ISSN 1463-6840
Proceedings of the British
Society for
Research into Learning
Mathematics
Volume 29 Number 3
Proceedings of the Day Conference held at
The Loughborough University, Saturday 14th November 2009
These proceedings consist of short research reports which were written for the
BSRLM day conference on 14 November 2009. The aim of the proceedings is to
communicate to the research community the collective research represented at
BSRLM conferences, as quickly as possible.
We hope that members will use the proceedings to give feedback to the authors
and that through discussion and debate we will develop an energetic and critical
research community. We particularly welcome presentations and papers from
new researchers.
Published by the British Society for Research into Learning Mathematics.
Individual papers © contributing authors 2009
Other materials © BSRLM 2009
All rights reserved. No part of this publication may be produced or transmitted
in any form or by any means, electronic or mechanical, including photocopying,
recording or any information storage retrieval system, without prior permission
in writing from the publishers.
Editor: M. Joubert, Graduate School of Education, University of Bristol, BS8
1JA
ISSN 1463-6840
Informal Proceedings of the British Society for Research into Learning
Mathematics (BSRLM)
Volume 29 Number 3, November 2009
Proceedings of the Day Conference held at Loughborough
University on 14 th November 2009
Interpretations of, and orientations to, “understanding mathematics in depth”: students
in MEC programmes across institutions
1
Jill Adler1, Sarmin Hossain1, Mary Stevenson2, Barry Grantham2, John Clarke3, Rosa Archer4.
1
King’s College London, 2Liverpool Hope University, 3University of East London, 4St Marys College
Twickenham
Symbolic addition tasks, the approximate number system and dyscalculia
Nina Attridgea, Camilla Gilmorea and Matthew Inglisb
7
a
Learning Sciences Research Institute, University of Nottingham, UK bMathematics Education
Centre, Loughborough University, UK
The T-shirt task: Using a mathematical task as a means to get insights into the nature of
the collaboration between in-service teachers and researchers
13
Claire Vaugelade Berg
University of Agder, Kristiansand, Norway
Motivating Years 12 and 13 study of Mathematics: researching pathways in Year 11 19
Rod Bond, David Green and Barbara Jaworski
Mathematics Education Centre – Loughborough University
Computer Based Revision
25
Edmund Furse
Swansea Metropolitan University
Children's Difficulties with Mathematical Word Problems.
31
Sara Gooding
University of Cambridge, UK
Some initial findings from a study of children’s understanding of the Order of
Operations
37
Carrie Headlam and Ted Graham
University of Plymouth
The role of attention in the learning of formal algebraic notation: the case of a mixed
ability Year 5 using the software Grid Algebra
43
Dave Hewitt
School of Education, University of Birmingham
Lower secondary school students’ attitudes to mathematics: Evidence from a large-scale
survey in England
49
Jeremy Hodgena*, Dietmar Küchemanna, Margaret Browna & Robert Coeb
a
King’s College London, bUniversity of Durham
Simon Says: Direction in Dialogue
55
Jenni Ingram, Mary Briggs and Peter Johnston-Wilder
University of Warwick
The relationship between number knowledge and strategy use: what we can learn from
the priming paradigm
61
Tim Jay
Graduate School of Education, University of Bristol
Aspects of a teacher’s mathematical knowledge in a lesson on fractions
67
Bodil Kleve
Oslo University College
Post-16 maths and university courses: numbers and subject interpretation
73
Peter Osmon
Department of Education and Professional Studies, King’s College London
The role of proof validation in students' mathematical learning
79
Kirsten Pfeiffer,
School of Mathematics, Statistics and Applied Mathematics, NUI Galway
An exploration of mathematics students’ distinguishing between function and arbitrary
relation.
85
Panagiotis Spyrou, Andonis Zagorianakos
Department of Mathematics, University of Athens, Greece
Identifying and developing the mathematical apprehensions of beginning primary
school teachers
91
Fay Turner
Faculty of Education, University of Cambridge
What Might We Learn From the Prodigals? Exploring the Decisions and Experiences of
Adults Returning to Mathematics
97
Robert Ward-Penny
Institute of Education, University of Warwick
Design Decisions: A Microworld for Mathematical Generalisation
103
Eirini Geraniou, Manolis Mavrikis, Celia Hoyles, Richard Noss
Institute of Education, London Knowledge Lab
Do students’ beliefs relating to the teaching of primary mathematics match their
practices in school?
Caroline Rickard
Initial Teacher Education, University of Chichester, UK
109
BSRLM Geometry working group: tasks that support the development of geometric
reasoning at KS3
115
Sue Forsythe; Keith Jones
University of Leicester; University of Southampton
Working group on trigonometry: meeting 4
121
Notes by Anne Watson
Department of Education, University of Oxford
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Interpretations of, and orientations to, “understanding mathematics in depth”: students
in MEC programmes across institutions
Jill Adler1, Sarmin Hossain1, Mary Stevenson2, Barry Grantham2, John Clarke3, Rosa Archer4.
1
King’s College London, 2Liverpool Hope University, 3University of East London, 4St Marys College
Twickenham
In this paper we present initial findings from our study of interpretations and orientations
to ‘understanding mathematics in depth’ among students in selected Mathematics
Enhancement Courses (MEC) in the UK. The MEC is a 26-week pre-Initial Teacher
Education (ITE) ‘mathematics subject knowledge for teaching’ course designed for, and
undertaken by, graduates wishing to teach mathematics at secondary level, but do not
have a Mathematics degree. It is completed before commencing with a PGCE. A
common theme running through the MEC documentation is the importance of
‘understanding mathematics in depth’. We are interested in what and how MEC students
interpret and orient themselves towards ‘understanding mathematics in depth’. In
designing and conducting our empirical work we have drawn upon a related project in
South Africa, which is exploring ‘mathematics for teaching’, specifically what and how
mathematics and teaching are co-constituted in mathematics teacher education
programmes. The MEC is an interesting empirical context for such study, as it is a
mathematics course, or set of courses, specifically designed for future teachers. We have
collected data through guided, semi-structured interviews with 18 students and 4
lecturing staff at three different institutions. The interpretations and orientations of MEC
students towards mathematics and the notion of ‘understanding mathematics in depth’,
we contend, provide additional insight into the developing notion of mathematical
knowledge in and for teaching.
Keywords: Understanding mathematics in depth, Mathematics Enhancement Course,
Mathematics initial teacher education
Introduction
This study conducted in the UK extends from the QUANTUM project which is
currently on-going in South Africa. Our focus here is on ‘understanding mathematics in
depth’ as interpreted within the Mathematics Enhancement Course, (MEC). The MEC
provides an alternative route into mathematics teacher education. It has been designed for
graduates who do not have a mathematics degree but wish to teach mathematics at secondary
level. It is a is a 26-week pre-Initial Teacher Education (ITE) ‘mathematics subject
knowledge for teaching’ course which is completed before commencing with a PGCE. The
programme has been running across a number of institutions in the UK for the past four years.
The motivation in the UK, has been to encourage and attract more graduates into retraining as
a secondary mathematics teacher. Graduates entering these programmes are required to have
an A-level pass in mathematics, or some indication of post secondary study with mathematics.
This seems to vary across institutions. Overall, students in these courses are moving from
only some post school mathematics, to preparation for being a secondary mathematics
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 1
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
teacher. The MEC programmes are thus focused both on deepening and extending
mathematical knowledge in ways appropriate to the profession of teaching.
The empirical field of QUANTUM in South Africa (SA) has been focused to date on
upgrading programmes for teachers whose qualifications were limited by apartheid teacher
education policy and practice. Orientations in South Africa similarly contain intentions to
deepen teachers’ subject knowledge in ways that are appropriate and useful in teaching.
Despite differences between the UK and SA, programmes share the phenomenon of providing
mathematical education specifically geared to the profession of teaching. The UK can be seen
as another context where ideas and experience about content knowledge for teaching
mathematics are being developed, particularly within the MEC. While the comparative
advantage offered by looking across these two contexts will illuminate the field in interesting
ways, the study in the UK will be of interest in the UK context itself, and more directly of
benefit to the shared understanding that we hope develops across MEC course participants
through their activity in the study. Of particular interest to us is the expressed commitment in
MEC course materials to “a deep understanding of fundamental mathematics”. We are
interested in what and how MEC students interpret and orient themselves towards
‘understanding mathematics in depth’.
Our empirical work has been conducted across three UK institutions: a MEC tutor and
six students from each of the institutions have been interviewed. We report here on our
findings from the student interviews only. Data has been gathered related to: MEC students’
motivations for and concerns in pursuing a teaching career in mathematics and joining the
MEC; the structure of the MEC and its activities; students’ orientations to learning on the
MEC and the meanings (interpretations) they attach to ‘understanding mathematics in depth'.
We are in the early stages of analyzing the data and in this paper we present and discuss
emerging trends that lead us to suggest that MEC students’ ‘understanding of mathematics in
depth’ is a discourse of mathematics interwoven with discourses of teaching and learning.
Background
In official and institutionally specific documentation for the MEC, there is a common
theme that mathematics teachers need to know and understand ‘mathematics in depth’. This is
a particular description of the specific ways in which teachers need to know and use
mathematics in order to teach well. Within the field of research on mathematics teacher
education (see Sullivan & Wood 2008; Even & Ball 2009) over the last 30 years there has
been focus and a growing interest in content knowledge for teaching. There is increasing
evidence of a positive relationship between student learning gains and what is being referred
to as teacher’s mathematical knowledge for teaching (Hill, Rowan & Ball 2005) and that the
nature of teachers’ mathematical knowledge and its use in practice matters for effective
teaching. Ball, Thames and Phelps (2008) for example, have examined what teachers know
and are able to use in practice and through this, elaborated and extended Shulman’s (1986)
categories of content knowledge for teaching that included: subject matter knowledge,
pedagogical content knowledge and curriculum knowledge. They distinguish common,
specialized, and horizon content knowledge, as forms of subject matter knowledge; and
knowledge of mathematics and students, mathematics and teaching, and mathematics and
curriculum, as forms of pedagogical content knowledge. Rowland et al. (2005) also built on
Shulman (1986) and developed a grounded model of mathematical knowledge used in
teaching – the Knowledge Quartet.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 2
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
These and other studies related to mathematical knowledge for teaching (e.g.
Neubrand 2008) do not cohere into a unified frame. However, they consistently reinforce
Shulman’s insight that knowing mathematics for oneself is not synonymous with enabling
others to learn mathematics. An expansive knowing of mathematics is illuminated in the
MEC through its intention that prospective teachers need to understand mathematics ‘in
depth’ and hence provides an interesting context in which to investigate what and how
‘understanding mathematics in depth’ is interpreted by MEC students who are close to
completing their MEC.
Research Method
Three institutions named A, B, C, for the purpose of anonymity, took part in this study. The
sample included 6 MEC students from each institution. The selection of students was
purposive, and guided by a set of criteria so that the sample included students with:
mathematical and non-mathematical backgrounds; different cultural and educational
backgrounds; and ranging participation and performance in the course. A total of 18 semistructured interviews were conducted, guided by an interview agenda lasting approximately
an hour. The interviews were recorded and transcribed. Each interview was conducted by a
researcher from Kings College London and one of the MEC tutors in an institution which was
not their own. This collaborative approach is beneficial. . Firstly, as insiders, the MEC tutors
are well placed to probe students’ interpretations by drawing on aspects/activities of the MEC
programme. Secondly, the tutors are all ‘new’ researchers and are enjoying participation in a
collaborative research community.
Sample Description- Within the study sample 14 students were educated in the
UK and 4 educated abroad (2 students from Nigeria, 1 from Cyprus, 1 from Pakistan).
Furthermore the point at which the students joined the MEC in their lives varied: 3 students
joined straight after finishing their degrees at University. 4 students had had a short career and
11 students had had a long career before entry into the programme. Out of the 18 students,
only 5 students had some form of teaching career or teaching experience before joining the
MEC. In terms of educational qualifications 16 students had bachelor’s degrees, 1 qualified
in a PG Programme and 1 qualified in Access to Primary School Teaching. The subjects
studied by the students at degree level ranged from: Educational Studies, Business, Computer
Science, Engineering, to Sports Science. In regard to mathematical background/qualifications:
12 students had a Level 3 Mathematics qualification (i.e. A Level); 3 students had an
equivalent of a Level 3 Mathematics qualification and, interestingly, 4 students had a Level 2
Mathematics qualification (i.e. GCSE or its equivalent), with some mathematics related study
or experience in their educational and work histories.
The Interviews- Students were probed on: a). How and in what ways their learning of
mathematics in the MEC has been different to their learning of mathematics at school and/or
at University b). What and how they interpreted “understanding mathematics in depth”? c).
How they rated five different statements related to “understanding mathematics in depth”.
These statements had been elicited from earlier interviews with MEC tutors where the MEC
tutors had defined in their own terms what it was to “understand mathematics in depth”. The
MEC students were asked to put these five statements in the order of importance and then
probed on their specific ordering. The statements included: Understanding mathematics in
depth means being able to: 1). Justify your mathematical thinking. 2). Explain and/or
communicate mathematical ideas and thinking to others. 3). Understand why and how these
procedures work. 4). Make the connections between concepts and between procedures. 5).
Identify structure and generality.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 3
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Data Analysis- We are in the initial processes of the data analysis, and will be
concerned with 1). MEC students’ backgrounds, concerns and motivations for pursuing the
MEC. 2). Their preferences within MEC activities. 3). Their orientations to mathematics in
the MEC and their learning and interpretations of understanding of mathematics in depth. As
a mathematics teaching focused programme, we anticipate interpretations and hence
discourses (by which we mean representations of social practice, Van Leeuwen, (2008) ) of
mathematics, of teaching and of learning to be present in the data. We are also interested in
whether and how the students affiliate or dissociate themselves in their talk about
mathematics, teaching and learning. We have developed an initial coding based on these
theoretically informed assumptions, and our coding has been extended through our interaction
with the data. For example, we found that students often associate mathematics and teaching
elements when making reference to their Self and their Environment. For example, “I am
more mathsy now” and “I am more confident to teach” in these instances for example the text
would be coded as the “Self and mathematics”, and the “Self and teaching” respectively.
These instances often show how the students are foregrounding themselves either in
mathematics, in teaching or in both mathematics and teaching as they journey through on the
MEC programme. So far the data has been analysed thematically. Our next step is to refine
our analysis by conducting a critical discourse analysis, attending more closely to students’
positive and negative descriptions (e.g. what mathematics in depth IS and what it is NOT);
and to whether and how they associate or distance themselves in what they say (e.g. I think
this means vs. they say it means).
Preliminary Findings
As noted, to date we have focused our analysis on students’ spontaneous responses to what
“understanding mathematics in depth” means for . As anticipated, there are mathematical,
teaching and learning discourses that thread through the transcripts.
Below is an example of some of the mathematical discourses that were prevalent
across the data.
Em, what it means to me is understanding, you know, the different concepts, how they
originated, you know, what the idea was, therefore, a concept came about. How it can be
applied is a useful lesson in everyday life, and how it’s interconnected with other aspects of
mathematics. So that’s how I feel. We will have a deeper understanding of something.
And, er, it’s like an open box – we don’t just look at the problem and say, ‘Okay, this is the
calculus,’ ... No, you can have, you can try different methods, you can be flexible, you...
and see what applies. So if you have a deeper understanding of it, you’re not going to be
scared of anything because you know there is a solution, or you can attain to solution.
–
–
–
Here understanding Mathematics in depth includes
Discourses of knowing the history of mathematical concepts
–
Discourse of mathematics as connected knowledge
Other mathematically related discourses in the data were:
Discourses of proof, knowing basic proofs or first principles
Discourses of understanding formulae and their parts; when these are used
Also reflected in the above quote is a discourse of learning, of acquiring a disposition
towards mathematics as something you can do; knowing mathematics in depth means being
able to be flexible. Across the data there were learning related discourses, where students
talked about understanding mathematics in depth in terms of relearning mathematics, or
learning new maths in particular ways. For example,
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 4
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
[Understanding mathematics in depth] means that whenever it comes to understanding it ...
you’re not blinkered by what you’re relaying to someone else. ... you’ve actually looked at
something, tried to look at it in depth... you may have been led down some paths that come to
dead ends, you may have gone down some further paths that leads you to something else. Er, I
think through doing that you're more able to build analogies within your own mind, so
therefore if you’re trying to explain something to somebody. Because you’ve gone a bit
beyond or possibly a bit more beyond what’s expected of you, or what you think should be
expected of a student, you actually look at it from different perspectives. So I think whenever it
comes to actually passing that knowledge on, you actually get an enthusiasm for it, and I think
if you can actually gain the enthusiasm for what you’re looking at, there is a point beyond
which it loses an effect, but if you get it just right, I think you can actually – I wouldn’t, I’m
not going to say inspired because I think that’s an utterly wrong thing to say, but I think you
can actually enthuse someone...
And similarly, discourses of teaching thread through discussion of learning –
indicating that students interpret understanding mathematics in depth as needed for teaching.
For example:
But the fact that, you know, you should never give a... a kind of an equation and there’s...
there’s always like a practical example or an example in layman’s terms that can help
somebody understand it rather than, em, put it into a lot of letters or numbers or something
straightaway, because, when you... when you... when you... you often put, em... or express
it in terms of, em, letters, for example. I’m... I’m... I’m one of these people that if I just see
a lot of letters, it’s very intimidating, especially for children, like for somebody to see that,
it’s very intimidating, which can even... even stop them starting the problem at all. So put
that into a numerical example and then, for example, linking for patterns, for example, to
start off with, and then going on to try and express that in terms of first principles or, em, or
algebra...... it’s a way... well, it works for me, I think, em, and I think... I think it works well
with children.
Here and across the data is a representation that understanding mathematics in depth is
enabling others to know mathematics. Within this are
– Discourses of dealing with ‘difference’ (learners are diverse)
– Discourses of making mathematics ‘practical’
Discussion/Conclusion
In summary, our initial analysis of students’ orientations to ‘understanding
mathematics in depth’ in these interviews reflects an amalgam of mathematical, teaching and
learning discourses. Understanding mathematics in depth meant: knowing that and knowing
why (knowing basic proofs or being able to work from first principles; understanding
formulae, their parts and when these are used; knowing the history of mathematical concepts;
and being able to connect different aspects of mathematics and its applications. These are
resonant with Shulman’s elaboration of the Subject Matter Knowledge (SMK) that forms part
of a teacher’s professional knowledge base (Shulman, 1986). Of course, these were not
uniformly described with apparent different emphases a function of different schooling
histories (e.g. some students were schooled in very different educational cultures outside the
UK), different mathematical histories and different experiences in the MEC course. Across
the interviews, however, was a relatively strong affiliation to teaching requiring depth
understanding of mathematics, appreciation of opportunities to engage with mathematics in
extended ways in the MEC, and particular appreciation of activities in the MEC that closely
aligned with teaching (e.g. peer teaching activities). Many of the students pointed to the social
relations in the MEC and the care taken by lecturers in their learning of mathematics. Some
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 5
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
expressed directly that they believed the MEC would give them a ‘leg-up’ in the PGCE, as
they had had opportunity to revisit school mathematics and relearn it ‘in depth’. The next
question of course, is how their appreciation for understanding mathematics in depth
recontextualises into the actual teaching experience and practices and this is what we hope to
be pursuing next.
References
Adler, J., D. Ball, et al. 2005. Reflections on an emerging field: Researching mathematics
teacher education. Educational Studies in Mathematics 60(2): 359-381.
Adler, J. and Davis, Z. 2006. Opening another black box: Researching mathematics for
teaching in mathematics teacher education. Journal for Research in Mathematics
Education 37(4): 270-296.
Ball, D.L., Thames, M.H., and Phelps, G. 2008. Content knowledge for teaching: What makes
it special? Journal of Teacher Education, 59(5): 389-407.
Even, R., & Ball, D. L. 2009. The Professional Education and Development of Teachers of
Mathematics. The 15th ICMI Study Springer.
Hill, H. C., B. Rowan, and Ball, D.L. 2005. Effects of Teachers' Mathematical Knowledge for
Teaching on Student Achievement. American Educational Research Journal 42(2):
371-406.
Neubrand, M. 2008 Knowledge of teachers—Knowledge of students: Conceptualisations and
outcomes of a Mathematics Teacher Education Study in Germany. Proceedings of the
Symposium on the 100th Anniversary of ICMI.
Rowland, T., P. Huckstep, et al. 2005. Elementary teachers' mathematics subject knowledge:
the knowledge quartet and the case of Naomi. Journal of Mathematics Teacher
Education 8(3): 255-281
Shulman, L. S. 1986. Those who understand knowledge growth in teaching. Educational
Researcher 15(2): 4-14.
Sullivan, P., and Wood, T. (Eds.). 2008. The Handbook of Mathematics Teacher Education.
Knowledge and Beliefs in Mathematics Teaching and Teaching Development. (Vol.
1). Rotterdam: Sense Publishers.
Van Leeuwen, T. (2008) Discourse and practice: New tools for critical discourse analysis.
Oxford. Oxford University Press.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 6
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Symbolic addition tasks, the approximate number system and dyscalculia
Nina Attridgea, Camilla Gilmorea and Matthew Inglisb
a
Learning Sciences Research Institute, University of Nottingham, UK
bMathematics Education Centre, Loughborough University, UK
Several recent theorists have proposed that dyscalculia is the consequence of a
disconnect between the so-called ‘approximate number system’ and formal symbolic
mathematics. Such theories propose that symbolic exact mathematics is built out of
approximate representations of quantity. Here we investigate this proposal by testing
whether non-dyscalculic adults appear to use their approximate number systems when
tackling symbolic tasks. We find a strikingly similar pattern of responses on two
approximate addition tasks, one where participants saw numerosities represented as dots
and one where the numerosities were represented with Arabic symbols. These findings
are consistent with the view that non-dyscalculic adults do indeed use the approximate
number system when dealing with symbolic mathematics.
Keywords: approximate number system, arithmetic, dyscalculia, number sense.
There is an increasing body of evidence that humans have an inbuilt ‘number sense’ – or
approximate number system (ANS) – which supports approximate numerical operations (e.g.,
Cordes, Gelman, Gallistel & Whalen 2005, Dehaene 1992, 1997). The ANS, which is present
in infants, children and adults, involves approximate, abstract representations of number that
support both the comparison and manipulation of numerosities (Barth, Kanwisher, & Spelke
2003, Barth et al. 2006; Pica, Lemer, Izard, & Dehaene 2004). When adults and children
compare or add symbolic numerals, these approximate representations seem to be activated
(Dehaene 1997, Gilmore, McCarthy, & Spelke 2007, Moyer & Landauer 1967).
It is even suggested that a disconnect between the ANS and formal symbolic
mathematics is the root cause for so-called mathematics disorder or dyscalculia. The
Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association
2000) describes mathematics disorder as a condition which causes a person to have
substantially lower scores on mathematics tests than would be expected given their age,
intelligence and educational background. The Department for Education and Skills (2001),
using the term dyscalculia, offered a similar characterisation, emphasising how difficult it is
for dyscalculic students to acquire “simple number concepts”. Rouselle & Noël (2008) have
suggested that dyscalculia is the result of a disconnect between approximate numerical
representations held in the ANS and the symbols used in formal mathematics. They found that
children classified as dyscalculic could perform well on non-symbolic comparison tasks (i.e.
tasks of the form “are there more blue dots or red dots?”), but could not successfully tackle
the equivalent tasks represented symbolically.
