils’ active involvement to be obtained. It will not be sufficient to
appreciate that a pupil strategy will exist. As work in self-esteem and
motivation shows most clearly, the strategies developed by pupils are
often designed to deal with short-term objectives rather than with longer-
term ones. Effective short-term strategies may have less than desirable
250
longer-term outcomes. However, attempts to alter these are likely to fail

unless the suggested alternatives are also able to meet the functional
requirements of the pupils’ initial short-term concerns. As Galton has
suggested, pupils will want to maintain face-saving strategies, and,
therefore, greater educational rewards will not be sought out if the risk is
perceived to be too great.
On the active-passive pupil dimension then, it is clear that current
social psychological thinking firmly locates concern at the active end. In
switching to the next dimension of concern, it is important to bear this
prior observation in mind. The articles contained within this volume have
also clearly advocated (although in various ways) an interventionist
rather than a laissez-faire strategy on the part of teachers. Pupils may take
an active role in their own development and education but this is not seen
to imply that teachers need not strive to ensure that appropriate actions
are taken to direct and influence that development. However, it is clear
that interventions that are likely to be maximally effective are going to
have to work with, and not against, the grain of the pupils’ own strategies.
It is worth repeating the point made above that teacher strategies will be
most effectively developed by those with an understanding of the
strategies being adopted by pupils.
However, it is also well understood that teachers are themselves
constrained and pressured. At times of major change in the educational
system and with much public questioning of the effectiveness of schools
and teachers, such pressures and constraints are likely to be operating at
high levels. One would be unwise to add to these by proposing an
additional set of concerns and responsibilities. As has been demonstrated
by Bennett et al. (1984), it is already the case that teachers find it very
difficult to judge accurately the appropriateness of the interventions that
they do make.
Given present resource levels within education (and those that could
be realistically anticipated over the next decade or so), there are no easy
solutions to this problem. However, the difficulties are greater when
teacher interventions are exclusively conceptualized as operating at the
level of the individual pupil. Research into school effectiveness
(Mortimore et al. 1988: Smith and Tomlinson 1989) shows us that
schools and individual classes within schools can make a difference to
pupil performance. While there is still much to discover about the
processes involved here, there is some tentative agreement that it is the
ethos, style or character of the school or classroom that matters. Effective
schools do not carefully direct each individual pupil along their own
most appropriate path (but this is not to argue that this would be unwise if
it were clearly possible), rather they would seem to create the conditions
that facilitate the decision-making strategies of pupils (and presumably
of teachers also) themselves. In the terminology being adopted here, the
251
interventions of effective schools would be seen to be collective rather

than individualistic. They build on the social life of the school and turn it
to educational advantage. Less effective schools either fail to do this or,
worse, create an ethos that actually prevents effective social intercourse
from taking place.
This is not just a matter of good management leading to good
behaviour and good results (although Croll and Moses in their article
have argued that poor classroom management can be seen to lead to a
number of unintended and often unrecognized effects). Rather it is the
beginnings of a recognition that schools are essentially social places,
children’s learning and development are essentially social in nature,
knowledge itself (and, thereby, the contents of the curriculum) cannot be
separated from the effects of social life, and that good schools will
therefore seek to capitalize on this.
Prospect
If the potential implicated throughout this volume for developing and
improving present practice is to be realized, then it is essential that the
social psychology of education develops along two closely-related yet
separate tracks.
The first of these, naturally, is the continuing development of clear
social scientific models and theories germane to relevant aspects of
human social behaviour. Not all of this work, involving a variety of
research methods and drawing upon a wide range of theory, will have a
clear and immediate application to education.
For present purposes the second track is the more important one and
involves the development of good planning models of social process to
set aside the full-blown social scientific models. It is the development of
these planning models that will move us away from looking at relevant
bits of social psychology towards dealing with a social psychology of
education. Such planning models need to be developed in conjunction
with teachers and others with close and detailed experience of classroom
practice, for they need to dwell upon the concerns of the educator as
much as the concerns of the social scientist. Planning models will
provide teachers with ways of applying the fruits of the social scientific
study within the context of their everyday classroom concerns.
Slavin’s work on co-operative learning is one of the better examples
currently available. It is a model that provides ways of relating educational
objectives to social scientific understanding, and enables the classroom
teacher to develop her or his own applications within a framework that
attempts to establish the central features of any successful instance of co-
operative group work. It is this understanding of essential features that is
important as this enables the teacher to develop practice in ways that are
252
suited to the precise settings in which she or he has to operate. Social