The suggestion that dyscalculia is the result of a disconnect between the ANS and
symbolic mathematics is consistent with other recent claims that the ANS provides the basis
of all exact mathematics (e.g. Gilmore, McCarthy & Spelke 2007). But if these accounts were
correct, we would expect that non-dyscalculic adults, who are fluent in simple arithmetic
tasks, would somehow harness the ANS when tackling approximate symbolic mathematics
tasks. Our goal in this paper is to explore this account of dyscalculia by examining the
behaviour of non-dyscalculic students during both symbolic and non-symbolic addition tasks.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 7
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Experiment 1
The primary goal of Experiment 1 was to provide a baseline for the ANS in the context of
non-symbolic addition tasks. Participants saw three numerosities, n1, n2, n3, represented as
coloured dots. Their task was to determine which was greater out of n1+n2 and n3. Previous
research has found that tasks during which the ANS is used to compare numerosities tend to
show a ratio-effect. That is to say that, because representations within the ANS are only
approximate (internal representations of a numerosity n are assumed to be distributed
normally around n), comparisons of numerosities which have a ratio close to one lead to more
errors than those of numerosities with a ratio far from one (e.g., Barth et al. 2006, Gilmore et
al. 2007). We would expect to find such an effect in the current task.
Method
Twelve staff and students (seven male) from the University of Nottingham (aged 23-36,
M=29) received £4 for participating in the study.
Participants were tested individually. Displays were presented on a 17'' Philips 170B
LCD placed at eye level, and were viewed from approximately 60cm away. The stimuli
consisted of three dot arrays. The two addend arrays were blue dots against a white
background and the comparison set array was red dots against a white background. To
prevent participants using strategies based on continuous quantities correlated with number
(dot size, luminance, total enclosure area), the stimuli were created following the method of
Pica et al. (2004). For each problem two sets of stimuli were created: one in which the dot
size and total enclosure area decreased with numerosity, and one in which the dot size and
total enclosure area increased with numerosity.
Fifty problems were used for each approximate ratio. Sum totals for number to sum
ratios less than 1 were the integers from 21 to 70, sum totals for number to sum ratios greater
than 1 were the integers from 11 to 60. Comparison numbers were related to sum totals by
approximate ratios 8:5, 7:5, 6:5, 5:6, 5:7, 5:8. The addends were randomly chosen in such a
way that neither was larger than the comparison number, or less than 5 (for those problems
with a sum total less than 30) or 10 (for all other problems). Each set of dots was displayed
for 300msec, in the order shown in Figure 1.
Figure 1: A non-symbolic addition task.
Each problem appeared twice: once with each stimuli set (according to Pica et al.’s
method (2004)). Responses (and response times) were recorded via coloured blue (leftmost)
and red (rightmost) buttons on a five-button response box.
In summary, the experiment followed a 3 (approximate ratio: 8:5, 7:5, 6:5) × 2 (ratiodirection: number-larger, sum-larger) design. This yielded a total of 600 trials for each
participant. The experiment was preceded by a practice block of 10 trials.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 8
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Results
Mean accuracy rates for each ratio and ratio-direction are presented in Figure 2. These were
analyzed using a 3 (approximate ratio: 5:6, 5:7, 5:8) ×2 (ratio-direction: number-larger, sumlarger) repeated-measures analysis of variance (ANOVA). There were main effects of ratio,
F(2,10)=110.72, p<.001, and ratio-direction, F(1,11)=6.09, p=.031. Accuracy was higher for
problems with the number larger, M=79%, SD=7.6%, than the sum larger, M=64%,
SD=15.9%. As predicted, we found the characteristic ratio-effect, as there was a significant
linear trend of ratio, F(1,11) = 188.6, p<.001.
Figure 2: Accuracy rates for sum-larger and number-larger problems in Experiment 1, by ratio. Error bars show
±1 SE of the mean.
Discussion
The results from Experiment 1 clearly showed that participants were able to accurately
tackle non-symbolic addition tasks at accuracies well above chance. Furthermore we found
the so-called ratio-effect: accuracies on problems of ratio 5:6 were lower than those of ratio
5:8. Surprisingly, however, we also found an unpredicted main effect of ratio-direction:
number-larger problems had higher accuracy rates than sum-larger problems. Note however,
that like earlier researchers (Barth et al. 2006) we did not counterbalance the order in which
the numerosities were presented (the sum was always presented first, followed by the
comparison number). Consequently, it is possible that this effect might be a consequence of a
confound in the experimental design.
Experiment 2
Our goal in Experiment 2 was to ask non-dyscalculic participants to tackle problems similar
to those used in Experiment 1, but where the numerosities were represented symbolically. If,
as we hypothesised, the ANS is used in such tasks we would expect to see similar patterns of
results as in Experiment 1.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 9
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Method
Twelve staff and students (five male) from the University of Nottingham (aged 19-37, M=24)
received £4 for participating in the study.
Participants were tested individually. Displays were presented on a 17'' Philips 170B
LCD placed at eye level, and were viewed from approximately 60cm away. Stimuli consisted
of two symbolic items, a sum and a comparison number, as shown in Figure 3.
Figure 2: A symbolic addition task.
Numerosities were from the same range as those used in Experiment 1. A total of 25
problems were used for each approximate ratio, with the task being to determine as rapidly as
possible which item (left or right) was numerically larger. Again six comparison ratios were
used, 8:5, 7:5, 6:5, 5:6, 5:7 and 5:8. Each problem appeared four times (once as each of a+b
vs. c, b+a vs c, c vs a+b and c vs b+a). Both responses and response times were again
recorded via the leftmost (left-larger) and rightmost (right-larger) buttons on a five-button
response box.
In summary, the experiment followed a 3 (approximate ratio: 8:5, 7:5, 6:5) × 2 (ratiodirection: number-larger, sum-larger) design. This yielded a total of 600 trials for each
participant. The experiment was preceded by a practice block of 10 trials.
Results
Mean accuracy rates for each ratio and ratio-direction are presented in Figure 4. These were
analyzed using a 3 (approximate ratio: 5:6, 5:7, 5:8) ×2 (ratio-direction: number-larger, sumlarger) repeated-measures ANOVA. There were main effects of ratio, F(2,22)=22.08, p<.001,
and ratio-direction F(1,11)=6.17, p=.030. Accuracy was higher for problems with the number
larger, M=97%, SD=2.6%, than the sum larger, M=93%, SD=5.2%. There was a significant
linear trend of ratio, F(1,11) = 6.17, p=.030.
Discussion
The data from the symbolic tasks we used in Experiment 2 followed essentially the
same pattern as that from the non-symbolic tasks we used in Experiment 1. We found the
ratio-effect characteristic of the ANS, as well as the unexpected main effect of ratio-direction
(this latter finding casts doubt upon the suggestion that it was the result of insufficient
counterbalancing, as all stimuli were presented simultaneously in Experiment 2). This pattern
of results is exactly what we would expect if the ANS was being used by participants in
Experiment 2 to help them tackle the symbolic addition and comparison tasks we used.
However, unlike with the non-symbolic tasks used in Experiment 1, participants had various
different strategies available to them when tackling the symbolic tasks in Experiment 2.
Might it be that adopting non-ANS strategies would result in this pattern of results? To test
for this possibility we conducted several additional analyses.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 10
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Figure 4: Accuracy rates for sum-larger and number-larger problems in Experiment 2, by ratio. Error bars show
±1 SE of the mean.
Space constraints prevent a detailed presentation of the statistical tests we used to rule
out the possibility that participants were using alternative strategies. Our method in each case
however, was to compare mean accuracy rates and response times on the subset of the
problems where the alternative strategy would have worked with those problems where it
would not have. If participants had consistently been using the given alternative strategy we
would have expected to see a difference in the means associated with these two subsets of
problems. In those cases where such a difference was also predicted by the ANS-based
account we instead looked to see whether performance was above chance on those problems
where the alternative strategy would not have worked. No evidence was found that any of the
following strategies were being used by participants:
• Rounding Down. A participant using this strategy would have added the two leftmost
digits of each of the addends and compared to the leftmost digit of the comparison
number.
• Comparison Number Range. Here a participant would have based their answer on the
size of the comparison number alone. If it was higher than the median comparison
number used across the experiment it would have been selected.
• Addend Range. Alternatively, participants could have based their answer on the size
of the largest addend. If it was above the median it would have been selected.
• Addend-Comparison Difference. Participants may simply have compared the size of
the largest addend with the comparison number, and assumed that the sum is largest
when that difference was low.
General Discussion
The ANS appears to be an inbuilt cognitive system that supports rapid – but
approximate – numerical calculations. The evolutionary benefit of such a system is clear, for
example one can easily imagine how a predator might find it advantageous to be able to
rapidly decide which of two groups of prey is the larger. But is the ability to connect the ANS
to formal symbolism a prerequisite for higher level mathematics? Here we explored the
suggestion that dyscalculia is the result of a disconnect between the ANS and symbolic
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 11
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
mathematics. If this were the case we would predict that participants tackling tasks where
numerosities are represented non-symbolically would show a similar pattern of behaviour to
those who tackle the equivalent tasks where the numerosities are represented with Arabic
numerals. This is exactly what we found.
The ANS-symbolism disconnect proposal is a strong hypothesis, and further work is
needed to investigate its potential. If it is correct, however, then it seems that an important
early goal for mathematics education is to form and strengthen connections between various
different representations of numerosity and the internal representations of the ANS (e.g.
Wilson et al. 2008).
References
American Psychiatric Association 2000. Diagnostic and statistical manual of mental disorders
(4th ed, text revision). Washington DC: Author.
Barth, H., N. Kanwisher and E. S. Spelke. 2003. The construction of large number
representations in adults. Cognition 86: 201-221.
Barth, H., K. La Mont, J. Lipton, S. Dehaene, N. Kanwisher and E. S. Spelke. 2006. Nonsymbolic arithmetic in adults and young children. Cognition 98: 199-222.
Butterworth, B. 2005. Developmental dyscalculia. In J. I. D. Campbell (Ed.), Handbook of
mathematical cognition (pp. 455-468). New York: Psychology Press.
Cordes, S., R. Gelman, C. R. Gallistel and J. Whalen. 2005. Abstract number and arithmetic
in preschool children. Proceedings of the National Academy of Sciences 102: 1411614121.
Dehaene, S. 1992. Varieties of numerical abilities. Cognition 44: 1-42.
Dehaene, S. 1997. The number sense. Oxford: Oxford University Press.
Department for Education and Skills 2001. Guidance to support pupils with dyslexia and
dyscalculia (No. DfES 0512/2001). London: DfES.
Gilmore, C. K., S. McCarthy and E. S. Spelke. 2007. Symbolic arithmetic knowledge without
instruction. Nature 447: 589-591.
Moyer, R. S., and T. K. Landauer. 1967. Time required for judgements of numerical
inequality. Nature 215: 1519-1520.
Pica, P., C. Lemer, V. Izard, and S. Dehaene. 2004. Exact and approximate arithmetic in an
Amazonian indigene group. Science 306: 499-503.
Rousselle, L., and M. P. Noël. 2007. Basic numerical skills in children with mathematics
learning disabilities: a comparison of symbolic vs. non-symbolic number magnitude
processing. Cognition 102: 361–395.
Wilson, A. J., S. Dehaene, P. Pinel, S. K. Revkin, L. Cohen and D. Cohen. 2006. Principles
underlying the design of “The Number Race”, an adaptive computer game for
mediation of dyscalculia. Behavioral and Brain Functions 2: #20.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 12
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The T-shirt task: Using a mathematical task as a means to get insights into the nature of
the collaboration between in-service teachers and researchers
Claire Vaugelade Berg
University of Agder, Kristiansand, Norway
By following a mathematical task, from its design by researchers to its implementation
by a teacher, it is possible to get some insights into the collaboration between researchers
and teachers. Activity Theory is used as a theoretical approach in this research.
Keywords: designing and implementing tasks, collaboration between researchers
and teachers, activity theory.
The research setting
The research reported here examines the way a mathematical task is transformed and
adapted, from its design among researchers to its implementation by a teacher in lower
secondary school level. This research is situated within an ongoing research project at the
University of Agder (UiA) called TBM (Teaching Better Mathematics). The name of the
project reflects two goals: the first one concerns Teaching (Better Mathematics) and aims at
developing better understanding of, and competency in mathematics for pupils in schools.
The second one concerns (Teaching Better) Mathematics and aims at exploring better
teaching approaches in order to achieve the first aim. The research involves collaboration
between researchers from university and teachers working at different levels, from
kindergarten up to upper secondary school. Within the project, we collaborate with 4
kindergarten, 6 primary schools and lower secondary schools, and 3 upper secondary schools.
The collaboration with the teachers is organised around workshops which happen
approximately once a month and consist of a plenary presentation of a theme within
mathematics and some group work where the teachers have the opportunity to work
collaboratively with colleagues at the same school level. In addition, within each school, a
group of two or three teachers (called the TBM group) is responsible for the continuity
between the work done at the university during the workshops and the teaching of
mathematics in their respective schools. An important feature of this research is the
recognition of the researchers and the teachers as working together as co-learners and getting
the opportunity to develop a better understanding of each others “world and its connections to
institutions and schooling” (Wagner 1997, 16).
The centrality of inquiry
The idea of inquiry plays a crucial role in the TBM project, as we consider that
through the project we, as researchers, are able to address inquiry at three different levels
(Jaworski 2006): at a first level, as pupils engage with a task and inquire into the
mathematics, at a second level, as teachers engage with inquiring into the developmental
process of planning for their teaching, and at a third level, as researchers engage with
inquiring into the research process of systematically exploring the developmental process and
practices as presented at the two first levels. Thereby, all participants of the TBM project
engage with inquiring into how to improve mathematics learning and teaching in classrooms.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 13
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Furthermore, we consider that inquiry could form a basis for the teachers’ teaching practice in
mathematics.
Activity Theory as a means to characterize development
In our project we use Activity Theory as a means to describe and characterise both the
researchers’ and the teachers’ development. Within Activity Theory, the idea of “activity”
endorses a precise meaning: human activity is understood as directed by a motive and is
firmly rooted in actions and goals. Furthermore, these actions are carried out through
operations and conditions (Engeström 1999). Thereby, we see our motive within the TBM
project as being to engage, collaboratively with teachers, in inquiry about teaching and
learning of mathematics in order to improve pupils’ achievement in mathematics. Our actions
consist of organising workshops and designing suitable mathematical tasks, while the notions
of operations and conditions refer to searching and collecting ideas and adapting these to the
workshop environment and classroom situation. In order to deepen the ideas of actions and
operations, I propose to introduce the theoretical constructs of “didactical aim” and
“pedagogical means” (Berg 2009). Didactical aim refers to the choice of a particular area or
subject-matter as for example symmetry, algebra or proportionality, while pedagogical means
refers to a task which is chosen and used in order to address the chosen didactical aim. Here I
consider that the construct of “didactical aim” as a useful theoretical tool enabling me to pull
out, articulate, and make visible central issues in relation to both the collaboration with
teachers, and more specifically, the design and preparation of workshops, and the teachers’
planning and preparation of their own teaching. Thereby, “didactical aim” relates both to the
researchers’ actions (preparation of workshops) as it emphasises which mathematical goal the
researchers plan to address during the workshops, and to the teachers’ actions (preparation of
teaching) as it emphasises which mathematical goal the teachers plan to address during their
teaching. Likewise, “pedagogical means” serves as a theoretical tool describing the result of
the processes of choosing, transforming and adapting a particular task to a specific social
setting in order to address a chosen didactical aim. Thereby, “pedagogical means” refers both
to the researchers’ operations in the sense of assembling a set of tasks and adapting these to
the group session during the workshops, and to the teachers’ operations as they prepare and
adapt a particular mathematical task to their own classes. Therefore I consider that “by
presenting a particular task within a specific social setting, a didactician creates a
mathematical environment whose characteristics depends both on the mathematical task and
on the setting” (Berg 2009, 103). In this article, I present and compare two mathematical
environments: the first one relates to a specific task prepared by the researchers in order to
engage collaboratively with teachers during a particular workshop, the second one relates to
how a teacher implement this particular task in his teaching.
Focusing on a specific task
In my current research, I follow a particular task, the T-shirt task, from its design
during a meeting among researchers at UiA (TBM meeting, 26.11.08), its presentation to the
teachers during a workshop (03.12.08), and to its implementation in a primary school
(11.12.08) and in a lower secondary school (05.05.09). The rationale for choosing this
particularly task is the following: in December 08, a teacher from primary school contacted
me and invited me to follow her teaching as she wanted to implement a task from a previous
workshop in her teaching. Similarly, in May 09, another teacher (Per) from lower secondary
school contacted me as he planned to implement a task from a previous workshop in his
teaching. The fact that these teachers were referring to the same mathematical task (the T-
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 14
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
shirt task) encouraged me to focus my research on the processes behind the design and
implementation of this specific task. Because of space limitation, I focus on the results of the
analysis of the mathematical environment created by Per.
The TBM meeting
The T-shirt task was elaborated during the TBM meeting on November 26th . During
that meeting, the discussion among the group of researchers was focusing on how to address
“communication in mathematics classroom”. This theme had been chosen in advance by the
TBM group of the different schools. While exploring different possibilities for addressing this
subject, we decided to contextualise the discussion and to address “communication in
mathematics classroom” through engaging with a particular task. We chose the T-shirt task as
we agreed that this specific task offered a rich approach to communication. As one of the
participants emphasised:
… in communicating mathematics, questions are a far more effective way of
communicating than telling. In order to make sense of mathematical knowledge,
pupils need to take the responsibility for exploring which means questioning the
teacher, questioning others. The fundamental aspect about communication is
questioning. (TBM meeting, 26.11.08, translated from Norwegian by the author)
Thereby, our group decided to focus on the ability to ask “good” questions, in the
sense of engaging with task by inquiring into the mathematics (Jaworski 2006). The T-shirt
task was elaborated in the following way: the context is a phone call where one person has to
explain to another one the motive of a logo to be reproduced on a T-shirt (see Figure 1).
Figure 1: The T-shirt task presented within a grid system behind the logo
The workshop (03.12.08)
Usually the workshops are organised according to the following pattern: first one of
the didacticians from the university gives a plenary presentation on a chosen theme. Second, a
group session is organised where all participants are divided into groups according to the
level at which they teach. Finally, all participants gather together in order to exchange
experiences from the group sessions. The title of the workshop on December 3rd was “To ask
good questions in mathematics”, and it was one of the didactician from the university who
had the responsibility to introduce to the participants of the workshop (teachers from different
school levels, from kindergarten to upper secondary school) ideas related to questioning and
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 15
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
communicating in mathematics and to link these to the idea of “inquiry”. During the group
session, all teachers engaged with the T-shirt task and discussed different approaches to it
with emphasis on the kind of questions which are relevant to ask in order to reproduce the
logo in an accurate way. In addition, they discussed the possibility of integrating, changing
and adapting this task to their respective classes.
Summary of the design process
Looking at the design process of the T-shirt task, the discussion during the TBM
meeting concerned “Communication in mathematics” and we decided to emphasise
questioning and its relation to inquiry. Furthermore, we chose to contextualise the discussion
through presenting a mathematical task and the T-shirt task was selected as a relevant
example. Using the theoretical constructs from Activity Theory, I understand the idea of
“actions” as referring both to the researchers’ preparation of workshops, and to the teachers’
preparation of their own teaching. These actions are linked to the didactical aims chosen by
the researchers and by the teachers. Furthermore, I consider the construct of “operations” as
referring both to how the researchers assemble a set of tasks and adapt these to group sessions
during the workshops, and to how the teachers prepare and adapt tasks to their own classes.
Likewise, I consider that these operations offer a description both of the researchers’ and of
the teachers’ pedagogical means.
Implementation of the T-shirt task in lower secondary school
Before observing how the teacher, from lower secondary school, implemented the Tshirt task in his class, I had the opportunity to interview him and, thereby, to make visible his
aims for the teaching period. Likewise, I conducted an interview right after the teaching
period in order to summarise and evaluate it with the teacher. In this article I focus on the
interview before the teaching period.
Visiting a lower secondary school: in class with Per
On May 5th, I had the opportunity to visit Per and to observe his teaching in grade 8.
During the interview he explained the rationale for implementing the T-shirt task in his class
and he emphasised particularly on the following aspects:
From the curriculum, there are first of all two aspects which I would like to have as
goals for my teaching, it is the use of coordinate system, and the second is the
introduction of functions… And you can say, what I want to emphasise is
communication, I would like the pupils to have an understanding of how one
communicates in mathematics.
From his utterance, it seems that Per decided to implement the T-shirt task in his class
since he could recognise the possibility to address two aspects from the curriculum through
this task. The first one refers to the use of coordinate system, the second one relates to the
introduction of functions. In addition, Per emphasised communication in mathematics. It was
the first aspect (use of coordinate system) which was emphasised during the teaching period I
observed (right after the interview). Per organised his teaching by dividing the lesson into two
parts. During the first part, he asked one pupil to be responsible for explaining the logo of the
T-shirt task to another pupil who was sitting behind a blackboard and could not see the logo.
Here Per had prepared in advance a slightly different representation of the logo: the logo of
the T-shirt was drawn without a grid system behind it and he presented this version to the
pupil. From the classroom observation, it is possible to follow how the pupil struggled to
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 16
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
explain the respective positions of the circle and triangles, as drawn on the logo. During the
second part of the lesson Per asked another pupil to be responsible for explaining the logo of
the T-shirt task, this time with the grid system (see Figure 1). Here it is possible to follow how
the pupil used the grid as a coordinate system and could refer to the circle and the triangles by
indicating the coordinates of particular points. During both exchanges (description with and
without a grid system) the rest of the pupils could follow how the communication of the logo
was influenced by having the opportunity or not to use a coordinate system.
My interpretation of Per’s way of implementing the T-shirt task is the following: as
the T-shirt task was introduced during the workshop of December 3rd, Per was able to identify
the possibility to address two aspects from the curriculum, the use of coordinate system and
the introduction of functions. I understand these as didactical aims which Per plan to address
during his teaching, or using the theoretical tools from Activity Theory, these didactical aims
are the “goals” for his “actions”. Thereby, during the lesson I observed, Per was addressing
the use of coordinate system as one of his didactical aims. At the same time, as he
implemented the T-shirt task by contrasting the presentation of the logo with and without the
grid system behind it, he was in a position of emphasising communication in mathematics.
Thereby, my understanding of Per’s teaching is that his didactical aim for that lesson was to
address the use of coordinate system and he modified and adapted the T-shirt task to his class
in order to achieve his didactical aim. Here Per’s “operations” were to produce the modified
version of the T-shirt task (presentation with and without the grid system behind the logo) and
therefore this new task acted as a pedagogical means which was chosen and used in order to
address the chosen didactical aim. In addition, this comparison allowed Per to emphasise on
communication in mathematics.
Comparing the researchers’ and Per’s didactical aims and pedagogical means
Looking back to the researchers’ elaboration and preparation of the workshop
comparing to the interview with Per before his teaching, it is possible to observe an inversion
between the didactical aims and pedagogical means:
Researchers
Per
Didactical aim
Communication
Coordinate system
Pedagogical means
T-shirt task
(coordinate system)
T-shirt task
(communication)
Table 1: Comparing the researchers’ and Per’s didactical aims and pedagogical means
A possible explanation for the observed inversion consists of taking into consideration
and recognising the fact that the teacher and the researchers belong to two different activity
systems. Furthermore, this recognition begs the following question: how can this inversion be
understood using the theoretical constructs available within Activity Theory? As explained
earlier, I conceptualised our research group at UiA in terms of activity system where our
activity is motivated by a desire to engage, collaboratively with teachers, in inquiry about
teaching and learning of mathematics as a means to improve pupils’ achievement in
mathematics. The goals of our actions, consisting of workshops and school visits, were to
initiate and make possible the collaboration with teachers. In addition, according to
Engeström (1999), an activity system is defined by the “rules”, the “community”, and the
“division of labour” followed by the activity system. Concerning our research group, the
community of researchers consists of 5-6 persons working within the TBM project, and the
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 17
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
division of labour is made visible as we prepare the different workshops and are responsible
for school visits. Concerning the dimension called “rules”, one aspect of it is visible in our
community as our group takes into consideration the wishes emerging from the teachers’
TBM group at each school in relation to the choice of didactical aims. The teachers usually
suggest several themes to us and from these ideas we discuss how to organise the different
workshops in a coherent way. Thereby, within our activity system, one of the rules is to
recognise and elaborate on the different themes proposed by the teachers. Communication in
mathematics is one example of the teachers’ suggestions. Looking at the teachers’ activity
system at their respective schools, I understand the motive for their activity as creating
opportunities for engagement with mathematics, and offering critical guidance for what
mathematics achievement means. Concerning the goals of their actions (teaching), I see these
as being the organisation of pupils’ participation into mathematics classrooms. At the
teachers’ schools, the community consists of all teachers and colleagues working within the
administration, and the division of labour is clearly decided by the head teacher within each
school. Concerning the dimension “rules”, one of the constraints for the teachers is to follow
the curriculum. It seems that it is this rule which became visible through Per’s utterance.