engineering will never be very effective if every situation has to attempt to
accommodate to the exact same package.
Teachers need to be able to adapt and develop approaches to teaching
within a guiding framework. The task of social psychology is to develop
ways in which teachers can further develop their own understanding of
the nature of classroom life in order to be able to make more systematic,
sophisticated and effective judgements about the course of action to be
chosen. Above all else this involves an understanding of the importance
of process. It is here that Slavin’s model still falls short of the ideal as it
tells us too little about the critical elements of process. While his
identification of key features is a necessary step along this route to a full-
blown planning model, the lack of a detailed model of the dynamics of
pupil response to various teaching techniques still sets a limit on the
extent to which teachers are able to monitor and intervene in the process
once started.
While Wright has argued that teacher-pupil equality is often to be seen
as highly desirable, Edwards is right to point out that teaching and
learning take place within a relationship characterized by inequality. The
teacher’s greater power, influence and knowledge can all be harnessed as
forces for good, provided that teachers themselves have a sufficient
understanding of the ways in which these operate.
References
Bennett, S.N., Desforges, C., Cockburn, A. and Wilkinson, B. (1984) The
Quality of Pupil Learning Experiences, London: Lawrence Erlbaum
Associates.
Smith, D. and Tomlinson, S. (1989) The School Effect, London: Policy Studies
Institute.
Woods, P. (ed.) (1980) Pupil Strategies, London: Croom Helm.
253
Author index
Acker, S. 191, 207, 208 Bonhoeffer, D. 138