Thereby, I consider that by following a specific mathematical task from its design by
researchers to its implementation in Per’s class, the research reported here enables me to
compare and to observe an inversion between the researchers’ and Per’s didactical aims and
pedagogical means. A possible explanation for this inversion consists of recognising the
researchers’ and the teachers’ communities as belonging to two different activity systems
each of them having different rules. I argue that this recognition helps us, as researchers, to
get deeper understanding of each others’ world (Wagner 1997).
References
Berg, C. V. 2009. Developing algebraic thinking in a community of inquiry: Collaboration
between three techers and a didactician. Doctoral Dissertation at the University of
Agder. Kristiansand, Norway: University of Agder.
Engeström, Y. 1999. Activity theory and Individual social transformation. In Perspectives on
activity theory, ed.Y. Engeström, R. Miettinen and R-L Punamäki, 19-38. Cambridge:
Cambridge University Press.
Jaworski, B. 2006. Theory and practice in mathematics teaching development: Critical
inquiry as a mode of learning in teaching. Journal of Mathematics Teacher Education,
9(2), 187-211
Wagner, J. 1997. The unavoidable intervention of educational research: A framework for
reconsidering researcher-practitioner cooperation. Educational Researcher, 26(7), 1322
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 18
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Motivating Years 12 and 13 study of Mathematics: researching pathways in Year 11
Rod Bond, David Green and Barbara Jaworski
Mathematics Education Centre – Loughborough University
We report on a collaboration, between 4 teachers in 4 schools and a university team of 3,
over a period of 21 months, to enthuse Year 11 students (taken from the top 25% ability
range) about mathematics and encourage their further study of mathematics in Years 12
and 13. Each school used a different pathway to achieve these goals: this involved
acceleration, enrichment, the Free Standing Mathematics Qualification or an early start to
A level. The research was developmental in both studying the practices and processes
involved while contributing to teachers’ continuing professional development in
mathematics.
The project
Funded by the NCETM1, this project involved a 21 month investigation into how teachers can
motivate and enthuse able mathematics students (taken from the top 25% ability range as
judged by tests and examinations) by developing pedagogy in different approaches at KS4. It
was motivated by a desire to encourage more young people to enjoy mathematics and take it
further in their studies. From an NCETM perspective, the professional development of the
teachers concerned was paramount. To achieve these various aims, the project took a
developmental research approach which involved 4 teachers in 4 schools and 3 university
academics. Project activity included: For the teachers
• Creating a mathematical pathway in Year 11, designing and delivering the related
course, and studying its progress and outcomes;
• Participating in collaborative activity through meetings at the university and visits to
each others’ schools;
• Reflecting on, evaluating and reporting outcomes.
For the academics
• Working with the teachers to encourage and study the creation and implementation of a
pathway;
• Collecting data to chart progress and evaluate outcomes;
• Conducting analyses and reporting on outcomes.
Methodology
The project was conceived by Bond as a result of contacts with headteachers who wanted to
improve exam performance and Year 12 uptake, and discussions with Heads of Mathematics
faced with developing strategies for students who had taken GCSE (General Certificate of
Secondary Education) in mathematics at the end of Year 10. It was clear that schools used a
range of approaches and questions arose as to ways in which such approaches contributed to
achieving project aims. Four schools agreed to participate in the project, each with a lead
teacher and a different chosen pathway, as follows.
A. Entry for GCSE at the end of Year 10 followed by A/S Mathematics in Year 11
1
National Centre for Excellence in the Teaching of Mathematics.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 19
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
B. Entry for GCSE at the end of Year 11 and the Free Standing Mathematics Qualification
(FSMQ) at the end of Year 11
C. Entry for GCSE at the end of Year 10 followed by the FSMQ in Year 11
D. Entry for GCSE at the end of Year 11 with no additional qualification offered, but with
the course enhanced by practical work and ICT
The project provided, for each school, two hours per week off timetable for the lead teacher,
with release time for regular meetings as a group at the university to discuss progress and
share ideas, and £600 to buy resources of their choice. It provided also the support of a
Project Officer and staff at the university and the guidance of an advisory group which would
monitor the direction of the project.
The project used a developmental research methodology such that research activity
contributed to promotion of development (Jaworski, 2008). Research was designed to explore
the nature and outcomes of each particular pathway, and the associated development of the
lead teacher. Each lead teacher organized the activity of the project within their school,
liaising with their mathematics department as appropriate and teaching their own class of
students throughout the year. In two schools, two teachers were involved in the teaching.
Teachers were encouraged to keep a record of progress according to their aims in the project.
The Project Director (Bond) and the Project Officer (Green) liaised closely with schools to
support the initiation of the project and to collect relevant data (see below). Project meetings
of the four lead teachers and the university team were organised to take place at the university
9 times during the life of the project. Teachers were encouraged to reflect on activity and
progress and to discuss issues and concerns; the university team encouraged reflection and
asked probing questions. The team asked teachers to state research questions for their own
pathway and discussion in the meetings allowed these questions to be refined as the academic
year progressed. Towards the end of the project, theoretical perspectives were discussed to
enable teachers to consider the theories motivating their teaching and its development.
Data, both quantitative and qualitative, were collected throughout the project and
analysed by the university team. Quantitative data included school data, three student
questionnaires, three teacher questionnaires and public examination results. Data were also
collected from a fifth school to act as a ‘control’. Qualitative data included recordings of the
meetings (summarised), the developing research questions, two interviews with each of the
four teachers (first interviews transcribed), interviews with students in two schools, teachers’
written reports , and a diary (one teacher only).
Quantitative data were first coded by hand and entered into Microsoft Excel
spreadsheets for checking. Data from the second questionnaires, for both students and
teachers, were analysed and graphs produced using Excel. All other data were transferred to
SPSS for analysis and production of tables and graphs. Due to the small and highly specific
nature of the sample of schools, caution has been used in interpretation of these data. Results
which may appear significant (e.g. using a multi-dimensional chi-squared test on crosstabulated data) might well not generalise to a wider population or other circumstances. Our
hope is that teachers in other schools where circumstances seem similar to those in the project
schools will feel able to draw their own tentative conclusions and then to conduct research to
verify them.
Of the extensive qualitative data, the first set of interviews with teachers (June 2009,
transcribed) and the teachers’ written reports have so far provided the main data for analysis,
supported by the other forms of data. Analysis has involved a cyclic process of reading, rereading and categorisation of the data. This process is still ongoing and what we present here
are tentative initial categorisations. A detailed report has been written for the NCETM,
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 20
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
including detail of data and analyses and the full teacher reports. A copy can be obtained by
contacting one of the authors2.
Findings from the project
Quantitative analysis
School data, examination results and progression
Data collected from the four schools and the control school included information on students
numbers, gender, SATS and examination results. These data varied according to the nature of
the top set – for example in School A, where the top set was 1 out of 10, various results and
indices were not surprisingly higher than in the other three schools which had two or three
parallel four-class streams each with a top set (i.e. 1 out of 4). Also, the time allocated to
mathematics varied considerably between schools with School A having 5 hours per week
and School D only 2 hours and 20 minutes. The gender balance was quite even in all classes
except in School A (59% male) and School D (66% male). With regard to progression rates –
i.e. students transferring into Year 12 – complete data were hard to obtain due to transfers
between schools, students dropping out of Y12 mathematics courses at various points and
even students who leave school, only to return later. In almost all classes a decline is indicated
in uptake expectations between October 2008 and April 2009, and a further decline in the
actual uptake, after drop-outs. Results from the control school suggest that this decline is a
common feature to which the project made little difference.
Student questionnaires (initial and final)
These two questionnaires were identical, including: 12 questions on a 5 point scale to measure
perceptions of a) confidence, b) teacher supportiveness, and c) usefulness of mathematics for
themselves; and 10 questions on a 5 point scale to measure d) enjoyment and e) usefulness of
mathematics for society; each index was tested for reliability. Data and boxplots can be found
in our NCETM report.
There was no appreciable change in confidence levels (which were quite high)
between October 2008 and April 2009 although an increase in lower end outliers suggested
that some students were feeling examination pressure or starting to feel that mathematics was
not for them. Enjoyment levels remained quite high overall. Project classes had an increased
spread leading to a higher median than the control. Student views on teacher supportiveness,
which were largely positive, increased for classes A and D. Perceptions of control classes
were lower. Students’ views on the usefulness of mathematics for themselves and for society
remained at a high level for both project schools and control. Students were largely graded at
SATS levels 7 and 8. An analysis was done for each of the 5 indices against the two SATS
levels. SATS-8 students demonstrated mainly high confidence levels and positive enjoyment
levels, whereas there was a wider range of confidence for SATS-7 students whose enjoyment
levels were generally lower though still mainly positive. Levels of perceived teacher
supportiveness were generally high for SATS-8, and lower, although still positive, for the
SATS-7 students. The Spearman’s rho correlation of confidence with teacher supportiveness
was much higher for SATS-7 than for SATS-8, suggesting the crucial role that (perceived)
teacher attitude can play for the SATS-7 students.
2
Corresponding author: R.M.Bond@lboro.ac.uk
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 21
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Project teachers’ predictions of their students’ perceptions
Half way through the 2008/9 academic year, a second, shorter questionnaire was completed
by students with questions 1 to 5 on 3 levels (a lot, a bit, not at all) and question 8 on 2 levels
(Yes/No). The main statements for their consideration were:
1. I am enjoying Year 11 mathematics
2. The course is demanding in terms of workload
3. I am being stretched mathematically
4. My understanding of mathematics has improved in Year 11
5. The course has inspired me to carry on studying mathematics next year
8 Could the teaching be improved?
At the same time, their teachers were asked to predict the response profiles they
expected from their classes. Findings and predictions on enjoyment matched fairly well, with
teachers only slightly overestimating. However, teachers greatly overestimated their students’
workload. We noted interestingly that class B (taking GCSE and FSMQ simultaneously)
recorded a high level of enjoyment and also the highest incidence of the workload being very
demanding. Teachers substantially overestimated the intellectual demand level (“I am being
stretched mathematically”), although almost all students were finding the work challenging to
some degree. Almost all students reported some improvement in understanding, and teachers’
estimates were close to students’ recordings. Teachers slightly overestimated the numbers of
students who said they intended to continue to Year 12, although the overestimate was most
marked on just one school (C) which ultimately had a lower uptake than the other schools.
Regarding Question 8, remarkably, the prediction of Teacher A, that 100% of students would
say that teaching could not be improved was exactly correct. Teachers B and C rather
overemphasised the number who would say “yes”. Teacher D estimated his students’
responses very accurately.
Teaching styles
At the beginning and end of the academic year, teachers were asked to complete a
questionnaire designed to assess the strengths of three factors comprising their teaching
styles: transmission, connection and discovery (Askew et al, 1997). Briefly, transmission
teaching views mathematics as a body of knowledge and skills to be passed on from teacher
to student; connectionist teaching views mathematics as an interconnected body of ideas and
reasoning processes which the teacher and student construct together, and discovery teaching
views mathematics as a personal construction of the student. Responses indicated that
Teacher A moved from being just within the connectionist zone, to being substantially within
this zone; teachers B and C stayed in a very similar positions within their zone (B in the
transmission zone and C in the discovery zone) and D moved from being substantially within
the transmission zone to the boundary between transmission and discovery. Thus the project
seemed to have the most marked effect on Teacher D.
Qualitative analysis
While the quantitative data pointed mainly towards students’ perceptions and achievements,
and their teachers’ associated expectations, the qualitative data pointed mainly towards
teachers’ perceptions of their teaching and its development through the year. We present here
the broad picture of findings so far.
Teachers’ goals for the project
Teachers were asked to say something about their goals for the project (or the goals of their
school). Teacher A, who had taught his class (top set of 10 sets) since Year 9 and whose class
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 22
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
had already taken GCSE in Year 10, wanted to see if taking A/S level in Year 11 was “doable”. Could he take a class of 30 students through half an A level in Year 11? Teacher B,
who was new to the school and who had had the project thrust onto her in her first year in the
school, presented a school perspective. They wanted to “do something extra” with their “more
able” students (2 parallel top sets), “give them more experience with maths”. They wanted “to
push them a bit further than GCSE” (offering also the FSMQ). Teacher C, whose students had
taken GCSE at the end of Year 10 (some were trying to raise their GCSE grades) said that he
was looking for a course that would “stretch our more able pupils”, something that would take
them a bit further, give them insight into what A level maths would be like”. Thus the school
offered the FSMQ to parallel top sets alongside GCSE retakes. Teacher D also wanted to
“stretch our more able students” but without fast-tracking. The students were taking GCSE in
Year 11. He wanted to introduce them to mathematics that would give them a taste of A level,
and would reinforce and be complementary to GCSE, but without offering a separate
qualification.
Thus the goals expressed by the teachers fitted well with the stated goals of the
project. The pathways were different; two schools having already taken Year 10 classes
through GCSE were looking for a suitable course for Year 11; two schools were in the
process of preparing students for GCSE at the end of Year 11. One school in each category
decided to take on the FSMQ course, one (B) alongside GCSE in Year 11, and one (C) in
parallel, for some students, with some retaking GCSE and seeking higher grades. We now
report on particular issues or outcomes.
DO-ability – what works
Teacher A asked whether the plan to take 30 Year 11 students through half an A level in one
year was “do-able”. His research showed that it was do-able, and outcomes from the process
indicated considerable success as detailed in his diary. The idea of ‘do-ability’ seemed to
permeate the rhetoric of the school research for all four teachers. “What works” was a
common focus. Teachers initially all focused on what they would do, or what they had done,
and the extent to which it worked relative to their goals in the project and the context of their
school. What works included: planning for the classroom, types of activities and tasks, how
students respond, what issues arise and what all of this looks like in practice. In the early
stages, not much was said about what teachers learned or could learn from the project. The
focus for all of them was on what students would or could gain from the planned activity, and
on outcomes in terms of examination results and achievement of student targets.
Preparing for teaching
Teachers put a lot of time into their planning of activities and resources for their students.
This involved thinking hard about what would be interesting and motivating for students.
Teacher B, who was new to her school and to teaching A level mathematics, spent
considerable time working on mathematics herself. Teacher A prepared songs and quiz-based
tasks related to curriculum areas. Teacher D prepared computer-based activities such as
spreadsheets for numerical differentiation and activities with graphical calculators. Teacher C
engaged in collaborative planning with a colleague to find “different ways of doing the
topics”. All produced video-recorded examples of innovative practice for sharing in project
meetings
Valuable to work with colleagues (in school and in the project)
Teachers emphasised the importance for them of sharing their thinking, planning and
reflecting with colleagues, both in terms of sharing ideas and gaining ideas from others, and
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 23
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
in terms of gaining support where there were issues and problems. Teachers B and C valued
opportunities to share ideas with and/or gain support from colleagues in their own school. All
valued the opportunity to share ideas and issues with one other. Schools A, B and C made
choices at department and school level, whereas in School D it was largely the choice of the
one teacher, albeit supported by his Principal. Disappointingly, for Teacher D, his department
showed little interest.
Time pressure
Teachers B, C and D emphasized pressure of time on what they were able to achieve. The
time allocation for mathematics in School A was more generous than the other schools, with
School D the least generous. While we are aware that teachers generally experience time
pressure and that there are competing demands on allocation of time to subjects within a
school, it was clear that achieving the goals of this project was considerably circumscribed by
time factors and associated pressures.
Teachers’ learning through reflection
Project meetings put emphasis on what the teachers were learning from their activity and its
outcomes. For all of the teachers this seemed to require a refocusing of their attention and a
use of different language to describe what they were experiencing. Reflecting on their
experience and offering some analysis of it in terms of their own learning in a project
meeting required a more personal introspection. However, the supportive nature of project
meetings encouraged the teachers to share personal issues and concerns. All expressed growth
of understanding of, and confidence in, the new activity. The teachers commented overtly on
the value of project time in school in which to reflect, and the nature and outcomes of this
reflection: e.g., “I have certainly done more reflecting on what has happened.… You start
thinking about what would you do differently next time. I have also been trying to notice
when something has been successful and then try to come up with something similar next
time … ways of teaching … which would work in a similar way, so yes I have developed that
way”.
In conclusion
Perceptions of pathway success were strongly related to school factors such as time devoted
to mathematics and the degree of support for the project within a school. Schools A, B and D
indicated their intention to continue the same pathway in future years. School C saw problems
between students retaking GCSE and coping with the demands of the FSMQ. Early entry for
GCSE is being discontinued. Further research can usefully explore whether these findings
accord with practices more widely.
References
Askew, M., Brown, M., Rhodes, V., Johnson, D., & Wiliam, D. (1997). Effective Teachers of
Numeracy. London: King’s College, London.
Jaworski, B. (2006). Developmental research in mathematics teaching and learning:
Developing learning communities based on inquiry and design. In P. Liljedahl (Ed.),
Proceedings of the 2006 annual meeting of the Canadian Mathematics Education
Study Group. Calgary, Canada: University of Calgary.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 24
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Computer Based Revision
Edmund Furse
Swansea Metropolitan University
A computer system on the web known as xplus12 has been developed which supports
KS3 revision in Number and Algebra. This has been evaluated with a year 8 top-set
class on two trials in Number with a little Algebra. Since all the data is stored in a
database including timings and all student attempts at answers, it is possible to identify a
number of behaviour patterns among the pupils. These include the identification of
groups such as "rubbishers" and "rushers" who essentially abuse the system. Although an
effect size of 0.9 was found in the first trial it was not statistically significant due to the
low numbers of pupils completing the test, and no improvement was found in the second
trial. Misconceptions were also identified by the system. A number of suggestions are
discussed for improvements of the system including improved examples with animation
and explaining answers. Also included are techniques to handle rubbishers and rushers.
Keywords: web, revision, computer based learning
Introduction
The use of computer systems to assist in the teaching of mathematics is growing worldwide,
especially in the USA. Anderson's Cognitive Tutor (Algebra) from Carnegie Melon
University is installed in over 2000 schools. Computer remedial teaching of algebra has been
a growing market in American Universities for some time, and in the last few years this is
now filtering down into schools. It is convenient to use computers for remedial teaching
since the pupils can do it at their own pace and institutions may use cheaper staff.
Here we report on an evaluation of the xplus12 revision system that runs on the web
(www.xplus12.com). The system has been developed over a number of years and uses AI
techniques both in the automatic synthesis of worksheets, and in sophisticated answer
checking. The development system is built in LISP and Java, and the delivery system runs on
all computer platforms that support Java. The website has sections on worksheets, starters,
games and revision and has over 40 worksheets. A unique property of the system is automatic
transition from type-in questions into multiple-choice questions after a suitable number of
wrong attempts. Most systems available only support multiple-choice questions but these are
too reliant on recognition memory. But the danger of type-in questions is that a pupil can get
stuck on a question and not be able to progress.
Mathematics is an ideal subject for a computer based learning system since the
computer is able to correctly check if a pupil's answer is correct. This can even be done for
algebraic answers by use of suitable AI techniques. Most systems fudge this and cannot
handle all the possible algebraic answers. This study finds an impact of the revision software
in xplus12 but it is not statistically significant. Too many pupils had to be excluded from the
trials due to non-completion of the test or writing rubbish. The paper discusses in particular
two important groups of students: "rubbishers" and "rushers". The discussion section looks to
a number of improvements of the system, which should make learning more effective.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 25
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Literature Review
Several studies have been done on the effectiveness of the use of computers in the teaching
and learning of mathematics. The results appear to be mixed, with some finding no effect and
others a small positive effect. Tienken and Wilson (2007) have a review of the impact of CAI
on mathematics achievement. They found an effect size of 0.12, which was statistically
significant p < 0.05. Whereas, The Electronic Education Report (2007) reports on a large
study of seven maths computer packages and found none had a statistically significant
improvement on performance. Zhu and Polianskaia (2007) provide a comparison between
traditional lecture and computer-mediated instruction and find no statistically significant
difference over a ten-year period. Paper and pencil was more effective. Clearly computer
systems differ widely in many aspects and this has impacts on their effectiveness. One of the
highest effect sizes (of 1) is reported by Anderson (1995) for his Cognitive Tutor in
Geometry, but surprisingly there is no difference in performance with his Algebra Tutor
between controls and students using the system. Anderson attributes this to poor transfer
from the computer system to paper and pencil solutions.
One of the problems of some of the positive results studies is that the students were
self-selected. This is particularly true for revision systems, like SAM Learning. It is also an
issue that is recognised in the literature, for example Biesenger & Crippen (2008). However,
several systems have been used with whole classes that then encounter problems of nonengagement by some students. Egan, Jefferies and Johal (2006) introduce the classification of
lurkers, workers and shirkers in an online teaching system. There have been a number of
studies of guessing including Beal et al (2008). Several systems, including Aleven et al
(2006) attempt to get the computer system to identify the students abusing the system and get
it to react appropriately; for example students who click through hints at speed are told to
slow down. Abusing the computer system is part of a more general problem of disaffected
pupils and those who avoid engaging with lessons (Dowson & McInerney 2001).
Most studies of the effectiveness of computer systems compare with a control group
that uses paper and pencil. But paper and pencil questions do not always easily translate into
a computer form. This is especially a problem with assessment and is one of the reasons for
the prevalence of multiple-choice type questions. There is also the problem of the assessment
of partially correct answers, since this is usually dealt with in an explicit way in paper-based
assessments; but it is difficult to do on computers. Lindsay (1999) compared the use of a
computer algebra system with paper and pencil techniques. Paper and pencil was more
effective. Threlfall et al (2007) analysed the differences between paper versions and computer
versions of KS2 and KS3 questions. Pupils sometimes do better with the computer versions
because there is more opportunity to explore different answers even though the system did not
give them feedback to tell them if it was correct or not. Ashton et al (2006) argue the
importance of computer systems being able to evaluate partially correct answers to
mathematics questions, but there are few if any that can do this. Computer systems may also
teach a subject in a completely different way to how a student may do it in class using pencil
and paper. For example, Anderson's Algebra Tutor is used in over 2000 American schools
but uses box diagrams to explain algebraic simplification which does not naturally translate to
a paper based method.
The quality of feedback from a teacher or a computer system is an important factor in
how well pupils learn. Many systems only give a correct/incorrect response, which may not
be very effective to aid learning unless they can eventually discover the correct result. Some
systems give the correct answer and this can be done in a three ways: (1) immediately when
the pupil gets the question wrong; (2) after a fixed number of attempts; (3) via a help system
which has run to the end of helpful advice and just tells the student the answer. A few
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 26
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
systems go further and give an explanation of why the answer is correct. There is also an
issue of the timing of feedback. Some systems, including MyMaths, do not give feedback
until all the questions have been answered. But Tallent-Runnels et al (2005) give a review on
how to teach online and identify the importance of prompt feedback.
Received wisdom in Tutoring Systems suggested that users should be given a large
amount of user control in navigation of the system (e.g. Wenger (1987)). But more recent
research suggests that weaker students need stronger control by the system in order to learn
effectively (Kopcha and Sullivan 2008). Ketamo and Alajaaski (2008) found that in the use of
a multiple-choice based system there was a lot of guessing and students did not have the skill
to choose appropriate materials to study. Similarly, Mezirow (1995) found that students
tended to choose topics to study that they were already familiar with rather than the ones they
needed to work on.
Computer System and Method
The xplus12 revision system covers a subset of KS3 Number and Algebra by means of
interactive worksheets. The pupil is given simple feedback to each question on typing a return
after the answer in the field. They are normally given five attempts, after which a multiplechoice version of the question replaces the type-in question.
When the pupil has completed the first worksheet and attempted all the questions, the
computer determines which are the pupil's weakest sections. The pupil then does a short
remedial worksheet for each topic they are weak on. In Number 3 these worksheets just
contain 4 interactive questions, but in Number/Algebra 2 there was an example as well.
Once the pupil has completed the remedial worksheets the pupil then goes on to the
last worksheet that is very similar to the first one. It has the same sections and also 2
questions per section. But there is no feedback and it works as a test.
Qualitative Study Results
A questionnaire was given to the pupils and there was also a discussion with four pupils. The
closed questions had a five point Likert scale response. The question "How easy do you find
xplus12 to use?" had a median of 4 as did the question "How much did xplus12 help you to
learn maths?". The results indicated that the pupils liked the xplus12 system, found it easy to
use, and felt they learned mathematics from it. There were also open questions such as "What
do you particularly like about xplus12?" and several pupils indicated they liked the type-in
questions turning into multiple-choice questions. On the negative side they thought there was
more need of colour, animation and more games.
Quantitative Study Results
First Trial
22 students did the revision exercises but only 14 students provided reliable data that can be
used for analysis. Eight students had to be excluded since they did not complete the revision
exercises and so it is not possible to compare pre and post test results. The pre test scores are
based on the pupils' first attempt at the questions. They may go on to attempt each question
several times, and since the system eventually ends up with a multiple-choice question they
eventually get every question correct. The post-test scores are just based on the answer since
they are only given one attempt with no feedback. The pre test mean was 12.9 and the posttest mean of 15.6 with a standard deviation of 3, which represents an effect size of 0.9.
Unfortunately this difference is not statistically significant at the 0.05 level.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 27
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
One of the problems encountered was that some pupils would attempt to get to the
multiple-choice version of a question as soon as possible by guessing or writing rubbish. Five
pupils were identified using this tactic and they may be labelled with the name "rubbishers".
Second Trial
In order to try to decrease the number of pupils "rubbishing", the mechanism for deciding
when to present the multiple-choice question was changed from a threshold of 5 attempts.
Instead, whenever the pupil answered a question too fast (less than a second) and had the
wrong answer, it was assumed they were guessing. The penalty for this was that the threshold
number of attempts was increased by one. Unfortunately this change was a disaster with two
pupils making 60 attempts at questions! Four pupils had at least 30 attempts at a question.
We can operationalise this by identifying pupils who have at least two questions with 15 or
more attempts as "rushers". An unfortunate consequence of the changed attempt threshold
policy was that the rushers spent a great deal of time on the pre test and most did not complete
the post-test. A few pupils also repeated the test thus invalidating the pre-scores.
One of the advantages of a computer system of this type is that it is possible to store
all the attempted answers a pupil makes to a question and to identify misconceptions. A
particular case of this occurred in question one of the pre test in BODMAS:
3 - 12 + 4 = ?
Eleven pupils gave the correct answer of -5 but ten students gave the answer of 13,
and three -13. Clearly there is a misconception here and pupils are taking the smaller number
from the larger one. Interestingly this misconception did not occur in the pre-test with the
BODMAS questions:
2+3×5=?
2+7-3=?
and since the post test question is easier, they are not matched questions.
Discussion
It is clear that there is a need for the computer system to be improved. Too many pupils are
disaffected and not enough learning is taking place. I take the theoretical stance that if pupils
want to learn, then it should be possible to design a computer system to help them to learn.
It can be argued that pupils behave as rushers and rubbishers because they cannot do
the questions however hard they try. There is evidence of rusher and rubbisher personality
types from the data since these behaviours persist, even across trials. Furthermore, several
pupils type rubbish immediately they start a question, indicating no effort to try to answer the
question correctly. On the other hand, no pupil is 100% a rusher/rubbisher (RR), but RRs are
likely to attempt questions they find easier. It appears that they have a low threshold of effort
and may resort to RR if their first 3 attempts are wrong. It is also true that rushers are likely
to exacerbate rubbishing since the computer system was postponing the multiple-choice
question as a result of their behaviour.
A simple solution to engaging and helping RRs is to provide interactive help. A later
version of the system provides optional interactive help after three attempts. This is also the
approach taken by Anderson’s Cognitive Tutor, although their help system is much more
complex. It is also clear that the multiple-choice version should be triggered always after five
attempts, and not penalise RRs by making it later.
Rushers can be slowed down by giving them a pop up message that tells them to slow
down and they have to click to continue. It is easy for the system to identify if the pupil is
answering questions too fast and incorrectly. It is more difficult to check if the pupil is
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 28
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
writing rubbish but long answers of consecutive numbers are usually indicative. A strong
rubbish checker would be safer than a weak one since a genuine pupil would take a dim view
of being challenged at writing rubbish.
It is also clear that not enough pupils are learning from their mistakes. It is not clear if
the pupils are actually benefiting from the multiple choice questions apart from letting them
get onto the next question rather than being permanently stuck. The system does not
currently record the timing data of the multiple-choice attempts but this could be changed.
Thus, again the pupils could be slowed down with a suitable dialog if they are clicking the
options too fast.
More importantly, the pupils need to have an explanation of the answer. It is not
sufficient just to be told the correct answer, and rushers are likely to just want to get onto the
next question rather than digest the answer. One simple solution is to leave the correct
answer and its explanation up for 5 seconds before they can proceed. A more sophisticated
approach is to use animated text. Also, it is probably desirable for the system to check if the
pupils think they understand their mistakes.
Animation has also been used to improve the examples. Direct observation in class
suggests that many pupils rush straight into answering the questions in a worksheet without
reading the examples. Animating the examples makes them more interesting and encourages
them to read them. Admittedly a disaffected pupil may not look at the animation, but there is
nothing else for them to do on the computer. The system now also asks them how well they
understand the example before proceeding to the questions. In principle, their answer can
inform future navigation.
It is straightforward to incorporate misconception handling in the system if the author
knows of suitable misconceptions. These can be included as alternative multiple choice
questions and checked for also during the type-in stage.
It is clear from the qualitative study that the colour and appearance of the system
needs to be improved. This came out clearly from the qualitative study, but it is possible that
deeper issues may be more important. They also wanted more games. Some might argue that
provision of games having completed the revision might provide motivation to do the work.
But there is a danger that this will encourage some pupils to rush their work to get to the
games. It is better to provide some form of intrinsic motivation within the system.
Conclusion
The xplus12 revision system improves learning but the quantitative results are not
statistically significant. The new improvements of animated examples, preventing rushing,
misconception handling and explanations of answers have been implemented and should
improve learning further. If more schools use the system then further research should prove
the advantages of these enhancements.
References
Aleven V., B. McLaren , I. Roll, and K. Koedinger. 2006. 'Toward Meta-cognitive Tutoring:
A Model of Help Seeking with a Cognitive Tutor. International Jnl. of Artificial
Intelligence in Education; Jul2006, Vol. 16 Issue 2, 101-128
Anderson J.R., A. T. Corbett, K. R. Koedinger and R. Pelletier. 1995. Cognitive Tutors:
Lessons Learned. Journal of the Learning Sciences; 1995, Vol. 4 Issue 2, p167, 41p
Ashton, H.S., C. E. Beevers, A. A. Korabinski, and M. A. Youngson. 2006. Incorporating
partial credit in computer-aided assessment of Mathematics in secondary education.
British Journal of Educational Technology, Vol 37, No 1, 2006, 93–119.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 29
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Beal, C. R., L. Qu, and H. Lee. 2008. Mathematics motivation and achievement as predictors
of high school students' guessing and help-seeking with instructional software.
Journal of Computer Assisted Learning, Dec2008, Vol. 24 Issue 6, p507-514
Biesenger K. & K. Crippen. 2008. The Impact of an Online Remediation Site on
Performance Related to High School Mathematics Proficiency. Journal of Computers
in Mathematics and Science Teaching (2008)27(1), 5-17
Dowson, M., & D. M. McInerney. 2001. Psychological Parameters of Students' Social and
Work Avoidance Goals: A Qualitative Investigation. Journal of Educational
Psychology, 93, 35-42.
Egan C., A. Jefferies and J. Johal. 2006. Providing Fine-grained Feedback Within an On-line
Learning System – Identifying the Workers from the Lurkers and the Shirkers. The
Electronic Journal of e-Learning Volume 4 Issue 1, 15-24.
Electronic Education Report. 2007. Study Reports Software Did Not Raise Test Scores in
Reading and Math, Simba Information, Stanford, April 20 2007, 1-3.
Kopcha T.J. and H. Sullivan. 2008. Learner preferences and prior knowledge.
In Learner-controlled computer-based instruction. Education Tech Research Dev (2008)
56:265–286.
Mezirow, J. 1995. Fostering critical reflection in adulthood : A guide to trans-formative and
emancipatory learning. Jossey-Bass.
Nguyen, D., J. Yi-Chuan, and D. Allen. 2006. The Impact of Web-Based Assessment and
Practice on Students' Mathematics Learning Attitudes. Jnl. of Computers in
Mathematics & Science Teaching; 2006, Vol. 25 Issue 3, 251-279
Tallent-Runnels M.K., S. Cooper, W. Y. Lan , J. A. Thomas, and C. Busby. 2005. How to
Teach Online: What the Research Says. Distance Learning Vol 2. Issue 1
Thelfall J., P. Pool, M. Homer and B. Swinnerton. 2007. Implicit aspects of paper and pencil
mathematics assessment that come to light through the use of the computer.
Educational Studies in Mathematics (2007) 66:335–348.
Tienken C.H. and M. J. Wilson. 2007. The Impact of Computer Assisted Instruction on
Seventh-Grade Students Mathematics Achievement. Planning and Changing, Vol. 38,
No. 3 & 4, 2007, 181-190.
Wenger, E. 1987. Artificial Intelligence and Tutoring Systems. Morgan Kaufmann.
Zhu, Q and G. Polianskaia. 2007. A Comparison of Traditional Lecture and Computermediated Instruction in Developmental Mathematics, Research and Teaching in
Developmental Education. Fall2007, Vol. 24 Issue 1, 63-82
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 30
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Children's Difficulties with Mathematical Word Problems.
Sara Gooding
University of Cambridge, UK
This article reports a study of the difficulties that primary school children experience
whilst tackling school mathematical word-problems. A case study of four Year 5 children
was conducted; this involved interviews which probed the children’s views of their own
difficulties and discussions with the children as they tackled word problems. The data
were qualitatively analysed using a thematic analysis approach based on categories of
difficulty identified from existing literature. Examples of transcripts and responses which
show the children experiencing difficulties are included, as well as the children's
opinions on their difficulties. My interpretation of these findings, including proposed
subcategories of difficulty, is also given. The report concludes with suggestions of
methods – subject to further research – that teachers may use to help children overcome
their difficulties with school mathematical word problems.
Background
Children’s poor performance with mathematical word problems is a trend that I became aware
of very early on in my teaching career and one that an interest has been taken in by many who
are involved in Mathematics education.
By looking at the existing literature on children’s difficulties with mathematical word
problems, I was able to gain a more detailed insight into the causes of children’s difficulties.
Using the evidence from existing research, I formulated five categories of difficulties that
children may experience whilst tackling mathematical word problems. These categories are
presented below.
Reading and Understanding the Language Used Within a Word Problem
Difficulties in this category involve children not being able to decode the words used in a
word problem, not comprehending a sentence, not understanding specific vocabulary and not
having confidence or the ability to concentrate when reading. (Ballew and Cunningham 1982:
Shuard and Rothery 1984: Cummins et al 1988: Bernardo 1999).
Recognising and Imagining the Context in Which a Word Problem is Set
These difficulties arise when children cannot imagine the context in which a word problem is
set or their approach is altered by the context in which the word problem is given. (Caldwell
and Goldin 1979: Nunes 1993).
Forming a Number Sentence to Represent the Mathematics Involved in the Word Problem
Children appear to find it harder to form a number sentence for some word problems
structures than others. These difficulties can result in children not being able to select a
calculation to perform or selecting an incorrect calculation. (Carey 1991: English 1998).
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 31
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Carrying Out the Mathematical Calculation
Difficulties can occur here with children’s selection of, and aptitude with calculation
strategies (for example formal algorithms, pencil and paper methods and calculators). The
context in which a word problem is given and the size of numbers involved can impact on
children’s choice of a calculation strategy. (Verschaffel, De Corte and Vierstraete 1999:
Nunes 1993: Anghileri 2001).
Interpreting the Answer in the Context of the Question
Children have been shown to not consider real-life factors and constraints when giving an
answer to word problems which can result in giving an answer that is impossible in the
context and therefore incorrect. (Verschaffel, De Corte and Lasure 1994; Wyndham and Säljö
1997; Cooper and Dunne 2000).
Method
Aims
The aims of the study were to establish whether difficulties within the identified
categories occur in English Primary Schools and, if they do, to find examples of children
experiencing difficulties within the categories. I hoped that examples of children experiencing
the range of difficulties may provide a resource for increasing teachers’ awareness of the
difficulties.
Data Collection
Four children were selected to take part in the study. These children were from the
Year 5 class that I taught; hence, I knew them well. They were selected on the criteria that
they were willing and able to discuss the mathematics that they were doing and were working
at a range of attainment levels in Mathematics lessons.
The first element of the data collection involved interviews on the children’s views of
difficulties they had experienced with mathematical word problems.
The second element involved the children working individually through sets of
equivalent word problems and discussing their processes and difficulties with me. There were
five sets of equivalent word problems that each child attempted. Each set was given in a
different condition, with a different form of help given in each. Each form of help
corresponded to one of the previously identified category of difficulty. For example I read the
word problem to the child, explained any vocabulary and simplified sentences in condition
one to correspond to the first category. I offered forms of help in the belief that if I gave a
specific type of help and children then solved a problem, I could identify where the original
difficulty lay and be aware of which kinds of help allow children to overcome certain types of
difficulty.
Data analysis
The first stage of data analysis involved analysing and coding interview transcripts
and recordings. Excerpts were coded under a category of difficulty if they showed opinions
on that difficulty, a child experiencing that difficulty, or a child competently completing a
process, therefore not having that difficulty. Any un-coded data were then checked for a need
for new categories or reported as ‘Other Findings’.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 32
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The second stage of data analysis involved analysing and coding all incorrect or no
responses to word problems. Transcripts and children’s jottings or workings were used to
code as to which difficulty prevented a correct answer being given.
Using the coded data, I was able to create subcategories within some of the five main
previously identified categories of difficulty.
Finally, I selected illustrative examples of children experiencing difficulties or giving
opinions on difficulties from each category and subcategory. Examples were picked using the
criteria of being typical and not extreme.
The Results and Discussion
•
•
•
•
•
•
•
The finalised categories and subcategories of difficulties formed are:
Reading and Comprehension
o Decoding the Words in a Word-Problem
o Understanding the Meaning of the Words and Sentences
Reading All of the Information
Distracting Information
Imagining the Context
Writing a Number Sentence
Carrying Out the Calculation
o Lack of Accurate Methods for Calculating
o Making a Mistake When Calculating
Interpreting the Answer in the Context of the Question
o Giving an Answer that is Possible or Likely
o Transferring an Answer into the Required Format
I have selected examples of children experiencing the above difficulties, or opinions on
difficulties from the categories ‘Reading and Comprehension’ and ‘Interpreting the Answer in
the Context of the Question’ to present below.
Reading All of the Information
The following example shows Liam giving an incorrect response to a word problem because
he has not read or comprehended all of the text in the question.
Figure 1, a word problem given to Liam and the transcript of the conversation that followed.
Liam:
This is a tricky one. I’m gonna have to say it’s the big pack
there.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 33
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Researcher:
Why is that then?
Liam:
Well the big pack is 80p, now half price so 40p. But these
packs cost 30 plus 30 plus 30 which is … 90 so yeah 40p.
Researcher:
How much did you say they would be [small packs]?
Liam:
90p and you could buy 2 big packs for that!
I acknowledge the possibility that Liam may not have understood the term ‘3 packets
for the price of 2’ but his confidence in his final answer leads me to believe that he has simply
not read or disregarded the information in the second star, leading him to get the correct
answer, but for the wrong reasons. His mistakes here may be related to how the word problem
is arranged on the page as all of the required information for the larger packet of pencils is in
the star, but the information for the smaller packets is not.
Distracting Information
Two comments related to this category are shown below:
Liam:
I like the little flashy.
Liam:
Hah Patrick! Either you watch SpongeBob or you have a kid that
watches SpongeBob.
Liam’s first comment is related to the illustrations in the word problem in Figure 1 and
his second to a word problem featuring a character called Patrick. Although Liam’s
observations and comments may not lead to incorrect answers, they show that his attention
may not be focused on the mathematics required to answer the problem and therefore these
distractions may cause him difficulties.
Giving an Answer that is Possible or Likely
Rachel was given a calculator to use to answer a word problem about the number of
children going on a school trip and gave the answer of ‘8.333333 children’.
Rachel:
Researcher:
8 .333333 so 8.3 dot [recurring].
Is that the number of children?
Rachel:
Yeah.
Here, although Rachel had carried out an appropriate calculation, she has not given a
correct answer and does not appear to realise, or consider it important that it is impossible to
have a third of a child on a school trip.
Another example of a child experiencing difficulties within this category is below:
Liam:
I can tell you this drink’s gonna cost loads more than a Mars bar.
It’s 35 for the Mars bars so take away 35 which would be …that would be
85. It’s £85. £85 for a fizzy drink.
Researcher:
That sounds a lot doesn’t it?
Liam:
Mmmm [in agreement] I’m not going to this shop.
Researcher:
Do you think that this sounds like a realistic shop then?
Liam:
Yes cos I’ve seen things like my Pokemon cards and they cost £3.99
for one pack.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 34
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Researcher:
So you think that there could be a shop that sells a fizzy drink for
£85?
Liam:
Yes.
Figure 2, a word problem given to Liam and the transcript of the conversation that followed.
Liam’s difficulties are caused by problems transferring between pounds and pence,
but his perception that £85 is a possible answer, even after attention was drawn to it, mean
that he could not identify his error and went on to give an incorrect answer. This example is
closely linked to the subcategory of ‘Transferring an Answer into the Required Format’ due
to the difficulties he had with dealing with money. Conversely to this example, when
answering an equivalent word problem to the one shown in Figure 2, Liam experienced
similar difficulties with knowing when values were in pounds or pence, but was able to use
his judgment of what is realistic to identify his error and go on to give a correct answer.
Transferring an Answer into the Required Format
The following excerpt shows Fiona having difficulties transferring her answer from a
decimal number into money:
Fiona:
Researcher:
Fiona:
Researcher:
Fiona:
He has to pay 15 for four.
But it’s half price isn’t it so how much is that?
Uuuuh 7.5.
OK. What’s that in money?
£7. 05.
Although Fiona was able to carry out the calculation of ‘15 ÷ 2 =’ correctly, she had
difficulties when trying to write that value in the standard format for money.
Other Findings
Other findings were also identified. These involved a child having difficulties because
he was not using jottings. When prompted to write numbers down as he was calculating
mentally, he was able to carry out a calculation more effectively.
Manipulatives were also shown to help a child to find a correct answer to a word
problem after previously not being able to for equivalent word problems. This shows that
manipulatives may be a useful tool for helping children to answer word problems marginally
beyond their current grasp.
Children also made comments which suggested that they were able to identify the
equivalence between word problems. This ability to recognise equivalence could imply that
showing children how to correctly find an answer to a word problem may help those children
to also solve equivalent word problems.
Recommendations
As a result of examining my findings alongside existing literature, I have compiled a
list of strategies that teachers and researchers could trial to help children to overcome
difficulties with mathematical word-problems:
• Encourage children to read the word-problems thoroughly;
• Teach children which kinds of information may be important;
• Ensure that children practise solving word-problems to allow them to be able to
recognise the structure of word-problems and therefore know when to use each
calculation;
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 35
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
•
•
•
•
•
Consider giving children manipulatives to support the solving of word-problems
currently beyond the scope of their ability;
Encourage children to write down their workings so that they do not become
unnecessarily confused;
Encourage children to check if their answer satisfies the criteria of a question. For
example if it is in the correct format;
Teach children to calculate with monetary values;
Encourage children to check if an answer is possible in the context of the question.
References
Anghileri, J. 2001. Development of Division Strategies for Year 5 Pupils in Ten English
Schools. British Educational Research Journal, 27(1), 85-103.
Ballew ,H., and Cunningham, J. 1982. Diagnosing Strengths and Weaknesses of Sixth-Grade
Students in Solving Word Problems. Journal for Research in Mathematics Education,
13(3), 202-210.
Bernardo, A. 1999. Overcoming Obstacles to Understanding and Solving Word Problems in
Mathematics. Educational Psychology, 19(2), 149-163.
Caldwell, J., and Goldin, G. 1979. Variables Affecting Word Problem Difficulty in
Elementary School Mathematics. Journal for Research in Mathematics Education,
10(5), 323-336.
Carey, D. 1991. Number Sentences: Linking Addition and Subtraction Word Problems and
Symbols. Journal for Research in Mathematics Education, 22(4), 266-280.
Cooper, B., and Dunne, M. 2000. Assessing Children’s Mathematical Knowledge: Social
Class, Sex and Problem-solving. Buckingham: Open University Press.
Cummins, D., Kintsch, W., Reusser, K. and Weimer, R. 1988. The Role of Understanding in
Solving Word Problems. Cognitive Psychology, 20, 405-438.
English, L. (1998). Children’s Problem Posing within Formal and Informal Contexts. Journal
for research in Mathematics Education, 29(1), 83-106.
Nunes, T., Schliemann, A.D., and Carraher, D.W. 1993. Mathematics in the Streets and in
Schools. Cambridge: Cambridge University Press.
Shuard, H., and Rothery, A. 1984. Children Reading Mathematics. London: John Murray
(Publishers) Ltd.
Verschaffel, L., De Corte, E and Lasure, S. 1994. Realistic Considerations in Mathematical
Modeling of School Arithmetic Word Problems. Learning and Instruction, 4(4), 273294.
Verschaffel, L., De Corte, E. and Vierstraete, H. 1999. Upper Elementary School Pupils’
Difficulties in Modelling and Solving Nonstandard Additive Word Problems
Involving Ordinal Numbers. Journal for Research In Mathematics Education, 30 (3),
265-285.
Wyndhamn, J., and Säljö, R. 1997 Word Problems and Mathematical Reasoning: A Study of
Children’s Mastery of Reference and Meaning in Textual Realities. Learning and
Instruction, 7(4), 361-382.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 36
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Some initial findings from a study of children’s understanding of the Order of
Operations
Carrie Headlam and Ted Graham
University of Plymouth
This paper presents some of the initial findings of a study into the strategies used by
children to solve arithmetic and algebraic problems requiring the appropriate use of the
order of arithmetic operations. The research has utilised graphics calculators which have
been programmed with Key Recorder Software as a data collection tool. This has
enabled the researchers to analyse the children’s approaches to some of the questions
posed by observing their calculator keystrokes. Interviews with both teachers and pupils
will be used to link the pupils’ strategies with the teaching methods used, and an initial
analysis of observed misconceptions has been carried out. Initially this study has
involved children in the UK and in Japan, where teaching methods differ substantially.
Introduction
The principle of the Order of Operations is a cornerstone of the understanding of arithmetic.
It is necessary in order to correctly perform arithmetic calculations and it is also an essential
prerequisite to the beginnings of the understanding of algebraic structure and the ability to
understand and apply the principles of algebraic convention correctly.
More fundamentally, it could be argued that an acknowledgement of the need for a
convention in arithmetic is an important step in the development of an appreciation of
mathematical convention in other areas of mathematics and indeed to the sense of learning the
language of mathematics, where standard rules are necessary in order to assist in clear
communication.