Ainsworth, M. 116 Borkowski, J.G. 82, 86, 87, 88
Alexander, R. 15, 16, 28 Boydell, D. 204, 208
Allport, G. 237, 244 Brennan, W. 217
Ames, C. 87, 88, 94, 99, 102, 106, Brook, A. 54
107 Brooks-Gunn, J. 172
Ames, R. 99, 107 Brophy, J. 16, 28
Andrews, G.R. 79, 88, 100, 107 Brown, A.L. 86, 88
APU 200, 201, 202, 205, 208 Bruner, J. 52, 66, 122
Aronson, E. 17–18, 28, 234, 240, 244 Buswell, C. 194, 208
Asher, S.R. 171, 174 Butterworth, G. 52
Ashton, P. 199, 203, 208
Atkinson, J. 94–6, 98, 99, 107 Cantor, N. 73, 88
Atkinson, J.M. 36, 45, 46 Carey, S. 52
Christie, J.F. 159
Baker, L. 82, 88 Clark, A.H. 181
Ballard, M. 238, 244 Cole, D. 146, 153
Bandura, A. 71, 75–8, 81, 88, 174 Collins, J. 77, 88
Bar-Tal, D. 100, 107 Connolly, J.A. 153
Baumrind, D. 116, 126 Cooper, H.M. 101, 107
Bealing, D. 12, 28 Cooper, L. 238, 244
Bearison, D. 119, 121 Corno, L. 71, 77, 79, 88
Beery, R. 223 Cotton, J. 18, 28
Bennet, A. 43, 46 Covington, M.V. 71, 88, 105, 106,
Bennett, S.N. 2, 16, 17, 28, 123, 251, 107
253 Craske, M.L. 222
Berger, M.M. 220 Croll, P. 31, 40, 46, 92, 126, 190–210,
Berndt, T.J. 178 252
Bigelow, B.J. 178 Cronbach, L.J. 77, 88
Billig, M. 53, 62, 65, 67 Crystal, D. 45, 46
Black, P. 19, 28
Blaney, N.T. 238, 244 Damon, W. 170, 172
Block, J. 40, 46 Davie, C.E. 153
Blumer, H. 33, 45, 46 Deci, E. 112, 125
Blurton-Jones, N. 118 Deem, R. 207–8, 209
254
Author index
Delamont, S. 32, 45, 46, 53, 191, 193, Grolnick, W.S. 125
207, 209 Guralnick, M.J. 215
DeVries, D.L. 17, 28, 231, 239, 244
Diener, C. 103, 107 Hallahan, D.P. 82, 89
Donaldson, M. 52 Hallinan, M.T. 173, 175
Douglas, J.W.B. 201, 205–6, 209 Hammersley, M. 37, 45, 47
Doyle, W. 219 Harari, O. 78, 89
Driver, R. 53, 54, 66 Hargreaves, D.H. 34, 47, 216, 220,
Dusek, J.B. 100, 101, 107 221, 222
Dweck, C.S. 79, 88, 100–3, 107 Harris, K.R. 82, 89
Hartley, D. 192–4, 199, 209, 239
Eccles, L. 101, 108 Hartup, W.W. 117, 170, 171, 173
Eder, G. 176 Haviland, D, 217
Edwards, A.D. 33–5, 39, 41, 46, Hayes, D.S. 182
49–69, 115, 118, 250, 253 Hegarty, S. 238, 245
Ellis, S. 87, 88 Heider, F. 96, 108
Erikson, E.H. 171 Heritage, J. 36, 45, 47, 53
Evans, M.A. 161 Higgins, E.T. 73, 89
Hinde, R.A. 181
Fabricius, W.V. 83, 89 Hodgeon, J. 40, 47
Feldman, R.S. 87, 89 Holt, J. 61
Festinger, L. 72, 73, 78, 89 Horrocks, J.E. 175, 183
Field, T. 117 Howes, C. 180, 182
Fishman, P. 40, 46 Humphreys, A.P. 165
Flanders 37 Huston-Stein, A. 159
Fodor, J.A. 52 Hymel, D. 174, 181
Foot, H.C. 182
French, J. and P. 7, 31–48, 198, 209, ILEA 205, 209
250 Ingram, J. 126
Freud, A. 117 Isaacs, S. 151
Freud, S. 135, 136–9
Frieze, I.H. 78, 89 Jackson, P.W. 34, 47
Furth, H.G. 115, 172 Janke, R. 238, 239, 245
Fuson, K.C. 82, 89 Johnson, D.W. 17–19, 29, 154, 234,
240, 245
Gaa, J.P. 86, 89 Jones, D. 12, 29
Galloway, D. 9, 211–25, 247, 248 Jones, E.E. 96, 108
Gallon, M. 2, 6, 11–30, 123, 196, Jowett, S. 160
199, 200–1, 209, 249, 251
Gardner, D. 160 Kelley, H.H. 96, 108
Garfinkel, H. 33, 45, 46 Kelly, A. 195–8, 209
Gerard, H.B. 237, 245 Keil, F.C. 52
Good, T.L. 40, 47 Kerry, T. 15, 29
Gilbert, G.N. 53, 54, 62 King, R. 192–4, 205–6, 209
Glasser, W. 27, 28 Kohlberg, L 142
Gordon, T. 27, 29 Krasnor, L.R. 145
Gottman J.M. 174 Kutnick, P. 70, 110–30, 172, 219,
Grabrucker, M. 40, 47 247, 250
255
Author index
Lacey, V.C. 34, 47 Paris 86, 90