In considering the ways in which algebra is developed in different countries, the
relationship between algebra and arithmetic is usually characterised by the definition of
algebra as generalised arithmetic. Thus the view of ‘arithmetic then algebra’ dominates
school curricula in most countries. The reason for this, according to Lins and Kaput (2004)
can be found in the strong dominance of Piagetian constructivism. As algebra would require
formal thinking, while arithmetic would not, and as formal thinking would correspond to a
later developmental stage, algebra should come later than arithmetic. (page 50) This is seen
in the work of Kuchemann in Hart (ed) (1981) for the Concepts in Secondary Mathematics
and Science (CSMS) project who combined the view of algebra as generalised arithmetic with
the Piagetian developmental view. Lins and Kaput (2004) argue that the most visible result of
Kuchemann’s work is a reported link between different uses of letters in ‘generalised
arithmetic’ and Piaget’s levels of intellectual development. (page 50)
Hewitt (2003) considered students’ reading of formal algebraic notation and he
observed that many errors made by students could be accounted for by the strict left-to-right
reading of formally written arithmetic statements. He also considered how students read
word statements, acknowledging that expressing non left-to-right order in written words can
be problematic as well since words do not possess a set of notational conventions, such as
brackets (page 34)
In the National Strategies Secondary Mathematics Exemplification (DCSF, 2008) the
learning objective relating to this states that pupils should be taught to: Use the order of
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 37
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
operations, including brackets. (page 86) One example of a learning outcome is that a year 7
student should be able to perform the calculation
either mentally or using jottings. This objective is also linked to calculator methods,
with the expectation that a pupil should be taught to carry out more complex calculations
using the facilities on a calculator (page 108) and with the order of algebraic operations,
where pupils are expected to understand that algebraic operations follow the same
conventions and order as arithmetic operations. (page 114) The exemplification makes it
clear that pupils are expected to be able to use a scientific calculator efficiently when
evaluating more complex mixed operations.
The methods for teaching this topic can vary, but one common theme in some
countries is to use a mnemonic to aid the memorisation of the order of operations. In the UK
this is commonly BIDMAS or BODMAS:
Brackets Index Division Multiplication Addition Subtraction
Or
Brackets Of (Order) Division Multiplication Addition Subtraction
In the USA the mnemonic PEMDAS is commonly used:
Parentheses Exponents Multiplication Division Addition Subtraction
Clearly this may have its uses in remembering the “rule” once the concept has been
understood, but it is the clear understanding of the underlying principle and conventions that
enable it to be put into practice. This includes the understanding of index notation and the
recognition of a fraction for division.
Thus it is far from merely being a case of learning a mnemonic; a sound understanding
of mathematical notation and structure is required in order to carry out a calculation of the
type given in the National Strategies Mathematics Exemplification. It is this deep
understanding that lays the foundations for an understanding of algebraic structure.
It is interesting to note that the use of mnemonics is not referred to at all in the
National Strategies documentation, and yet many text books and other resources used in the
UK and in the USA encourage it.
In contrast, from conversations with Japanese teachers it would seem that mnemonics
are never used in Japan. Indeed in the Japan National Mathematics Program (2000) the order
of operations is not specifically referred to at all. The teaching methods are very didactic with
a large emphasis on whole-class teaching and repetition of questions, focusing on algebraic
structure.
The study: Context and Methods
The primary aim of this study is to examine the ways that pupils perform calculations which
require the correct use of the order of operations and to study the misconceptions that may
arise. One tool that will be utilised in order to carry out the research will be a piece of
software that was developed as a research tool by Texas Instruments in conjunction with the
University of Plymouth. This software is called the Key Recorder and can be loaded onto the
more recent models of the TI graphics calculator. It has been used as a data collection tool in
a small number of research projects (For example: Graham, Headlam, Honey, Sharp and
Smith, (2003), Berry, Graham and Smith (2003), Smith (2003) Berry, Graham and Smith
(2005), Berry, Graham and Smith (2006), Sheryn (2005), Sheryn (2006a), Sheryn (2006b)
Graham, Headlam., Sharp and Watson (2007) )
This study involves classes of children who have been taught about the Order of
Operations and who would therefore be expected to be able to perform calculations based
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 38
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
upon these principles. These children are in the age range 12 – 14 (Years 8 and 9 in UK
schools). The children complete worksheets of questions involving a variety of calculations,
some with and some without a calculator. For the calculator-based questions the children are
provided with a TI-84 graphics calculator which has the Key Record software running.
When the children’s work has been initially analysed, some of the children are then
interviewed in order to follow up and pursue questions that arise.
Initial Findings
A pilot study was carried out in the UK and in Japan. In each country one class of students
was involved. In the UK this was a class of 20 middle ability students in year 8 (age 13). The
Japanese class consisted of 33 mixed ability students aged 14.
Both classes had been taught the principles of the order of operations as part of their
scheme of work, and had also been taught simple algebraic conventions, including
substitution of letters for numbers in algebraic expressions.
In the pilot study the graphics calculators were not used; the children were given one
worksheet to complete without using a calculator. As a result of this study the worksheets
were adapted and a second worksheet produced. The second worksheet contained questions
which were identical in structure to those in the first worksheet but involving decimal
numbers which would encourage the use of a calculator. The children would be given a
graphics calculator with the Key Record Software running which they were asked to use when
completing the second worksheet. The main study has now been carried out in a further two
classes in UK schools, both middle ability year 8 classes. The children completed both
worksheets, and afterwards their worksheets was analysed alongside the Key Record data.
Some children were then interviewed and the teachers were also interviewed.
From the pilot study it was interesting to investigate the questions that the Japanese
children got wrong, and to analyse their ways of working. There was a general tendency to
treat all the questions as algebraic, even though they were mainly numerical. The calculation
that was answered incorrectly by most Japanese pupils was question 10:
the calculation (1 + 2)2 was in many cases calculated by expanding the brackets first:
Figure 1 Examples of three Japanese pupils’ work on question 10
and it was observed that the incorrect answers were more likely to be due to careless
errors rather than revealing misconceptions. When calculated in this way, the need to
remember a rule such as BIDMAS becomes unnecessary, although there is still a need to
know that indices are evaluated first in the numerator.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 39
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
This question was approached differently by the children in the UK. One girl’s
attempts at questions 9 and 10 of the non- calculator worksheet are shown in figure 2:
Figure 2 One UK pupil’s work on questions 9 and 10 on non-calculator worksheet
In question 10, although she correctly evaluated the denominator, involving brackets,
she did not evaluate the index first in the numerator. When interviewed, she was asked what
she was thinking about when doing the worksheet, she immediately answered “I was thinking
of BIDMAS” but when asked what this stood for she hesitated and then answered Brackets,
individual, divided, multiply, addition and subtraction” When prompted about what the “I”
stood for she did not know, and even when asked about the word “Index” she was not sure
what this meant, although when she was shown the number she immediately said “ oh – to
the power of 2?” which revealed that she understood what a power was, but had not related
this to the word “index” and therefore could make no sense of the I in BIDMAS. The same
misconception is also revealed in her answer to question 9. Her attempts at the corresponding
questions on the calculator paper are shown in figure 3:
Figure 3 The same pupil’s work for questions 9 and 10 of the with-calculator worksheet
Analysing her keystrokes revealed that she used her calculator efficiently with a good
grasp of the need to evaluate each of the numerator and denominator first before dividing.
Her incomplete understanding of the BIDMAS rule was overcome by using the calculator
efficiently. For question 9 the pupil correctly evaluated the numerator on her calculator, and
then evaluated the denominator:
Once happy with the denominator she proceeded to finish the calculation:
It would seem that she wanted to check that the answer from the first line was the
same as evaluating the power first, then adding, thus indicating that she had an idea of the
correct order, even though she had got the equivalent non-calculator question wrong. When
she was using her calculator she was able to investigate the effect of calculating the power
first and successfully confirm that this was the correct way to carry out the calculation. In
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 40
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
question 10 the pupil now seemed satisfied that the calculator would produce the correct
result for both numerator and denominator and worked efficiently to produce the correct
answer:
Conclusions
In this first part of the study it would seem that there are substantial differences between
Japanese and British childrens’ ability to carry out arithmetic calculations and the approaches
used. The Japanese pupils relied upon algebraic approaches which they generally employed
correctly, but this sometimes caused them to make the calculation unnecessarily complicated
and they made algebraic mistakes. The children in the UK relied heavily upon remembering
BIDMAS which worked well if they remembered it correctly but broke down if they did not
fully understand what all the letters stood for. However the use of a calculator enabled the
pupils to experiment and discover the conventions, which reflects the teaching approaches
used.
References
Berry, J., T. Graham and A.Smith. 2006. Observing Student Working Styles When Using
Graphic Calculators to Solve Mathematics Problems International Journal for
Technology in Mathematics Education 37 no.3 291-308
DCSF 2008. The National Strategies: Secondary Mathematics Exemplification
onlinehttp://nationalstrategies.standards.dcsf.gov.uk/strands/881/66/110129
Graham, T., C. Headlam, S. Honey, J. Sharp and A.Smith. 2003. The Use of Graphics
Calculators by Students in an Examination: What do they really do? International
Journal of Mathematical Education in Science and Technology 34 no.3 319 – 344
Graham.T., C. Headlam, J. Sharp, and B. Watson. 2007. An investigation into whether
student use of graphics calculators matches their teacher’s expectations International
Journal of Mathematical Education in Science and Technology 9 no.2 179 – 196
Hart, K.M. (Ed). 1981. Children’s Understanding of Mathematics: 11-16 Murray
Hewitt, D. 2003. Some issues regarding formal algebraic notation in Pope,S. (Ed)
Proceedings of the British Society for Research into Learning Mathematics’ 23 no.1
Japan Society of Mathematical Education. 2000. Mathematics Program in Japan
Lins, R and J. Kaput. 2004. The Early Development of Algebraic Reasoning: The Current
State of the Field New ICMI Studies Series 8 47 – 70
Sheryn, S.L. 2005. Getting an Insight into How Students Use their Graphical Calculators
Proceedings of the British Society for Research into Learning Mathematics 25 no.2
103 – 108
Sheryn , S.L. 2006a. What do Students Do with Personal Technology and How Do We
Know? How one student uses her graphical calculator International Journal of
Technology in Mathematics Education 13 no.3 151 – 158
Sheryn, S.L. 2006b. Investigating the Appropriation of Graphical Calculators by
Mathematics Students Ed.D. Thesis University of Leeds School of Education
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 41
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 42
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The role of attention in the learning of formal algebraic notation: the case of a mixed
ability Year 5 using the software Grid Algebra
Dave Hewitt
School of Education, University of Birmingham
The learning of formal algebraic notation is seen as a challenge for many students (Van
Amerom, 2003). The act of symbolising is not so much a problem. Hughes (1990) showed
that very young children are able to symbolise in order to record how many items were placed
into a tin. The problem is more concerned with interpreting and using someone else’s
notation, in this case the socially agreed convention of formal algebraic notation. In activities
where students are asked to find rules for pictorial patterns, they can often find rules but
expressing those rules in formal notation is seen as difficult. An indication of this is seen in
the levels of “patterning abilities” used by Ma (2008) and based upon Orton and Orton (1999)
where the highest level of this scale is students’ ability to express their rule in formal notation.
The difference between finding rules and expressing those rules in formal notation
highlights a difference I see between these two aspects of mathematics. Spotting patterns and
finding rules is algebraic in nature whereas how to express those rules is a matter of language
and notation. This is an example of the arbitrary and necessary divide (Hewitt, 1999) where I
call those things which are socially agreed, such as names and conventions, arbitrary as they
can appear to feel so for a learner and are a matter of choice. Those things which are
necessary concern properties and relationships and are not a matter of choice. For students to
learn the arbitrary they need to be informed of these socially agreed names and conventions,
whereas students can come to know the necessary through their own mathematical activity.
This implies there are different pedagogic challenges for a teacher between the arbitrary and
the necessary. The learning of algebraic notation is essentially about accepting and adopting a
socially agreed convention and thus lies in the realm of the arbitrary. The finding out of rules
is a different matter and lies in the realm of the necessary. Thus, the teaching of notation is
pedagogically a different challenge to the teaching of algebra and requires me providing
notation one way or another and helping students to accept, rather than question, notation and
adopt it within their work.
The Software
Figure 1: the first two rows of the grid
The vehicle I used to provide notation is the software Grid Algebra3 which is based
upon a multiplication grid (see Figure 1 where just the first two rows are shown). Movements
between the numbers can be made through dragging a number from one cell to another cell. A
move to the right would involve addition, to the left subtraction, down involves multiplication
3
Available from the Association of Teachers of Mathematics
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 43
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
and up division. These movements can be carried out with the software and result in formal
notation as indicated in Figure 2.
Figure 2: some movements on the grid
As movements were made students began to see the formal expressions as historical
artefacts, representing journeys which had been made on the grid. At times the notation was
the only evidence available for journeys which had taken place and thus there was a need for
students to use this notation in order to engage with tasks such as trying to re-create the
journey made. Brown and Coles (1999) have talked about setting up activities so that there is
a need to use algebra and here activities were set up where there was a need to use the
notation. Indeed the formal notation was subordinate (Hewitt, 1996) to some of the tasks as
the notation was not a requirement in students’ understanding the task yet the students were
forced to go through the notation in order to carry out those tasks. Such activities involve
students meaningfully practising interpretation of the formal notation even if that notation is
relatively new to them.
The study
The study took place in an inner city multi-cultural primary school which has a greater
percentage of children with free school meals than the national average and whose KS2
results are below the national average. There were 21 students in the group taught and these
were the total number from two Year 5 classes who returned parental permission slips and
who were personally happy to be part of the study. They were a mixed ability group ranging
from teacher assessed National Curriculum levels two to five. The class was taught by myself
on three consecutive days for one hour on the first day and one and half hours on the other
two days, giving a total of four hours. Some of this time was spent with them working on pen
and paper tasks and some of this time included two half hour sessions with them working
individually or in pairs in a computer room. The rest of the time the class worked with a
whole-class focus on the Interactive Whiteboard. Except for the pen and paper tasks, all the
time was spent working with the Grid Algebra software, and even the pen and paper tasks
were largely based upon the software, some of them being printout of sheets accompanying
the software. Their class teachers reported that none of them had been taught any formal
algebra, including use of a letter, or used formal algebraic notation.
Some key aspects of the way in which I worked with the students included a general
absence of anything being explained by myself. Instead there was use of questioning and an
expectation that students would notice and abstract rules concerning how notation was written
along with the mathematical processes required to solve linear equations.
There were a number of stages in the foci of the teaching sessions, which were:
students meeting notation for multiplication (brackets) and division (division line) and how
addition and subtraction fit in with those in larger expressions (placing of an
addition/subtraction sign following a division); order of operations within an expression;
introduction of letters; substitution; multiplying out brackets; inverse operations and solving
equations.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 44
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Several of the tasks involved students re-creating journeys given final expressions, or
finding where a journey began given only the final expression. These involved students
having to work through the notation as it was the only information provided and in this way
the notation was subordinate to the task in that reading the notation, and order within the
notation, was a necessary aspect of carrying out the task. This work I am not reporting here
but I will comment that all students at the end of the lessons were writing their work in
correct algebraic notation. What I do report here is a small window of time when the students
first met and began to work on writing expressions involving division.
Attention
Attention has been discussed as part of the teaching and learning dynamic in different ways.
Mason (1989) has talked of a shift of students’ attention which is indicative of a learning
process whilst Ainley and Luntley (2007) have discussed the “attentional” skills which
experienced teachers exhibit. Wilson (2009) has discussed the relationship between teacher
and student in relation to attention. She talks of the notion of alignment between teachers’
practices and students’ focus of attention. My interest in attention concerns pedagogic
decisions about where a teacher might wish students’ attention to be placed at particular
moments in a lesson. This will involve deliberate teaching acts which attempt to direct
students’ attention onto particular aspects of what is being discussed. My interest is in
catching such teaching moments but also following through and examining evidence of where
students’ attention might actually be placed over time following such teaching acts.
Methodology
Lessons were video recorded and written work was collected. The later viewing of these I
describe through the metaphor of glasses. The viewing of anything is never carried out
neutrally. In viewing something I bring with me my experience and the particular interests
which are currently present for me. In this case my viewing glasses contained lenses which
represented my particular interest in the placement of attention. Thus the viewing of video
and students’ work was carried out through these lenses. The act of noticing (Mason, 2002)
particular incidents or aspects of writing is indicative of links made between those artefacts
and my particular interest. What follows is a description of some of those artefacts along with
my accounting for their significance to me in relation to my interest in the placement of
attention. In particular, this paper will focus the notation associated with the division line.
The notation for division: attention on particularities of expressions
The notation of addition and subtraction was familiar to the students and so the focus up to
this point in time was on seeing an expression as an object rather than a process to be carried
out. For example, seeing ‘2+1’ as how it is rather than wanting to carry out that addition and
say three. However, the notation of division was new to these students and attention was now
focused onto particular aspects of the expression which were novel for the students.
Having made a downwards movement to create a multiplication I asked them what
was the opposite of doing this. They responded almost in unison with “divide” and after me
asking “…by what?” they said “Divide by two”. This continued as follows:
DH: Divide by two? OK. So when I go from 8 to 4 I’ve got to do eight divided
by two. Is that right?
Students: Yes.
DH: OK, and here it is [movement made resulting in ,8-2. ]
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 45
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Students: Oh... [A few seconds of quiet]... Fraction [several voices] / It’s a
fraction but it stands for divide [one voice].
DH: So, say this to me [pointing at the 8].
Students: Eight… [DH points to the divide line] … divide/over/add [DH points
to the 2] … two.
DH: OK. Just say divide. OK? Ready? [Points to the 8 then the divide line and
then 2].
Students: [In time with the pointing] Eight... divided by... two.
DH: OK? Eight divided by two.
For many students this notation appeared to be a surprise. There were also several
voices saying that ,8-2. was a fraction and I would conjecture that this form of notation had
previously only been used under the topic of fractions, whereas the notation 8÷2 had been
used for division. Here attention was brought to the individual parts of ,8-2. in order to learn
to read how this expression was going to be said within these lessons, as a division rather than
as a fraction. This required taking something which the students viewed as an object, a
fraction, and seeing it as a process, a division. This was the opposite shift of attention which
was involved earlier when I wanted to help students see an expression as an object rather than
a process.
Following the above sequence, students were asked to write down what it would look
like if I dragged the ‘6’ in row two up to the cell above, then (having already hidden the ,8-2.
) when the ‘8’ would look like dragged up (so a repeat of what we had done above) and then,
having dragged the ‘8’ in row two to the left to produce ‘8-4’, what it would look like if ‘8-4’
was dragged up to the cell above. For each of these, students wrote down what they thought it
would look like and then the action was carried out on the grid to reveal what it did actually
look like. Out of the 18 booklets which were finally collected in on the last day, 12 of them
wrote down ,6-3. , two of them wrote ,3-6. , another wrote 6÷3 and the remainder wrote ,6-2. .
My conjecture is that the attention of many of the students was on the grid and the start and
finish cells of this little journey. It started with the number ‘6’ and the cell which it would end
on had ‘3’ in it. What was not visually present on the grid was the fact that the operation was
division by two. Instead many students have used numbers which were visible. In particular
the ‘3’ is visually above the ‘6’ and this might be a factor why two of them wrote ,3-6. . What
was adopted by all but one student was the use of a horizontal line for division.
The challenge of ‘8-4’ divided by two was new to them as they had only seen single
numbers being divided. This was successfully written by 11 of the 18 students. However,
since the finishing cell of this movement contained the number ‘2’ it was not clear whether
these students wrote dividing by two because they were aware of the mathematical operation
or whether the ‘2’ came from them attending to the final cell. So, in retrospect, this was not a
good example to have chosen.
Four of the students who wrote something different seemed to carry out the
calculation of 8-4 and so wrote either ,4-2. or ,2-4. . Two other students wrote:
This appeared to indicate the two stage nature of what was happening. The first
operation of subtraction was carried out and the second operation of division shown with that
answer. This reminded me of the common misuse of the equals sign when someone writes
8−4=4÷2=2, only this time without the final division calculation taking place. One other
person wrote:
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 46
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The fact that previous examples had a division line underneath a single number might
have been generalised to this example. Since notation is arbitrary, there would be no way of
someone knowing how the case for single numbers is generalised to cases for expressions,
and so this seemed a perfectly reasonable way to generalise the use of the division line.
Division not only brings with it the issue of the division line but also adds a notational
issue when it is followed by addition or subtraction. Shortly after the sequence above I said I
would drag ,8−4+6-2. one cell to the left (where this would produce ,8−4+6-2.−1 ). Of the
13 who attempted this, six wrote down the conventionally correct answer with five writing
,8−4+6−1-2. . This is again a novel situation for them and so how would they be expected to
know what is the conventional way of writing this? When I moved the expression across to
produce the result there were a number of students who said “Yessss!” followed by a small
gap of a couple of seconds followed by several voices saying “Uh?” with one student saying
“Why’d the one have to go there?” I did not respond to this and I am not convinced the
question was actually directed at me. It appeared more an expression of surprise. I did,
however, want to focus attention, in a factual way rather than an explanatory way, as to this
feature of where the subtraction sign was written. In doing so I offered a visual and aural
image; I pointed the pen at the left hand end of the division sign and moved to the right along
the line saying “Just notice, zzzzzippp,...” and then continued along this line taking my pen
off briefly before putting it back down again along the subtraction line and continued by
saying “... ping!” This aural image was heard being repeated by a number of the students as I
then went about clearing the grid and beginning to start the next challenge. None of the
students asked again for any explanation of why the notation was how it was.
In the expressions the students wrote down for the above, there was only one which
involved a calculation: ,8−4+6−1=4-2. . The ‘=4’ is where we finished up after this
movement and so this equation could be considered mathematically correct and just not
written conventionally. This whole sequence of the lesson from when the first division was
carried out lasted under seven minutes and in this time there was a significant shift towards
writing expressions conventionally; the division line was adopted and there were nearly no
calculations being carried out, indicating a shift back from process to object.
Final remarks
Coming to accept and adopt formal notation involved a development where attention was
shifting from process to object, and occasionally vice versa. At times attention appeared to be
on what was visible within the grid rather than the operations which were visually implicit
even if verbally explicit. Pedagogic techniques were occasionally used to draw attention onto
particular detail and at other times to ignore detail and treat an expression as an object upon
which further operations were being carried out. I note that there was a general sense of
acceptance of the notation, rather than questioning why it appeared how it did. Even when
this did happen, with the initial creating of a division and then with a subtraction following a
division, any comments did not appear to be directed towards me. I conjecture that this was
because I was not doing the writing of those expressions. Instead the computer created them
and students are quite used to accepting all sorts of arbitrary conventions which form part of
interactions with technology.
References
Ainley, J. and Luntley, M. (2007), 'The role of attention in expert classroom practice', Journal
of Mathematics Teacher Education, 10(1), pp. 3-22.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 47
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Brown, L. and Coles, A. (1999), 'Needing to use algebra - a case study', in O. Zaslavsky (Ed.),
Proceedings of the 23rd Conference of the International Group for the Psychology of
Mathematics Education, Vol. 2, Haifa, Israel Institute of Technology, pp. 153-160.
Hewitt, D. (1996), 'Mathematical fluency: the nature of practice and the role of subordination',
For the learning of mathematics, 16(2), pp. 28-35.
Hewitt, D. (1999), 'Arbitrary and Necessary: Part 1 a Way of Viewing the Mathematics
Curriculum', For the Learning of Mathematics, 19(3), pp. 2-9.
Hughes, M. (1990) Children and Number. Difficulties in Learning Mathematics, Oxford:
Basil Blackwell.
Ma, H.-L. (2008), 'The algebraic thinking of 5th and 6th graders to solve linear patterns with
pictorial contents ', Research and Development in Science Education Quarterly, 50,
pp. 35-52. (In Chinese).