Ladd, G.W. 174 Payne, G.C.F. 37–8, 47
Latour, B. 53 Peery, J.C. 173
Leiter, K, 45, 47 Peisach, E. 146, 153
Licht, B.G. 82, 89 Pellegrini, A.D. 164
Little, A. 97, 103, 108 Perret-Clermont, A.N. 121
Locke, E.A. 81, 82, 89 Peterson, P.L. 108
Piaget, J. 49, 51, 52, 65–6, 112, 116,
MacClure, M. 35, 47 118, 119, 121, 134–6, 138–42,
McGuire, K.D. 180, 182 146, 171, 177–8
McHoul, A. 36, 38, 47 Pintrich, P.R. 71, 90
Madden, N.A. 233, 238, 239, 240, Pollard, A. 22, 29, 124, 126, 127, 193,
245 194, 209
Maehr, M.L. 100, 106, 108 Potashin, R. 173
Mannarino, A.P. 180 Potter, J. 53, 54, 62
Manning, K, 152 Prescott, E. 159
Markus, H. 73, 89 Pring, R. 221
Marsh, H.W. 72, 89 Pringle, M.L.K. 203, 209
Marshall, H.R. 173 Pullantz, M. 118, 174
Maxwell, W.S. 9, 169–89, 250
Mead, G.H. 33, 72, 89, 171 Raphael, T.E. 86, 90
Measor, L. 22, 29 Renshaw, P.D. 174
Meggary, J. 198, 209 Rock, P. 45, 47
Mehan, H. 33, 37, 44, 45, 47, 53 Rogers, C.G. 8, 70, 78, 90, 92–109,
Miller, L.B. 163 190, 247, 249
Mitroff, I.I. 54 Rose, A.M. 33, 45, 47
Mize, J. 87, 89 Rosenberg, M. 71, 90
Montemayor, R. 72, 90 Rosenthal, T.L. 78, 90
Moore, S.G. 173 Rosenthal, R. 101, 108
Moreno, J.L. 172 Rowland, S. 26, 29
Morgan, V. 40, 47 Ruble, D.N. 73, 90
Morrison, A. 2 Rubin, K.H. 145, 180
Mortimore, P. 13, 14, 19, 29, 196, 197 Rubin, Z. 171, 182
200, 201, 205–6, 209, 220, 251, Rutter, M. 203, 209
253 Rumelhart, D.E. 73, 90
Moses, D. 190–210, 252 Russell, J. 52
Mosatche, H.S. 73, 90 Ryan, R.M. 125, 127
Nash, 34, 47, 126 St John, N. 175

Nias, J. 26–7, 29 Sacks, H. 36, 44, 47
Nicholls, J.G. 71, 78, 79, 90, 93, 102, Sadker, D. 40, 48
108 Sapp, G.L. 132
Scheirer, M.A. 75, 90
O’Connor, R.D. 174 Schunk, D.H. 7, 70–91, 92, 190, 249
Oden, S. 174 Schutz, 33
Okebukola, P.A. 235, 245 Schweinhart, L.J. 163
Olson, D.R. 52, 53 Selman, R.L. 177
Osborne, A.F. 118, 162 Serafica, F.C. 182
256
Author index
Serbin, L. 192, 209 Thorne, B. 40, 48

Shantz, C. 115 Tizard, B. 13, 19, 29, 37, 48, 117, 196
Shapira, A. 121 Tuma, N. 197, 199, 200, 205, 209,
Sharan, S. 18, 29, 234, 236, 240, 245 175
Sharp, R. 194, 209
Shavelson, R.J. 71–5, 91 Vandell, D.L. 117, 179
Simmel 33 Veroff, J. 73, 91
Simon, B. 112 Vygotsky, L.S. 49, 52, 66, 122
Simon, T. 154
Sinclair, J.McH. 37, 43, 48, 53 Walberg, H. 18. 29
Singleton, L.C. 175 Waldrop, M.F. 176, 183
Slavin, R.E. 4, 9, 17–18, 29, 70, 121, Walkerdine, V. 52, 65
127, 226–46, 247–9, 252–3 Waller, W. 45, 48
Smilansky, S. 146, 155–6 Waters, E. 116
Smith, D. 251, 253 Wertsch, J.V. 52
Smith, L. 34, 48 Wheldall, K. 16, 29
Smith, P.K. 8, 144–68, 145, 154, 163, Webb, N. 236, 246
165, 250 Weiner, B. 71, 77, 78, 86, 91, 94,
Speier, M. 36, 48 96–9, 102, 108–9
Spender, D. 193–4, 198, 209 Whiting, B. 121
Stanworth, M. 40, 48, 194, 209 Willes, M.J. 35, 48
Stevens, R.J. 233, 246 Willis, P. 222
Stipeck, D.J. 93, 108 Winne, P.H. 71, 77, 91
Stott, D.H. 220 Wood, D. 52
Strain, P.S. 87, 91, 213, 215 Woods, P. 22, 30, 249, 253
Strayer, F. 118 Woolgar, S. 53
Stubbs, M. 45, 48 Wright, D. 8, 131–43, 250, 253
Sullivan, H.S. 176 Wylie, R.C. 74, 91
Swarm, M. 40, 48
Sylva, K. 160 Yearley, S. 62
Yeomans, A. 15, 30
Tajfel, H. 122 Youniss, J. 115, 118, 121
Taylor, P.H. 147
Thompson, G.G. 162 Ziegler, S. 237, 246
257
Subject index
accountability 5 discourse 49–69