Mason, J. (1989), 'Mathematical abstraction as the result of a delicate shift of attention', For
the learning of mathematics, 9(2), pp. 2-8.
Mason, J. (2002) Researching your own practice: the Discipline of Noticing, London:
RoutledgeFalmer.
Orton, A. and Orton, J. (1999), 'Pattern and the approach to algebra', in A. Orton (Ed.),
Pattern in the Teaching and Learning of Mathematics, London: Cassell, pp. 104-120.
Van Amerom, B. A. (2003), 'Focusing on Informal Strategies When Linking Arithmetic to
Early Algebra', Educational Studies in Mathematics, 54(1), pp. 63-75.
Wilson, K. (2009), 'Alignment between teachers’ practices and pupils’ attention in a
spreadsheet environment', in M. Tzekaki, M. Kaldrimidou and H. Sakonidis (Eds),
Proceedings of the 33rd Conference of the International Group for the Psychology of
Mathematics Education, Vol. 5, Thessaloniki, Greece, 353-360.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 48
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Lower secondary school students’ attitudes to mathematics: Evidence from a large-scale
survey in England
Jeremy Hodgena*, Dietmar Küchemanna, Margaret Browna & Robert Coeb
a
King’s College London, bUniversity of Durham
In this paper we present some preliminary data from the ESRC funded ICCAMS project
about current student attitudes to mathematics at Key Stage 3 in England. We compare
attitudes by sex and by attainment. Whilst the data largely confirms existing findings, an
unexpected result was that a very high proportion of students responded that, for
mathematical success, effort was more important than ability. We also present some
interview data concerning student attitudes.
Keywords: Attitudes, attainment, gender
Background
Increasing Student Competence and Confidence in Algebra and Multiplicative Structures
(ICCAMS) is a 4-year research project funded by the Economic and Social Research Council
in the UK (Hodgen et al. 2008, 2009). In this paper, we report and discuss early findings of
the study regarding students’ attitudes drawing on both survey and interview data.
Methods and theoretical framework
Phase 1 of the ICCAMS project consists of a large-scale survey of 11-14 years olds’
understandings of algebra and multiplicative reasoning in England using three tests of
mathematical understanding and an attitudes questionnaire. The three mathematics tests,
covering algebra, decimals and ratio, were originally used in the late-1970s as part of the
Concepts in Secondary Mathematics and Science (CSMS) study. (See Hart 1981, for a
discussion of the test development.) The attitudes test is adapted from previous work (Boaler,
Wiliam, and Brown 2000). In Phase 2 of the study we are conducting a collaborative research
study with eight teachers extending the investigation to classroom / group settings and
examining how assessment can be used to improve attainment and attitudes.
Participants
In June and July 2008, tests were administered to a sample of approximately 3000 students
across Key Stage 3 (KS3) from 10 schools and approximately 90 classes. In England, KS3
refers to the first three years of secondary school: Years 7 (ages 11-12), Year 8 (ages 12-13)
and Year 9 (ages 13-14). Since the survey was conducted at the end of the school year, the
vast majority of these students were at the older end of these age ranges: 12, 13 and 14 years
old, respectively. We report here on the attitudes of a sub-sample of 1422 students for whom
we have linked data on attitudes and attainment on the Ratio test, enabling us to report on the
relationship between students’ attitudes and attainment. The sub-sample consisted of 494
Year 7 students, 524 Year 8 students and 394 Year 9 students. Of the total, 748 were boys and
674 were girls.
The Ratio test reports students’ mathematical attainment using a hierarchy of levels
from Level 0 up to Level 4 (Hart 1981; Brown, Küchemann, and Hodgen Forthcoming). In
the sub-sample, the attainment of boys was slightly higher than girls (see Table 1).
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 49
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Boys
Girls
Ratio Level
0
1
2
3
4
0
1
2
3
4
Year 7 [12]
24%
49%
16%
8%
4%
23%
56%
13%
6%
2%
Year 8 [13]
16%
49%
14%
16%
6%
20%
52%
15%
9%
5%
Year 9 [14]
10%
41%
15%
19%
15%
9%
48%
19%
16%
9%
Table 1: Comparison of the attainment of boys and girls on the Ratio test by Year group [age] for the subsample.
We note that these early results should be treated with caution. In particular, we note
that this sub-sample of students appears to be slightly higher attaining than the general KS3
population in England. A further sample of approximately 3000 students took part on the
survey in Summer 2009 and these results are currently being analysed. When this process is
complete, the sample will be representative of schools and students in England. The full
sample was randomised and drawn from MidYIS, the Middle Years Information System.
MidYIS is a value added reporting system provided by Durham University, which is widely
used across England (Tymms and Coe 2003).
Research background: Attitudes, attainment and participation
In England, in common with many other countries, too few students choose to continue
studying mathematics once it ceases to be compulsory. There is considerable research in
England addressing reasons for non-participation in mathematics - students stop studying
mathematics because they experience it as difficult, abstract, boring and irrelevant (Osborne
et al. 1997). The most recent findings relating to 16 year-olds (Matthews and Pepper 2007;
Brown, Brown, and Bibby 2008) suggest that students’ attainment and attitudes are strongly
inter-related. A major factor is that even relatively successful students perceive that they have
failed at the subject and lack confidence in their ability to cope with it at more advanced
levels, especially in comparison to the perceived ‘clever core’ of fellow-students. When
pressed about the reasons for their feelings of failure, students suggest that they do not
understand parts of what they have been taught and point to the predominance of routine and
formal work on algebra and multiplicative reasoning (Nardi and Steward 2003). These
perceptions of failure appear to be strongly linked to ideologies of ‘ability stereotyping’
(Ruthven 1987) and ability grouping (Boaler, Wiliam, and Brown 2000). Girls’ attitudes to
mathematics tend to be more negative than boys. Boaler and Greeno (2000) link these more
negative attitudes to mathematical teaching practices that do not emphasise understanding.
However, the nature of the relationship between attitudes and attainment is poorly
understood. In common with other highly and relatively highly attaining countries in TIMSS
2007, English students’ attitudes fell in comparison to TIMSS 1999, the last comparable data
(Sturman et al. 2008). The fall in England at Grade 9, 25 percentage points, was greater than
for other comparable countries, despite an increase in attainment elative to previous TIMSS
surveys. One puzzling result is that, although within countries higher attainment is associated
with more positive attitudes, the between country effect is in the opposite direction –
countries with higher mathematical attainment tend to have more negative attitudes (Askew et
al. 2010).
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 50
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Early analysis of survey data: Student attitudes to mathematics
This early analysis of the survey largely confirms existing findings. For example, students’
attitudes dropped as they got older. 63% of 12 year olds responded that they enjoyed
mathematics lessons, but this had fallen to 54% of 14 year olds. (See Table 2.) However, the
drop was greater for girls than for boys. Although a similar proportion of 12 year-old boys
and girls said that they enjoyed mathematics lessons (64% and 62%, respectively), by age 14
the proportions were 59% of boys compared to 50% of girls. This reflects the TIMSS 2007
finding of greater levels of self-confidence amongst Grade 9 boys than girls in England
(Sturman et al. 2008). Similarly, although boys’ perceptions of their own ability were largely
stable and positive across the age range, girls’ perceptions of their ability had dropped to
about half the sub-sample by age 14. (See Table 3.)
Year Group [Age]
Year 7 [12]
Year 8 [13]
Year 9 [14]
Boys
64%
61%
59%
Girls
62%
61%
50%
Total
63%
62%
54%
Table 2: Positive responses to ‘Do you enjoy maths lessons?’ by age and gender.
Boys
Girls
Year 7 [12]
78%
67%
Year 8 [13]
80%
66%
Year 9 [14]
81%
52%
Table 3: Positive responses to ‘Do you think you are good at maths?’ by age and gender.
Across the age range, more boys than girls thought that they would study mathematics
after GCSE and for both boys and girls this dropped. However the drop was greater for girls
with only a quarter of the sub-sample of 14 year olds saying that they would continue post-16.
One potentially positive finding is the relatively high proportion of students of both sexes
who were undecided. (See Table 4.)
Boys
Girls
Year [Age]
Year 7 [12]
Year 8 [13]
Year 9 [14]
Year 7 [12]
Year 8 [13]
Year 9 [14]
Yes
38%
34%
35%
30%
28%
26%
No
15%
15%
22%
16%
19%
32%
Don't know
46%
51%
44%
54%
53%
43%
Table 4: Responses to ‘Do you think you will continue to study maths after GCSE?’ by age and gender.
One surprising result was that a very high proportion of students responded that
working hard was more important for success in mathematics than natural ability: 89% of 12
year olds and 85% of 14 year olds, with the drop being almost wholly due to a change in boys
attitudes. (See Table 5.) This result appears to contradict previous findings.
Year Group [Age]
Year 7 [12]
Year 8 [13]
Year 9 [14]
Boys
88%
81%
82%
Girls
89%
89%
87%
Total
89%
85%
85%
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 51
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Table 5: ‘Working hard’ responses to ‘Which do you think is more important for success in maths? Working
hard or Being naturally clever’ by age and gender.
The relationship between attitudes and attainment
We now turn to examine the relationship between attainment and attitudes.
Unsurprisingly, a greater proportion of the highest attaining students said that they enjoyed
maths and, in contrast to other students, this proportion did not fall for older students. (See
Table 6.)
~ Bottom 75%
~ Top 25%
~ Top 10%
Year 7
62%
64%
71%
Year 9
52%
59%
69%
Table 6: Comparison of positive responses to ‘Do you enjoy maths lessons?’ between Year 7 [age 12] and Year
9 [age 14] by attainment. Rough attainment proportions calculated as follows: for bottom 75% by aggregating
Levels 0 and 1 for Year 7 [75%] and Levels 0, 1 and 2 for Year 9 [71%]; for Top 25% by aggregating Levels 2,
3 and 4 for Year 7 [25%] and Levels 3 and 4 for Year 9 [29%]; for Top 10% by aggregating Levels 3 and 4 for
Year 7 [10%] and taking Level 4 for Year 9 [12%].
Again, unsurprisingly, a greater proportion of high attaining students were intending
to continue to study mathematics after GCSE. However, the proportion of the highest
attaining 10% intending to continue studying post-16 was only 57%. It is worth noting that
24% of the bottom 75% said that they intended to continue with mathematics post-16. This
figure is relatively high in comparison to the limited options in English education for this
group of students post-16. (See Table 7.)
~ Bottom 75%
~ Top 25%
~ Top 10%
Year 7
32%
43%
59%
Year 9
24%
46%
57%
Table 7: Comparison of students intending to continue to study mathematics post-GCSE at Year 7 [age 12] and
Year 9 [age 14] and by attainment. See Table 6 for how rough attainment proportions calculated.
It is also noteworthy that the proportion of students who felt that the working hard was
more important than natural ability was high at all attainment levels. (See Table 8.)
Year Group [Age]
Year 7 [12]
Year 8 [13]
Year 9 [14]
0
89%
81%
86%
Attainment (as measured by Ratio Level)
1
2
3
89%
89%
91%
87%
87%
82%
86%
86%
81%
4
86%
82%
85%
Table #: Proportions of ‘working hard’ responses to ‘Which do you think is more important for success in
maths? Working hard or Being naturally clever’ by age and attainment.
Findings from interviews
We have conducted several group interviews with students from Phase 2 schools. These
interviews have followed a semi-structured format. Here, for reasons of space, we discuss just
one interview. This interview is of particular interest because it sheds light on the issue of
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 52
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
‘working hard’ and success at mathematics referred to in the analysis of survey data above.
The interview was with three Year 8 students from the second set of a relatively high
attaining state comprehensive school. The students were typical in disliking algebra and
describing it as “bad”. Similarly, when asked whether they themselves were “good at maths”,
they referred to their test results and levels from the end of Year 7 (Hodgen and Marks 2009).
The following is an extract about what it takes to be good at mathematics:
Researcher: What’s it take to be good at maths?
Student K: - dedication...
Student M: -knowing your numbers...
Student C: I think it just comes normal to you... with some people, like you might not be
good at many subjects, but when it comes to maths you could just be brilliant...
Researcher: And is that what those three were...
Student C: I reckon one them was... J, when it came to him, because he... I don’t think he’s
that great, like superb at all the others but when it comes to maths he’s superb, he gets
every answer first, so...
Researcher: So what has he got...? What’s it take to be good at maths?
Student M: -study...
Researcher: Do you think he worked hard, this boy...?
Student K: ...um, I think yeh, he worked hard.. I think ... if he didn’t know what he had to
do in class, if he didn’t understand it, then he would go home and like get hold of a maths
book or something which would explain it better and also give him questions that he could
do to make sure he actually knew what he was doing when he came back to class the next
day, so he could understand it
An interesting feature of this discussion was the way in which the student referred to J
as “great” and “superb” but also as someone with “dedication” who “worked hard”.
Conclusion
As we have already noted, the analysis is at an early stage. The next stage of analysis will
further investigate the relationship between students’ attitudes and attainment.
References
Askew, M., J. Hodgen, S. Hossain, and N. Bretscher. 2010. Values and variables: A review of
mathematics education in countries with high mathematics attainment. London: The
Nuffield Foundation.
Boaler, J., and J. G. Greeno. 2000. Identity, agency and knowing in mathematics worlds. In
Multiple perspectives on mathematics teaching and learning, edited by J. Boaler.
Westport, CT: Ablex Publishing.
Boaler, J., D. Wiliam, and M. Brown. 2000. Grouping - disaffection, polarisation and the
construction of failure. British Educational Research Journal 26 (5):631-648.
Brown, M., P. Brown, and T. Bibby. 2008. “I would rather die”: Attitudes of 16 year-olds
towards their future participation in mathematics. Research in Mathematics Education
10 (1):3-18.
Brown, M., D. E. Küchemann, and J. Hodgen. Forthcoming. The struggle to achieve
multiplicative reasoning 11-14, The Seventh British Congress of Mathematics
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 53
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Education (BCME7). Paper submitted to The Seventh British Congress of
Mathematics Education (BCME7), University of Manchester.
Hart, K., ed. 1981. Children's understanding of mathematics: 11-16. London: John Murray.
Hodgen, J., D. Küchemann, M. Brown, and R. Coe. 2008. Children’s understandings of
algebra 30 years on. Proceedings of the British Society for Research into Learning
Mathematics 28 (3):36-41.
———. 2009. Secondary students’ understanding of mathematics 30 years on. Paper read at
British Educational Research Association (BERA) Annual Conference, at University
of Manchester.
Hodgen, J., and R. Marks. 2009. Mathematical ‘ability’ and identity: a socio-cultural
perspective on assessment and selection. In Mathematical Relationships in Education:
Identities and Participation, edited by L. Black, H. Mendick and Y. Solomon.
London: Routledge.
Matthews, A., and D. Pepper. 2007. Evaluation of participation in GCE mathematics: Final
report. QCA/07/3388. London: Qualifications and Curriculum Authority.
Nardi, E., and S. Steward. 2003. Is mathematics T.I.R.E.D? A profile of quiet disaffection in
the secondary mathematics classroom. British Educational Research Journal 29
(3):345-367.
Osborne, J., P. Black, J. Boaler, M. Brown, R. Driver, and R. Murray. 1997. Attitudes to
Science, Mathematics and Technology: A review of research. London: King's College,
University of London.
Ruthven, K. 1987. Ability stereotyping in mathematics. Educational Studies in Mathematics
18:243-253.
Sturman, L., G. Ruddock, B. Burge, B. Styles, Y. Lin, and H. Vappula. 2008. England’s
Achievement in TIMSS 2007 National Report for England. Slough: NFER.
Tymms, P., and R. Coe. 2003. Celebration of the Success of Distributed Research with
Schools: the CEM Centre, Durham. British Educational Research Journal 29 (5):639653.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 54
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Simon Says: Direction in Dialogue
Jenni Ingram, Mary Briggs and Peter Johnston-Wilder
University of Warwick
There has been a steady increase in the quantity of mathematics education research
focusing on language, discourse and interaction. A wide variety of theoretical
frameworks and methodological approaches have been taken including discursive
psychology, commognition, and discourse analysis. This paper explores the use of a
conversation analysis approach to analyzing interactions in mathematics classrooms. In
particular what this approach can tell us about the structure of interactions and the use of
repair in the negotiation of mathematical meanings.
Keywords: classroom discourse, conversation analysis, repair
Introduction and Background
Several authors have focused on language, discourse and communication in the mathematics
education literature over the past twenty years. Some authors have explored the relationship
between discourse and identity (Boaler, Wiliam and Zevenbergen 2000, Lee 2006), beliefs
about mathematics and beliefs about teaching and learning. Others have focused on
interactional strategies and the implications of these for the learning of mathematics
(O'Connor and Michaels 1993). More recently, Sfard (2007) has proposed a theoretical
framework which conceptualises learning mathematics as a transformation and extension of
learner’s discourse. This “commognitive” framework treats a learner’s discourse as the object
of learning and not just the means of learning, raising the importance of research on
mathematics classroom discourse.
There are many methodological approaches available in the research on discourse:
discourse analysis including systemic functional linguistics, critical discourse analysis,
discursive psychology and conversation analysis to name a few. This paper explores the
conversation analytic approach as a means of explicating the complexity of interaction in the
classroom setting.
After outlining the conversation analysis methodology the notion of repair is explored
in the context of two extracts taken from transcripts of whole-class interactions. Differences
in the organisation of repair between the classroom contexts and everyday conversation are
then examined and the implications these differences have on the learning of mathematics are
discussed.
Conversation analysis
The origins of conversation analysis lie in the analysis of naturally occurring conversation but
have been extended to include the analysis of institutional settings such as courtrooms and
emergency help lines (see Drew and Heritage 1992 for more examples). McHoul (1990) used
a conversation analysis approach in his study of geography classrooms and Seedhouse (1996)
offers an in-depth analysis of second language classrooms.
Conversation analysis (CA) as a methodology assumes that interaction is structurally
organised and the goal of CA is the exposition of this structure from the perspective of the
participants themselves (Levinson 1983). Consequently, claims about the existence of
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 55
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
various structures of interactions need to be supported with evidence in the transcripts that the
participants orient themselves to these structures. This is often referred to as the next-turnproof-procedure and serves as an attempt to offer empirical proof and prevent the imposition
of the researcher’s preconceptions.
Of particular importance to the analysis of classroom discourse is the CA dynamic
view of context. The well-known initiation-response-feedback (IRF) triad from a discourse
analytic approach is proposed as a common feature of classrooms but this takes a static view
of the context of the classroom. The CA approach examines why the IRF triad shapes the
classroom context. In particular, each occurrence of an IRF sequence can only be understood
within the sequential context of the interactions as they are affected by previous utterances
and it influences those utterances that follow. Each part of the triad has multiple roles in
relation to what has occurred before and in what is to follow. A response, for example is
constrained by the nature of the initiation it follows but also constrains the feedback that is yet
to come. The CA approach therefore offers a tool for explicating the fluidity of classroom
interaction, in particular how the context (mathematical, task-based, management) can shift
during and after individual turns. A CA approach provides an in-depth analysis of a specific
context but also limits the generalisability of any analysis.
Talk in classrooms is usually goal oriented. The multidimensionality (Doyle 1990) of
classrooms means there are multiple goals influencing interaction in the classroom. These
include the pedagogic goals of the teacher and the wider goals associated with the behaviour,
motivation and learning of the participants. CA examines interactions with reference to these
goals, but the analysis seeks only to examine the interactional advantages and disadvantages
and does not evaluate the effectiveness of these interactions from a pedagogical perspective.
This could lead to a conflict between pedagogic goals and the interactional goals controlling
the discourse.
The structures of turn taking and repair are particularly relevant to the classroom
contexts and it is the latter that is the focus for this paper. There are three features of a repair:
the trouble source; the initiation; and the outcome. Trouble has a broader definition than that
of errors and mistakes, including difficulties in understanding, hearing or in the structure of
the interaction itself. The repair initiation and outcome are sequential and can be performed
by the speaker in which the trouble occurred (self) or other participants (other). Self-initiated
self-repair is the preferred type of repair, occurring most often in everyday conversation,
whilst other-initiated other-repair is very rare (Schegloff, Jefferson and Sacks 1977). In CA,
the preference for self-initiated self-repair is not that the participants like or want to do this,
but the act is socially affiliative. Preferred actions are normally bald and direct, without
hesitation or delay, whilst dispreferred actions usually include hesitation, mitigation, and
delay. There is some evidence that the preference organisation of repair is different in the
classroom context (McHoul 1990) with other-initiated repairs occurring more frequently.
This is perhaps unsurprising in many classrooms as it is often the teacher who not only has
the expertise to identify the trouble source but also the authority to initiate a repair.
The data discussed in this paper is taken from the transcript of a lesson with 12-13
year old students, considered to be high achievers in relation to their peers, focusing on
measures of central tendency, and is chosen to exemplify the features of repair outlined
above. The students have been asked to calculate the missing value if the mean is 70 and the
other values are 72, 43, 85 and 71. The extracts are taken from the whole-class interactions
that occurred after the students had been given some time to work on the problem
individually.
Extract 1
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 56
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
1
Sam:
I added them all up
2
Simon:
you added them all up
Sam has been nominated by the teacher (Simon) to explain what he did. His answer is
incorrect and Simon initiates a repair in line 2. Immediately we can see the importance of
knowing the interactions that preceded this interaction in order to identify Sam’s utterance as
causing trouble. However, these two utterances do not themselves offer evidence that this is
in fact an other-initiated repair. We need to look at the utterances that follow for that:
Extract 2
3
Sam:
yeah
4
Simon:
so you did (0.9) what that (0.6) plus that (0.5) plus
that, did you add that one on as well.
5
Sam:
u:m: no
6
Simon:
okay
7
Sam:
and then (0.6)I: (1.5) divided that by five (0.7) to
get the how much she needed (0.7) in the last (0.3)
um: (.) test.
8
Simon:
so you added up the four numbers, (0.8) you added up
four numbers (0.3) and then you divided by five?
(1.8) is that it?
9
Sam:
yeah
In line 3, Sam’s affirmative response indicates that Simon’s repetition of his answer in
line 1 has been understood by Sam as a check that Simon has heard and understood Sam
correctly. In lines 4 and 8, it becomes clear that Simon in fact meant his repeat as a repair
initiation by repeating, expanding and recasting Sam’s initial response further and
emphasising the words four and five in line 8, locating the source of the trouble. The repair
itself is not performed in the entire episode but the dispreference for other-repair is clear from
Simon’s final utterance:
Extract 3
18
Simon:
Sam added them up. okay (1.2) Sam added them up
shhshh shh shh. Sam added them up, (0.7) they added
up to two hundred and seventy one, that is a useful
bit of information (1.0) bu:t that thing about
dividing by five. that seemed to me, I don't know, a
little bit nonsensical cause you've only got four
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 57
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
numbers, dividing by five I'm not sure. Phillip.
Here Simon is now addressing the class as a whole. He starts by repeating Sam’s
response and follows it with a positive evaluation. Then he pauses before locating the trouble
again. This second half of the response is hesitant and mitigated with vague phrasing “that
thing about” and hedged comments “seemed to me”, “I don’t know”, “I’m not sure”. Simon
is avoiding making the repair, and in this example, the repair is not directly made until the
next student speaks and offers the expected answer:
Extract 4
19
Phillip:
um you need, if yo-, you can find th- like all the
numbers, the end mark, the end percentage means that
there's like three hundred and fifty percent
altogether divided by five it comes up to seventy.
20
Simon:
right hold on a sec. (0.5) three hundred and fifty
percent,(.) er I suppose, can you add percentage
together and then get three hundred and fifty
per[cent I suppose so ]ok (0.9)
21
Phillip:
22
Simon:
[no what we
]
so you're saying that if you've got five numbers
(0.4) and you want to get a mean (.) of seventy
In this example, Phillip is searching for the words he needs to explain what he did. In
line 2, Simon initiates a repair focused around the trouble associated with whether you can
have three hundred and fifty percent or not. There is no indication in the interaction as to
whether this is Simon’s interpretation of Phillip’s trouble (in which case it is an other-initated
repair) or a new trouble source which is Simon’s himself (in which case it is a self-initiated
repair). Phillip’s interruption indicates that this was not the source of trouble for Phillip;
Simon recasts Phillip’s answer in line 4, consequently performing the repair.
Each of the two repairs in extract 4 is clearly different in nature. The first is a
difficulty in communication revealed by Phillip’s word search, whilst the second is
mathematical (can you have three hundred and fifty percent?). The latter appears to be
repaired in the same turn as the trouble (line 2) yet in line 4, Simon’s recasting uses the word
number instead of percent, indicating that a repair of meaning has not actually been
performed. It is interesting to note that the trouble in the first extract was also mathematical
and in both cases, the repair was to some extent not performed.
In conclusion, a conversation analytic approach offers an effective tool for exploring
and exposing the structure and complexity of mathematics classrooms. In particular, the
preference organisation of repair explicates the roles adopted by the participants of teacher
and students. The more frequent prevalence of other-initiated repair defines the teacher as
expert and in a position of authority, similar to adult-child interactions in other contexts. The
dispreference for other-initiated other-repair remains evident in the extracts discussed above.
Future work will include the analysis of a broader data set to examine the structure of
repair in the secondary mathematics classroom. This includes exploring the similarities and
differences between the nature of the trouble source, the participants in the repair and the
repair trajectory. The implications of each of these on the learning of mathematics are of
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 58
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
particular interest. For example, the relationship between the dispreference of other-initiated
other-repair and the role of errors and mistakes in the learning of mathematics.
Bibliography
Boaler, J., D. Wiliam, and R. Zevenbergen. 2000. “The construction of identity in secondary
mathematics education.” Proceedings of the Second International Education and
Society Conferenc, ed by J. Matos and M. Santos. Univeridade de Lisban.
Doyle, W. 1990. “Classroom management techniques.” In Student Discipline Strategies:
Research and Practice, by O. C. Moles. SUNY Press.
Drew, P., and J. Heritage. 1992. Talk at Work: Interaction in Institutional Settings.
Cambridge: Cambridge University Press.
Lee, C. 2006. Language for Learning Mathematics: Assessment for Learning in Practice.
Buckingham: Open University Press.
Levinson, S. 1983. Pragmatics. Cambridge: Cambridge University Press.
McHoul, A. W. 1990. “The organization of repair in classroom talk.” Language in Society 19:
349-377.
O'Connor, M. C., and S. Michaels. 1993. “Aligning academic task and participation status
through revoicing: Analysis of a classroom discourse strategy.” Anthropology and
Education Quarterly 24, no. 4: 318-335.
Schegloff, E. A., G. Jefferson, and H. Sacks. 1977. “The preference for self-correction in the
organization of repair in conversation.” Language 53: 361-382.
Seedhouse, P. 1996. Learning Talk: A Study of the Interactional Organisation of the L2
Classroom from a CA Institutional Discourse Perspective. Unpublished PhD
Dissertation: University of York.
Sfard, A. 2007. “When the Rules of Discourse Change but Nobody Tells You: Making Sense
of Mathematics Learning From the Commognitive Standpoint.” The Journal of the
Learning Sciences 16, 4: 565-613.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 59
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 60
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The relationship between number knowledge and strategy use: what we can learn from
the priming paradigm
Tim Jay
Graduate School of Education, University of Bristol
Priming methods involve showing a stimulus for a short amount of time (the prime),
followed by a second stimulus (the target), which children are asked to perform some
operation on. If there is a strong association between the prime and target for a particular
child, then the operation on the target will be facilitated by the presence of the prime.
This paper describes a project in which priming methods are used to add to our
understanding of strategy development for simple addition problems. Children were
asked to complete two activities; a priming trial designed to demonstrate priming effects
for doubling, and a set of addition problems where participants were asked to explain
how they arrived at their answers. Approximately half of the participants used counting
strategies (count-on from first, count-on from smallest), while half used non-counting
strategies (decomposition, tie or retrieval). Results indicate that a priming effect for
doubling relationships but only for the group of children using non-counting strategies.
This result could help to explain the relationship between the development of number
knowledge and the development of new strategies.
Introduction
There is a well established understanding of the normal course of development of strategy-use
when solving single-digit arithmetic problems (e.g. Fuson 1992). Children begin this course
of development by counting both addends in an addition problem, often using concrete
objects such as fingers to aid the process. The next strategy to appear is the count-one
strategy, in which children start with one of the addends, then count on from there to find the
answer. The 'min' strategy usually comes next, which involves children choosing to count on
from the largest addend. These three strategies all involve children counting in order to arrive
at an answer to a problem. At some point, children will begin to be able to solve some simple
problems using retrieval – directly accessing answers to problems stored in memory.
Although strategies generally appear in this order, children maintain a repertoire of several
strategies any any given point during this development, and show a high degree of variability
in their application of strategies to problems (Siegler 2007).
Fewer studies have addressed the nature of strategies used to solve problems resulting
in answers greater than 10. However, this aspect of the literature is growing due to the recent
focus on children's adaptive expertise in selecting amongst strategies (Verschaffel, Torbeyns,
De Smedt, Luwel, & Van Dooren 2007). When children are solving addition problems that
bridge 10, there are heuristic strategies available that sit between the counting strategies and
direct retrieval in terms of efficency. The problem '8 + 7', for example, might be solved by
converting the problem to '7 + 7 + 1' – this is often referred to as the 'tie' strategy and is often
used in the case of 'near tie' problems where the addends differ by 1. Alternatively, '8 + 7'
might be solved by converting to '8 + 2 + 5', if a child knows their number-bonds to 10
(sometimes known as 'ten-friends') – this is often referred to as a 'decomposition' strategy. It
is not clear what factors are involved either in stimulating the adoption of new strategies in
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 61
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
response to problems or in determining the selection of one strategy over others that are
available in relation to a given problem.
The relationship between number knowledge and strategy development
Torbeyns, Verschaffel and Ghesqiere (2005) give a hypothetical example of a child who is
able to accurately retrieve the answer to 6+6, but not 7+7 or 8+8. This child would be
expected to be more likely to use the tie strategy when solving 6+7 (by transforming the
problem to 6+6+1) than when solving 7+8 or 8+9. Torbeyns et al. showed that children in
their study differed in the efficiency with which they carried out counting, decomposition and
tie strategies, but that children at a range of different ability levels all showed similar levels of
adaptivity, generally choosing the strategy that would generate a correct answer most quickly
in response to a particular problem. Torbeyns et al. only analysed data from children who
were already using either the decomposition or tie strategy – their aim was to study variation
in adaptivity related to differences in achievement in mathematics, not to investigate the
necessary conditions for the development of these strategies.
Very relevant to the current discussion is the existence of a “tie effect” (LeFevre,
Shanahan, & DeStefano 2004), whereby the problem-size effect (the fact that arithmetic
problems with larger answers tend to be answered more slowly than those with smaller
answers) can generally not be observed for tie problems (where both addends in an addition
problem are the same). LeFevre et al. showed that the tie effect is not due to facilitation of
encoding (the fact that the same number appears twice means it is more quickly encoded the
second time), but is due to calculation and memory access.
It seems reasonable to argue, as do Torbeyn, Verschaffel and Ghesquiere (2005), that
good knowledge of doubling relationships (pairings between 6 and 12, 7 and 14 and so on) is
required in order for children to begin using the tie strategy. However, this paper aims to go a
step further and make the claim that implicit knowledge of doubles is a prerequisite for use of
the tie strategy.
Using priming to investigate number knowledge
The first study of number knowledge that employed priming as a method was that of den
Heyer and Briand (1986), in which a priming distance effect (PDE) was observed. The PDE is
the phenomenon that a reponse to a target stimulus is facilitated by the presentation of a prime
that is similar in magnitude to the target. For example, in the lexical decision task used in den
Heyer and Briand's study, participants were quicker to respond to 'five' after the prime 'four'
than after the prime 'three'. The PDE has been shown to be equally strong in both directions –
so the prime '4' facilitates processing of a target '5' as well as it does '3' – and has also been
demonstrated in different modes, whereby the prime 'six' facilitates processing of 'seven' or
'7', for example (Reynvoet, Brysbaert, & Fias 2002).
There is some debate regarding the mechanism underlying the PDE. Some researchers
have explained the effect in terms of operations on a 'sub-symbolic' number-line, used in
order to compare number information in terms of magnitude. However, there is evidence to
suggest that a connectionist approach might generate a more satisfactory explanation. There is
evidence, for example, that as well as proximity on the number line, other relationship
amongst numbers can give rise to priming effects. Garcia-Orza, Damas-Lopez, Matas and
Rodriguez (2009) show that, for adult participants, the prime '2x3' facilitates processing of
target '6', using a masked prime protocol. This suggests that, rather than sub-symbolic
processing, these priming effects reflect symbolic processing within something like Collins
and Loftus' (1975) semantic activation network.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 62
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Method
Participants
57 children, from two primary schools, took part in this experiment. They were aged between
7 years, 2 months and 9 years, 11 months. In each school, the Mathematics Coordinator was
asked to select those children who were able to reliably solve single-digit addition problems,
but did not yet consistently use a retrieval strategy. All of the children who participated in the
study had experienced classroom instruction in the use of a range of strategies for solving
addition problems, including counting strategies, decomposition and tie.
Instruments and measures
Two tasks were prepared, using the DirectRT psychology experiment software package.
Stimuli were presented to participants in the centre of a 17 inch monitor, using a 48 point
font. A microphone was used in order to measure verbalisation latency.
Addition problem task
For this task, a set of addition problems was created. All single-digit addition
problems with two addends, where the two addends were different and the answer was greater
than 10, were included. Participants were asked to respond with an answer to each problem.
Following an answer, participants were prompted with the question, “How did you solve the
problem x + y?” Problems were presented to participants at random, without replacement.
Strategies were coded as being either “count-one”, “min”, “decomposition”, “tie”, “retrieval”
or “other – including don't know”. Three practice problems were given before starting the
main set of problems. The practice problems were “7 + 3”, “6 + 2” and “4 + 4”.
Priming task
60 prime-target pairs were created. Of these, 15 pairs related to the present study.
Primes used were “5”, “6”, “7”, “8” and “9”. The target stimulus in each pair was either the
exact double of a prime, or the double +/- 1. The remaining prime-target pairs presented to
participants were included in order to ensure that participants could not predict that the
purpose of the study was to assess knowledge of doubles, and were constructed as if intended
to address participants' knowledge of number bonds and proximity on the number line. Primetarget pairs were presented at random, without replacement.
Timings were as follows: Fixation “*”: 1000ms; Prime stimulus: 200ms; Fixation “*”:
500ms; Target stimulus. Participants were instructed to say into the microphone the second
number in each pair (the target), as quickly as possible. The reaction time recorded for each
trial was the time it took for the participant to begin reading the target number, following its
presentation.
Design
The experiment employed a mixed design, with two independent variables. The first
independent variable was the relationship between prime and target in a prime-target pair.
There were 3 conditions of this variable; the target was either double the prime minus one,
double the prime exactly, or double the prime plus one. The second independent variable was
whether or not participants claimed to use the tie strategy at least once whilst completing the
addition problem task. The dependent variable was the time that it took to read aloud the
target stimulus.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 63
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The hypothesis was that reaction time would be least when the target was the exact
double of the prime and that this effect would be observed only for those participants who
used the tie strategy at least once during the addition problem task.
Procedure
Participants completed the two activities individually, in a quiet room outside of their usual
classroom. A laptop computer was used to generate stimuli and record reaction times via a
microphone. The experimenter watched the laptop screen during each trial, whilst participants
watched a second monitor, synchronised with the laptop.
Half of the participants completed the addition problem task first, followed by the
priming task, while half completed the two tasks in the reverse order. The researcher
introduced the task, and gave the participant an opportunity to ask questions. Three practice
trials were completed, followed by a further opportunity to ask questions. The block of
experimental trials for the task were then completed.
Results and Discussion
The addition problem task was used in order to divide participants into two groups. 29
participants reported using the tie strategy on at least one occasion during this task, while 28
participants did not.
A 3 x 2 mixed ANOVA was carried out. The independent variables were prime-target
pair (repeated measures: target = 2 x prime – 1; target = 2 x prime; target = 2 x prime + 1) and
whether children used the tie strategy at least once during the addition problem solving
activity (independent groups). The dependent variable was the median time it took for a
participant to begin reading the target stimulus.
There was a significant main effect of prime-target relationship (F2, 84=4.867, p=0.01).
This indicates that participants were significantly quicker to read a target that was exactly
double the preceding prime than a target that was 1 greater or 1 less than the exact double of
the preceding prime. There was also a significant interaction between prime-target pair and
strategy use (F2, 84=6.879, p=0.002). As can be seen in Figure 1, the effect of variation in
prime-target pair on RT is accounted for entirely by the group of children using the tie
strategy. There was no significant main effect of strategy use on reaction time.
Participants who used the tie strategy to solve at least one of the set of addition
problems were quicker to read a target that was the exact double of the prime than a target
that was the exact double of the prime plus or minus 1. The participants in the study had no
way of predicting the relationship(s) under investigation. This means that on perceiving the
prime stimulus, the double of the prime (amongst other cognitive resources including
numbers and concepts) was automatically activated.
When the target stimulus was the exact double of the prime, participants' reading of
the target was facilitated due to that number already having been activated.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 64
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Figure 1: Graph to show effect of prime-target pair type on RT, by strategy use
Thus this study clearly demonstrates the fact that children using the tie strategy have
implicit knowledge of the relationship between numbers and their doubles (at least for
numbers between 5 and 9). This result contributes substantially to the literature on both
children's arithmetic strategy development and its relation to the literature on the nature of
children's representation of number.
The findings do not directly identify a causal relationship between the development of
knowledge of relationships between numbers and their doubles and the development of the tie
strategy. However, of the two possible interpretations (either the implicit knowledge of
doubles is a necessary condition for the development of the tie strategy, or children's
knowledge of the tie strategy encourages the rapid development of knowledge of doubles) it
is intuitively most likely that children must develop a knowledge of the relationships between
numbers and their doubles before they can add the tie strategy to their repertoire. This fits fits
well within a resource activation framework (Hammer, Elby, Scherr, & Redish 2005). Within
this framework, cognitive resources are activated in response to a problem situation. These
resources are used in the assembly of ad hoc theory in order to generate a solution.
Further work must be done in order to fully understand the relationship between the
development of implicit knowledge and the development of strategies, but some important
implications should be considered at this stage. Most importantly, the study calls into question
the claim that teachers should be helping children develop ways to select amongst available
strategies for solving problems. Torbeyns, Verschaffel and Ghesquiere (2005) found there
was no difference in levels of adaptivity (the ability to select the most efficient strategy from
a repertoire of available strategies for a given problem) between children across a range of
mathematical ability. The present study shows that children do not use the tie strategy if they
do not have implicit knowledge of relationships between numbers and their doubles. Taken
together, the evidence from these studies shows that the development of new strategies, and
the development of the adaptivity necessary to select amongst strategies, occur as a result of
the development of associated cognitive resources such as the knowledge of particular types
of relations amongst numbers.
These results help to provide an explanation for some effects observed in previous
work. For example, Siegler and Stern (1998) observed that a majority of children in a study
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 65
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
were using a new strategy for solving a number of problems before they were aware of using
it. Within the resource activation framework, the learner's use of a particular procedure
(resulting from the automatic activation of relevant cognitive resources) and the learner's
representation of that procedure are quite different things. The representation of a particular
strategy will always follow that strategy's first use (whether it follows immediately or at some
later point).
Conclusion
This study represents an important step in our growing understanding of children's
development of mathematical thinking. Its main contribution consists in the argument that
implicit knowledge of relationships between numbers and their doubles is a necessary
prerequisite for the development of the tie strategy.
References
Alibali, M. W. 1999. How children change their minds: strategy change can be gradual or
abrupt. Developmental Psychology, 35: 127-45.
Collins, A. M., and E.F. Loftus. 1975. A spreading-activation theory of semantic processing.
Psychological review 82: 407–428.
Fuson, K. 1992. Research on Whole Number Addition and Subtraction. In Handbook of
Research on Mathematics Teaching and Learning, ed. D. A. Grouws, 243-275. New
York: Macmillan.
Garcia-Orza, J., J. Damas-López, A. Matas, and J. M. Rodriguez. 2009. " 2 x 3" primes
naming" 6": Evidence from masked priming. Attention, perception & psychophysics
71: 471-80.
Hammer, D., A. Elby, R. E. Scherr, and E. F. Redish. 2005. Resources, framing, and transfer.
In Transfer of learning from a modern multidisciplinary perspective ed. J. P. Mestre,
89–120. Information Age Publishing Inc.
den Heyer, K., and K. Briand. 1986. Priming single digit numbers: Automatic spreading
activation dissipates as a function of semantic distance. The American Journal of
Psychology 99: 315–340.
LeFevre, J. A., T. Shanahan and D. DeStefano. 2004. The tie effect in simple arithmetic: An
access-based account. Memory and Cognition 32: 1019–1031.
Reynvoet, B., M. Brysbaert, and W. Fias. 2002. Semantic priming in number naming. The
Quarterly Journal of Experimental Psychology Section A 55: 1127–1139.
Siegler, R. S. 2007. Cognitive variability. Developmental Science 10: 104-109.
Siegler, R. S., and E. Stern. 1998. Conscious and unconscious strategy discoveries: A
microgenetic analysis. Journal of Experimental Psychology-General 127: 377-397.
Torbeyns, J., L. Verschaffel, and P. Ghesquiere. (2005). Simple addition strategies in a firstgrade class with multiple strategy instruction. Cognition and Instruction 23: 1-21.
Verschaffel, L., J. Torbeyns, B. De Smedt, K. Luwel, and W. Van Dooren. 2007. Strategy
flexibility in children with low achievement in mathematics. Educational & Child
Psychology 24: 16-27.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 66
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Aspects of a teacher’s mathematical knowledge in a lesson on fractions
Bodil Kleve
Oslo University College
This paper is about a mathematics teacher, and how aspects of his mathematical
knowledge surfaced in a 5th grade (11 years old) fractions lesson in Norway. The
teacher’s responses to pupils’ (unexpected) comments and questions, ‘contingent
moments’, are discussed. Difficulties in dealing with improper fractions, which were
mirrored in the pupils’ inputs in the lesson, are discussed. Considerations are made
whether the problems the pupils expressed can be traced back to aspects of the teacher’s
mathematical knowledge.
Keywords: Improper fractions, teacher knowledge, contingent moments
Background and introduction
What knowledge is required for the teaching of mathematics has been widely discussed
within mathematics educational research, both with regard to what comprises the knowledge,
and how this mathematical knowledge is made accessible to others. Through classroom
observations and focus-group meetings with four mathematics teachers in 5th grade
elementary school in Norway I have been studying how teachers drew on their knowledge in
mathematics and mathematical didactics in their teaching. Lessons were videotaped and I
have used the Knowledge Quartet developed by Rowland, Huckstep and Thwaites (2005) as
an analytical framework to study how a teacher’s (Hans’) mathematical knowledge surfaced
in the lesson. How examples and illustrations of improper fractions influence the pupils’
conceptions and difficulties are discussed. Before presenting the lesson with the teacher Hans,
I will report some research about mathematical knowledge for teaching which I have used in
my study, and also briefly report research about fractions which suggests some factors
explaining why pupils’ concepts of fractions only become partly developed.
Mathematical knowledge for teaching
By questioning how teachers’ use their knowledge in the subject they teach and where
teachers’ explanations come from, Shulman (1986) brought didactics into mathematics
educational research. Shulman suggested distinguishing among three categories of content
knowledge: Subject Matter Content Knowledge, Pedagogical Content Knowledge and
Curricular Knowledge. Subject Matter Content Knowledge (SMK) refers to the knowledge
the teacher has in mind, both substantive and syntactic. Pedagogical Content Knowledge
(PCK) goes beyond knowledge of the subject and refers to content knowledge for teaching,
“It is the capacity of a teacher to transform the content knowledge he or she possesses into
forms that are pedagogical powerful” (Shulman, 1987, p.15). Curricular Knowledge is both
lateral and vertical. Lateral curriculum knowledge is how the teacher is able to relate the
content and issues discussed in his/her subject to that being discussed in other subjects.
Vertical curriculum knowledge is about what has been taught in earlier lessons (and years)
within a subject as well as what is relevant to be taught in the next lessons.
Rowland et al (2005) based their work on Shulman’s categories of knowledge.
Through a grounded approach to data from video studies, the Knowledge Quartet (KQ) was
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 67
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
identified. In the KQ the classification of the situations in which mathematical knowledge
surfaces in teaching is of importance (Rowland and Turner, 2008).
The Knowledge Quartet has four broad dimensions; Foundation, Transformation,
Connection and Contingency. Foundation is the mathematical knowledge the teacher has
gained through his/her own education, it is knowledge possessed and which can inform
pedagogical choices and strategies. It is the reservoir of pedagogical content knowledge you
draw from in planning and carrying out a lesson and thus informs pedagogical choices and
strategies. Transformation focuses on the teacher’s capacity to transform his or her
foundational knowledge into forms which can help someone else to learn it. It is about
examples and representations the teacher chooses to use. The third category, Connection,
binds together distinct parts of the mathematics and concerns the coherence in the teacher’s
planning of lessons and teaching over time and also coherence across single lessons.
Contingency is the category which concerns situations in mathematics classrooms that are
impossible for the teacher to plan for; the teacher’s ability to deviate from what s/he had
planned and the teacher’s readiness to respond to pupils’ ideas are important classroom events
within this category.
Fractions
As with decimals and percentages fractions occur with different meanings. These meanings
can also be seen in everyday life. A fraction can be a part of a whole, a place on the number
line, an answer to a division calculation or a way of comparing two sets or measures (part
group). Novillis (1976) studied the hierarchical development of various aspects of fractions
among American children. She found that the part-whole and part-group models were
significantly easier for the children to understand than the number line. Her study referred to
work with fractions not bigger than one. As opposed to the part whole or part group model,
the number line does not incorporate that a fraction can be thought of as a concrete object. But
according to Dickson, Brown and Gibson (1984), a number line makes improper fractions
appear more natural. They claimed that “the representation of fractions as sub areas of a unit
area does not lend itself very well to the representation of improper fractions” and that “The
acceptance of the definition of a fraction as meaning ‘part of a whole’ is inconsistent with the
very existence of such improper fractions” (p. 279).
According to Anghileri (2000) much of the focus when working with fractions in school is
identification of fractions as part of a whole. She claimed that success in working with
fractions depends on the ability to see the fraction both as representing a number and a ratio
which reflects the procedure for finding the number. She wrote:
Research suggests that an approach to fractions which identifies each as numbers to be
located on a number line, without emphasizing the way of partitioning a whole, will help to
establish the equivalence with decimals and percentages (p.115)
She thus warned against emphasizing fractions as parts of a whole in schools. This
is in accordance with Askew (2000) who claimed that if one focuses on fractions as
part of a whole so that becomes a social convention, possibilities for a obtaining a
well developed fraction concept are limited.
Hans
Based on an analysis of a lesson with Hans using the KQ as an analytical framework, I will
discuss how aspects of Hans’ mathematical knowledge became visible in this lesson, and if
difficulties in dealing with fractions bigger than one whole can be traced back to aspects of
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 68
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
the teaching. I will present an account of the lesson before I analyze it in terms of the aspects
of the KQ, followed by a discussion.
The lesson objective was written on the smart board when the lesson started: ‘To be
able to calculate with fractions which are bigger than one whole’. Hans asked the class what
the objective involved and a pupil suggested 4/3 or 1 1/3. The first task was 6/8+5/8 and it
was illustrated by two rectangles on the smart board. A pupil was taken to the board. He
worked it out and used the smart board to illustrate how one rectangle was filled up to a
whole and three remained in the other.