achievement 14, 17–18, 32, 39, disruption 142–204, 219
70–91, 101, 105, 191, 192,
199–203, 235, 238–9 education/knowledge 51
argumentation 50, 64, 118 Education Reform Act (1988) 1–2,
attachment 116, 135 217
attributions 77, 78, 83, 97–100, 101, Engineering Council 51
102, 104–6; attribution retraining Equal Opportunities Commission 191,
100; attributional feedback 79–80, 209
86 ethnic groups 17, 32, 39, 78, 228, 237
authority 116, 119–20, 134 ethnography 34, 44
autonomy 11, 111–14, 125, 136, 226,
248, 250 fear of failure 22, 95
friendship 119, 169–89, 192;
classroom studies: naturalistic, 123–4; behaviour with 179–81; choice of
experimental 125–6 175, 199; development of 176–9,
cognitive development 35, 63, 65, 226; in classrooms 175–6
112, 121–2, 148, 171–2, 250
conservation 119; and self-esteem and gender 31, 39, 40, 78, 92, 101, 104,
motivation 72, 73, 92, 102–3, 106 105, 190–210, 236
competition 17, 36, 106, 192–3, goal-setting 80–2, 86
227–28 grouping (in classrooms) 16–17, 122,
comprehensive schooling 12 124, 221
consensus (of pupil and teacher) 51, group work 11–30, 247
61–2, 127
co-operation; 25, 70, 106, 121, 183, hidden curriculum 216
190, 248; co-operative learning
11–12, 15, 17, 20, 127, 226–46, 252 individualism (and critiques of) 51,
Co-operative Integrated Reading and 216, 221
Composition (CIRC) 230, 233, individualistic goal structures 17
236–46 INSET 27, 39
Cox Report 39 integration (mainstreaming) 213–17,
cued elicitation 59 223, 237–8
curriculum 12, 92, 103, 199, 205, 207, interaction, in class 15, 31–48, 72, 87,
250; hidden 41 94, 100, 101, 191, 193–9, 250
258
Subject index
Jigsaw 234, 237–46 progressive (education) 51, 57, 123–4

Joint Council for the Education of pupils, as peers 63–4; dependence
Handicapped Children 214 124, 127
knowledge (common) 59 relationships (types) 121, 182; with

adults 115–17, 135; with peers 34,
language 33 117–19, 170
learning 49; children’s 50; theory of
112 scaffolding 66, 122
Schools Commission, Australia 40, 48
mastery 84, 103 science 62–3
meta-analysis 195–8 scientific knowledge 53, 62–3
Montessori 151 Scottish Education Department 172
moral development 131–43; climate self-concept, (self-efficacy, self-
(ethos) 215–16, 221; constraint esteem, self-worth) 11, 15, 17–18,
116, 134–9; mutuality 118, 24, 33, 39, 70–91, 92, 98–9, 102,
139–42 106, 190, 228, 238–9, 249–50
motivation 11, 22, 24, 70, 71, 78–82, self-fulfilling prophecy 39, 101
92–109, 190, 205, 226, 227–8, SEN (special educational needs) 190,
236, 249–250 196, 198, 203–4, 205, 211–25,
237–8, 248
National Curriculum 217–19, 247, social comparison 72, 73, 81, 87
250 social development 110–30, 170–2,
NFER (National Foundation of 213–17
Educational Research) 196, 197, socialization 33, 50, 61, 171
209 socio-economic class 39
non-verbal interaction 31 sociometry 172–6
Nuffield Foundation Mathematics STAD 17, 228, 230–1, 232–46
Project 51 symbolic interactionism 33
peer models (see relationships) 83–5, TAI 230, 231–3, 235–246

86; popularity 173 teacher control, 11, 54, 57, 63, 66,
play 144–68, 192, 250; aims 126; the role of 4, 5, 123, 133–4,
ineducation 147; and development 141–2, 175, 184–6, 219; and play
152–64; and school programme 148–50
159–64; categories 146; criteria TGAT 19
145; physical/rough-and-tumble TGT 230, 231–46
164–5
Plowden Report 1, 14–16, 29, 51, Warnock Report 212, 214, 220
112, 122, 151, 172, 218
problem solving 16
259

Galton, M. (1990) - Grouping and Group Work.

Uploaded by

Copyright:

Available Formats

Galton, M. (1990) - Grouping and Group Work.

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Galton, M. (1990) - Grouping and Group Work.

Uploaded by

Copyright:

Available Formats

The Social Psychology of the

London and New York

This edition published in the Taylor & Francis e-Library, 2002.

British Library Cataloguing in Publication Data

Library of Congress Cataloging in Publication Data

ISBN 0-203-19383-0 Master e-book ISBN

Figures and tables ix

11 Sex roles in the primary classroom

Paul Croll, Department of Education, Bristol

Process and structure in the primary

Peter Kutnick and Colin Rogers

individual attainment of that curriculum. We fear that teachers may

introduced in the 1920s and 1930s. At that time educators and

Practice and theory

Accountability and the reflective teacher

Reform Act. Accountability is undoubtedly of benefit to education

bases. Competitive versus co-operative learning processes will produce

Structure of the book

method for organizing children in classrooms. Group work has been

directly to further consideration of classroom motivation (Chapter 6),

succeeding chapters, especially those on friendship (Chapter 10) and sex

and laissez-faire groupings produce genuine learning amongst pupils

Grouping and group work

Definitions of group work

The use of group work in the primary classroom

there was some evidence that where pupils worked on the

Differing prescriptions of good practice

meanings clear to themselves by having to explain it to others and gain

Criticisms of group work

Alexander goes on to describe the work of a group of teachers who

The small group approach requires well chosen assignments

Research into group work and grouping in the United States

emerging according to the different focuses of these studies. Much of the

collaborative group work used earlier. Unlike Slavin’s approach, where

Effective group work in the primary classroom

Classroom. The main findings are described briefly in Galton (1987,

Further analysis of this data revealed an interesting pattern. An

Pupils’ perceptions of collaborative group work

‘My friends make me silly’.

Some put it even more strongly:

‘Look at the picture, get on!’

Ownership and risk during group work

Effective group work in the primary classroom

working independently. From the point of view of many pupils,

Teacher interviews and evaluations

including the improvement of understanding among slow learners, the

Learning and behaving during group work

opportunity to work co-operatively with their peers is contained in the

Towards a model of effective co-operative group work

practice was approved by the children later in the interview described

beginning to develop Inservice Educating Teachers (INSET)

Social interaction in the classroom

1. The nature of social interaction in classrooms.

The nature of social interaction in classrooms

arrangements. Within the social sciences this resulted in changes in

It is largely through talk that we develop our concepts of self,

Clearly the school classroom is a forum for this gradual process of

Thus, because of both theoretical principles, and the practical

The instructional context