The next task was illustrated by circles divided into eight pieces. There were two
circles on each side of an equal sign. Five pieces in the first and four in the other on each side
were shaded. To Hans’ question how much it was all together Jens first suggested 11/9. When
the teacher did not confirm, Jens suggested 10/9. Hans then asked how much was shaded in
the first circle, and Jens said 5/9. ‘Are there nine?’ Hans then asked, and he let a pupil in class
answer 8. What about the next circle? Hilde suggested ½. Hans confirmed but converted the
half into 4/8 ‘for the sake of clarity’ he said. The answer 9/8 was agreed upon, and a pupil
came to the board converting it into 1 1/8. A girl, Petra, then demonstrated uncertainty
(misconception) asking: ‘How is it possible to take nine eighths? When there are eight bits
and then take one more? How is that possible?’ Pupils in class explained.
For the next two examples, Hans chose not to use illustrations; ‘let us try without’, he
said. The tasks to be done were: 3/5+3/5=, and 7/10+5/10=. For the last one a pupil suggested
converting 12/10 into 1 1/10. After having clarified that it became 1 2/10, Hans wanted to go
on to the next task, when Mads had his hand up suggesting converting 1 2/10 into 1 1/5 ‘like
the first one’, he said.
The last task was to take away ¼ from 2. Hans had drawn 2 circles on the smart board,
and he wrote 2-1/4 under the circles. The circles were divided into four pieces which all were
shaded. Espen was taken to the board to work it out. He said: ‘it is two minus one fourth’ and
he erased two ¼ pieces of the second circle. Mads then shouted: ‘Now you are erasing one
half’, followed by: ‘not one fourth of two’. Hans asked Espen if he was sure and how much
he actually should have taken away and Espen answered: ‘If it is two so half of that’ (pointing
to one of the circles). Hans emphasized ‘what is the whole’. Then Espen realized that he had
not done what the teacher had expected. He shaded one of the quarters he had erased and
wrote 1 ¾ =7/8. Then Hans asked him to explain what he had been thinking.
After this episode Ella went to the board and pointed to the digit 1 in 1 ¾ and asked:
‘what does the one actually do there?’ Hans responded to her question by taking it all from
the beginning, asking questions for Ella to answer throughout his review. He started by asking
what the whole was and how many pieces they originally had. Four more times during Hans’
review, Ella repeated her question. When he had finished and Ella expressed not
understanding, Hans erased the board and tried again. This time he just focused on 4/4 being
one whole. Then Ella said she understood.
Foundation
What subject knowledge (SMK) did Hans have on which he could draw in this lesson? In the
textbook fractions were illustrated as part of rectangles, as part of circles and on number line.
Hans used rectangles and circles, but not the number line. He demonstrated that he knew how
to make fraction tasks adding up to more than one. Hans ‘converted’ ½ into 4/8 ‘for the sake
of clarity’. He foresaw that writing ½ could cause confusion and wanted to avoid the issue of
common denominator at this stage. He also inserted an equal sign when that was missing in
Espen’s work with the subtraction task on the board.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 69
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Transformation
How was Hans’ foundational knowledge set out in practice? The lesson seemed to be well
planned. He had prepared examples and illustrations in form of rectangles and circles. The
smart board was used to illustrate how it was possible to fill up one whole and then ‘see’ what
fraction was left. There was a dialogic approach, Hans invited the pupils to participate and he
let them come to the board to work out the exercises he had chosen. He also let other pupils
explain when errors and misconceptions surfaced. Thus the pupils took actively part in the
lesson.
Hans’ choice of examples and illustrations mirror a view on fractions as part of a
whole. He did not use the number line. Neither did he use examples which mirrored fractions
as part of a group or proportions.
In the first example Hans used the expression ‘eleven out of eight’ which may have
caused the question from a pupil about how it was possible to take nine eighths, when there
were eight bits and then take one more was grounded in that expression.
All exercises included converting an improper fraction into a mixed number. In all but
the first example (which was taken from the text book), the fraction part of the mixed number
was a unit fraction. This suggests why a pupil converted 12/10 into 1 1/10 which again
reveals that she had an undeveloped conception about the link between an improper fraction
and a mixed number. The reason why the tasks Hans made all had a unit fraction as the one
remained is not clear. One suggestion is that Hans only thought about making the answer
bigger than one whole, and one bigger was sufficient. Another suggestion is that he wanted to
avoid abbreviations since that was not the goal of the lesson.
With regard to the illustration of 2-1/4, it seemed that it was the illustration itself that
caused Espen the difficulties. I suggest that Espen would have calculated 2-1/4 without any
difficulty if it had not been for the circles being used to illustrate.
Connection
There seemed to be a logical coherence in this lesson. There were good links across the lesson
with regard to progression and examples and illustrations. Hans started by asking what the
goal meant, and went from there to adding two fractions (6/8+5/8) which he had illustrated
with rectangles. Then he went on to another addition with fractions (5/8+4/8), illustrated with
circles. For the next two tasks (3/5+3/5 and 7/10+5/10), he suggested ‘trying with out
illustrations’. Then he went on to subtraction and chose to illustrate with circles again. This
task was not only different from the others in terms of not being addition. Now he started
from a whole number, 2, taking away a unit fraction (1/4). He also chose to illustrate with
circles. There may seem to be a gap between the addition tasks and the subtraction task. The
subtraction task was not only another calculation it also involved an integer which none of the
addition tasks did.
Contingency
There were four contingent moments in this lesson on which I will comment. First, Jens’s
answer to Hans’ question ‘how much is it all together’ when having shaded five in one and
four in the other. Why did Jens answer ninths, and how did Hans respond to it? I suggest that
Jens saw the nine shaded pieces, five in the first and four in the second circle as a whole, and
that was why he answered 5/9 when Hans broke the task down asking how many are shaded
in the first. From Jens’ point of view 5/9 was correct. It did not seem that Hans understood
how Jens was thinking. He responded to Jens as if Jens had done an unintended error (counted
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 70
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
9 pieces in the circle instead of 8). Hans did not invite Jens to explain why he kept saying
ninths. He let another pupil say eight, and he never came back to Jens and his answers. In this
case there was a lack of compatibility between the teacher’s and the pupil’s thinking. The
teacher responded to this contingent moment without incorporating it further into the lesson.
Second, I will comment on Petra’s input when she asked how it was possible to
become nine eighths; when there were eight bits and then take one more. That the question
was unexpected to Hans and that he was not prepared to answer it, were mirrored in how he
responded to it. He first repeated the question, then he stumbled before Mads inserted ‘you
have more pizzas you know’, Hans confirmed saying yes, and when Petra again asked: ‘One
pizza, and there are eight pieces of that’, Mads again emphasised his view by saying ‘and then
you take a new pizza’. In this case, Hans acknowledged Petra’s question, but left it for
another pupil to explain. An unanswered question is if he did that on purpose, or if the pupil’s
input helped him out of a problem which he was not able to answer right on his feet.
The third contingent moment in this lesson which I will discuss is Espen’s first
response to 2- ¼ . Obviously, the way Espen first worked out the task was due to that he
perceived 2 as the whole and he took ¼ away from that. He took away ¼ of 2 which is ½.
Also here, as in the case with Jens, there was a lack of compatibility between the teacher’s
and the pupil’s thinking. However, this time the teacher acknowledged the pupil’s
contribution and challenged him on how he had been thinking. Hans demonstrated an open
way of asking the child, acknowledging his thinking and incorporated it into the lesson.
In the last contingent moment on which I will comment, Hans acknowledged Ella’s
question why the 1 was there in 1 ¾. However, he did not incorporate Ella’s thinking in his
answer. He worked out the task over again posing questions and funnelling Ella through his
doing. Also when Ella repeated her question twice (‘but why is the one number there?’) Hans
carried on with his explanation. When Ella still did not understand, she asked ‘but why can it
not just be deleted then?’ Like in the case with Jens, Hans broke the task down, concentrated
upon one of the circles and 4/4 in that being the same as one whole. Then Ella expressed her
understanding.
Discussion
Hans responded differently to the contingent moments discussed above. I suggest that the
reason why he responded differently to different pupils was based on his knowledge about
these pupils’ different mathematical abilities. As with Espen, who was one of the best pupils
in class, Hans knew that Espen neither was making an occasional error, nor that he did not
understand. Therefore he challenged Espen. But did Hans understand why Espen had done
what he did?
In Hans’ lessons there was a confident atmosphere. Due to the atmosphere, the pupils
were actively participating by asking questions and commenting on what their classmates
were doing on the board. This confident atmosphere was created by the teacher through the
way he invited the children’s contributions. Hans’ knowledge about the pupils’ different
abilities and the confident atmosphere of his lessons are not aspects of a teacher’s knowledge
as part of the knowledge quartet. However, analysing a lesson through the use of the
knowledge quartet made these aspects of a teacher’s knowledge visible and has contributed to
draw an even broader picture of how different aspects of a teacher’s knowledge surfaced in
this lesson.
Finally, I want to discuss the contingent moments which I suggest mirrored some
common difficulties the children had in dealing with improper fractions. I have suggested that
Espen’s way of interpreting the task was grounded in the way it was represented with circles.
Although the teacher kept reminding the pupils what the whole was, pupils seemed to have
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 71
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
difficulties with it. Jens looked upon the shaded pieces as the whole. Petra did not understand
how it was possible to take nine bits when eight was the whole. And Ella argued why the
whole had to be there. This suggests a shortcoming with regard to the foundation and
transition aspects of Hans’ knowledge which incorporates research about what factors that
have shown to be significant with regard to pupils’ understanding or lack of such of fractions
bigger than one. He chose to use the same illustrations of improper fractions and mixed
numbers as he had done when dealing with fractions smaller than one, always emphasising
that fractions are parts of a whole. In this matter the illustrations he used seemed to cause
more difficulties than help in pupils’ work with fractions bigger than one. This suggests that
emphasising fractions as part of a whole and illustrating fractions as such, can explain some
of the difficulties revealed in this lesson.
References
Anghileri, J. 2000. Teaching number sense. London: Continuum.
Askew, M. 2000. What does it mean to learn? What is effective teaching? In Principles and
Practices in Arithmetic teaching, ed. J. Anghileri, Buckingham Open University
Press.
Dickson, L., M. Brown and O.Gibson. 1984. Children Learning Mathematics A teacher’s
guide to recent research, 134-146. Oxford: The Alden Press Ltd.
Novillis, C. 1976. An Analysis of the Fraction Concept into a Hierarchy of Selected
Subconcepts and the Testing of the Hierarchical Dependencies. Journal of Research in
Mathematics Education, 7:131-144.
Rowland, T., P. Huckstep and A. Thwaites. 2005. Elementary Teachers' Mathematics Subject
Knowledge: The Knowledge Quartet and the Case of Naomi. Journal of Mathematics
Teacher Education, 8(3): 255-281.
Shulman, L. S. 1986. Those Who Understand: Knowledge Growth in Teaching. Educational
Researcher, 15(2):4-14.
Shulman, L. S. 1987. Knowledge and Teaching: Foundations of the New Reform. Harvard
Educational Review, 57(1).
Turner, F., and T. Rowland. 2008. The Knowledge Quartet: A Means of Developing and
Deepening Mathematical Knowledge in Teaching? http://www.mathsed.org.uk/mkit/MKiT5_Turner&Rowland.pdf
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 72
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Post-16 maths and university courses: numbers and subject interpretation
Peter Osmon
Department of Education and Professional Studies, King’s College London
The low take-up of mathematics post-16 and consequences for the traditional STEM
(science, technology, engineering, and maths) subjects in higher education are well
known. The effect on the newer IT-based subjects, like computing and communications
engineering, and the commerce-based subjects, like business and management,
economics, and finance is less widely recognised but is at least an equal cause for
concern. Most university courses in these subjects are populated with students with no
maths beyond GCSE, despite the evident need for better mathematical foundationsperhaps a year of post-16 maths. The scale of this effect and the consequences for these
subjects in many university courses are described along with potential implications for
the AS-level curriculum.
Introduction: quantitative subjects, courses and professions
My experience during a lifetime in HE doing teaching and research in physics, then in
electronics, and then in computer science, and then working in academic management has led
me to identify a collection of subjects where students need at least one year of post-16 maths.
I call these the quantitative subjects but this does not imply mere familiarity with numerical
ways of working: it is intended to mean some degree of general mathematical competence is
required. Quantitative courses teach the various quantitative subjects and these courses tend
to be gateways to various quantitative professions. In fact the route comprising post-16
maths, then quantitative course and then quantitative profession is so well trodden as to imply
that setting students on this path is the main role of post-16 maths. However, many students
without post-16 maths also progress to quantitative courses and thence to the quantitative
professions.
Quantitative subject groups
The quantitative subjects are separable into groups, as follows:
A. Traditional STEM: Mainly physics based (plus maths of course), includesMaths, Stats, Phys, AerEng, CivEng, MechEng, ElecEng, and also Chem, ChemEng;
B. Post-IT revolution: Computer Science and Electronics-Optics technology based, includesCS, InfSc, AI, Electronics, Communications Technology;
C. Commerce based: include Economics and Actuarial Sciences (relatively traditional
subjects) and others (whose development is heavily indebted to IT) including Accountancy,
Finance, Bus&Mgt,
D. Bio-Tec based: Biology in combination with subjects from the other groups, includesBioChem, BioEng, BioInf;
M. Miscellaneous: includes Architecture and Medicine.
The allocation of subjects to these groups is not absolutely clear-cut. Thus, there are
subjects, besides those in group D, that cross group boundaries. But the above separation into
groups, which also corresponds roughly to the temporal sequence A, then B, then C, and then
D in which subject groups have developed, albeit with considerable time overlap, is proving
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 73
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
helpful when considering the mathematical needs of the various quantitative subjects.
However group D is at an early stage of development and so its needs are not discussed.
UCAS data on admissions to quantitative courses
My appreciation of maths requirements for quantitative courses, derived from my own
experience and observations and discussions with colleagues, has been greatly extended and
quantified by studying the data published by UCAS (UCAS, 2009). In this paper I use 2008
data for national student numbers accepted onto HE courses, and maths requirements for
admission in 2010, across a sample of universities, across subject groups A, B, C. The chosen
sample universities are ones where I have had some form of inside knowledge and are
approximately one per decile of university ranking, where the ranking is according to average
UCAS points score as given in The Times Good University Guide for 2010 (The Times,
2009).
The UCAS data, while immensely helpful, is not published in precisely the categories
needed for my purpose- for example electronics and electrical engineering courses are lumped
together. Also, the courses data is complicated by the existence of combined subject courses.
Where these combinations fall within my groups, for example Economics & Management, the
overall picture remains clear but where they cross groups, for example Maths & Management,
they blur the picture somewhat.
Organisation and presentation of the data
So as to give as clear picture as possible of the maths knowledge universities are requiring for
their quantitative courses and how this varies according to university ranking and subject
group I have summarised and transcribed the published data. (See Table 1 below).
Table 1 Maths qualifications for entry to Quantitative Subject HE courses
**
For clarity the Table is divided into three parts- one for each subject group (A), (B)
and (C). The columns divide subject groups into subjects and there is a row for each
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 74
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
university in the sample. Table entries are the corresponding maths knowledge level required
for entry. The Table is the most recent snapshot.
Student numbers by subject
For most subjects the picture is confused by the many distinct codes for subject variations and
course modes. It is apparent that lower ranking universities offer more combined subject
courses, for example Maths with Stats and OR, occasionally to the exclusion of a core course,
and also subject combinations which cross my subject groups. In my opinion these are
generally marketing devices to attract better qualified students. Student numbers given for
each subject are summed across all the variations (which may give rise to some double
counting).
Maths entry requirements
For clarity, so far as possible, I have focussed on the core code for each subject- single
subject, three year full-time course. The requirements shown are minimum entry
requirements.
“-“ means the subject is not offered.
“?” means a minimum maths knowledge for entry is not specified.
Where a subject is only offered in combination with an adjoining subject in the Table the
requirement is shown straddling the two subjects, thus “.GCSE.”
For clarity the only maths requirements shown are GCSE and A-level. (Equivalent
IB, Scottish Higher, etc, are not mentioned.)
GCSE maths grades: GCSE means Pass at gradeC, higher grades are shown explicitly.
AS-level maths: not present in the Table- presumably because the majority of students
go on to full A-level so it is not used as an entry condition.
Full A-level maths: the required grade eg “B” is shown.
Commentary on the Table content
Subject group (B)
The table shows only CS courses in this group. This is a consequence of the way university
courses are presented and UCAS data is collated: specifically courses in Electrical
Engineering and Electronics are lumped together, and courses focussing on communications
are also recorded there or else under CS, rather than as a separate subject. These (CS,
Electronics, and Communications) are subjects where the author has first hand teaching
experience and is aware that the mathematics needs differ from Electrical Engineering.
Correlation with university rank
Subjects offered. Chemistry and Physics tend to be offered only by high ranking universities
and these tend not to offer Business.
Maths needed for particular subjects. Some variation of maths grade is to be expected, but
where the same subject, eg Economics or Computer Science, requires A-level grade A at a
high ranking university and GCSE grade C at lower ranking universities, then it is
remarkable.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 75
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Student numbers in the subject groups
The numbers of students in each of the three subject groups are comparable. The combined
numbers in the newer groups (B) and (C) are double the number in the traditional STEM
group (A) and this is justification for emphasising the hitherto neglected maths needs of these
groups.
Low maths requirements turn quantitative subjects into qualitative ones
The inevitable consequence of having a low maths hurdle for entry, observed at first hand by
the author, is that subjects which are essentially quantitative have to be interpreted
qualitatively. (The alternative for universities would be to close most of their groups B and C
courses!) This in turn means the quantitative professions are fed graduates who are
inappropriately prepared. In the opinion of the author this fact implies a significant
downgrading effect on the national skills-base.
Mathematical needs of the quantitative subject groups
The government sponsored STEM initiative (STEM, 2007) has drawn attention to the national
importance of the traditional STEM subjects (group A) and the (full) A-level curriculum
seems to be associated with the maths needs of this group. In this paper I am more concerned
with the maths needs of the newer quantitative subjects, groups B and C, where most of the
quantitative student numbers are to be found and where most courses are only viable because
they are filled with students without post-16 maths knowledge.
The table shows that in quantitative subject groups B and C most of the students are
on courses that require no more than GCSE grade C maths. From direct experience of
teaching group B courses and discussion with colleagues who teach group C courses it is clear
that these students need 1 year of post-16 maths to engage effectively with their subject.
(This contrasts with the group A courses which generally require 2 years of post-16 maths.)
Remedial/foundational maths classes during the first year of the university course are
not very successful. This can be attributed to two factors: (a) the students had their last
(GCSE) maths class more than two years previously, so any mathematical knowledge and
thinking habits they may have acquired are very rusty, (b) these students typically disliked
maths at school and resent being given a dose of it at university when they really want to be
getting on with learning their major (which no-one told them is a quantitative subject!)
The remedial/foundational maths typically occupies about 25% of the first year of the
university course. 25% of one year is about the time that would be allocated to an AS level at
school. But it is noticeable that the maths being taught is rather different from the A-level
curriculum. It s also noticeable that the maths taught in group B courses is different from the
group C maths.
The following two questions emerge. (a) Can an AS-level maths curriculum be
devised that meets the one-year of post-16 maths requirements of groups B and C and also
the first year of the two-year maths requirement of group A? (b) Can this AS-level, somehow
become a requirement for entry to quantitative courses?
The bigger picture
The information in the Table is a snapshot taken during a particular year. But quite big
changes in quantitative subjects are occurring over time: there seems to be a pattern whereby
new subject groups develop and demand for courses in older ones declines relatively. The
timescale for significant change seems to be about a decade.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 76
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Thus, Traditional STEM subject have been haemorrhaging numbers since the 1970s
evidenced, for example, by closure of Chemistry and Physics departments at City and more
recently Chemistry at QML, as well as closure of many engineering departments. The
students who might otherwise have populated these departments appear to have migrated over
time in the direction left to right across the table: from group A to group B to group C. .
The first group B courses demanded A-level maths. But, as is well known, the
number of student with post-16 maths has not increased in step with the general expansion of
HE and hence much of the growth of group B and then group C numbers, except at the higher
ranking universities, has depended on students with GCSE maths only, as the Table shows.
From their inception most group C courses generally expected no more than GCSE maths
from their students- presumably because the large cohorts of students wanting admission to
these courses had no more maths to offer. If, as seems plausible, the currently emerging group
D grows rapidly during the next decade- following the growth pattern of the earlier groups B
and C- then presumably this will be at the expense of all three older groups and presumably,
unless there are dramatic changes in take-up of post-16 maths, these courses too will mostly
be populated with students having GCSE maths only.
Besides the development of new disciplines, which may be more attractive to students
than established ones, there are other mechanisms governing the changes in quantitative
subject provision. Universities are now conscious of being in a league table- ranked
according to the average (across all courses) UCAS score of their undergraduate intake.
Universities can optimise their scores by allocating students numbers to courses according to
the UCAS points they can attract, leading to closure of some courses and departments.
(Courses with small intakes have always been at risk of pruning, but the league table provides
a spur and a metric.)
University departments then market their courses to attract students with maximum
UCAS points- points scored becoming more significant than subjects studied- which can be
rationalised by treating UCAS points as a measure of ability. This in turn implies a more
foundational role for the first year of university courses.
In their turn secondary students will come to recognise this situation and since they
are free to choose the subjects they study post-16 and, since they are likely to perform best in
the subjects they enjoy most, this state of affairs seems purely benevolent. But what of
mathematics, which is not generally well liked (Brown, 2008) and maybe harder (Smith,
2004) than most subjects? The above scenario suggests fewer students will study maths post16- unless something changes so that more they find they like doing it!
Conclusion
The government sponsored STEM initiative has drawn attention to the number of students
gaining full A-level maths and recent improvement in this number, emphasising its
importance for the future of STEM subjects and the national skills-base. In this paper I have
attempted to shift the spotlight towards the newer quantitative subjects (groups B and C)
where the student numbers are greater than in traditional STEM (group A) - twice as many in
fact- and where most courses are only viable because they are filled with students without
post-16 maths knowledge.
Students acquire a dislike of maths at school and so generally don’t study it post-16.
Schools then pass the problem up to the universities to deal with. For their part universities
fail to clarify the importance of post-16 maths knowledge for studying quantitative subjects.
Lack of mathematical preparation weakens the courses so that students enter the quantitative
professions with only qualitative understanding of their subject. The problem would be eased
if courses in these newer subjects required their intakes to have one year of post-16 maths
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 77
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
(AS level). But of course they won’t be able to do this until large numbers of students choose
to take AS level maths- which seems unlikely until it becomes a course requirement: chicken
and egg!
Perhaps the way to improve take-up of AS-level maths is by devising an AS-level
curriculum that meets the one-year of post-16 maths needs of groups B and C and can also
serve as the first year of the two-year maths requirement of group A.
References
Brown, M., P. Brown, and T. Bibby, 2008. “I would rather die”: Reasons given by 16-yearolds for not continuing their study of mathematics. Research in Mathematics
Education 10, no. 1: 3-18.
Smith, A. 2004. Making mathematics count. London: The Stationery Office.
STEM, 2007. http://www.dcsf.gov.uk/stem/
The Times, 2009. http://extras.timesonline.co.uk/tol_gug/good university guide
UCAS 2009. http://www.ucas.ac.uk/
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 78
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The role of proof validation in students' mathematical learning
Kirsten Pfeiffer,
School of Mathematics, Statistics and Applied Mathematics, NUI Galway
The study of proofs is a major obstacle in the transition from school mathematics to
university mathematics. Given the importance of argumentation and proof in the
spectrum of mathematical activities, the incoming students' understanding, appreciation
and knowledge of the nature and role of proof must be considered. I describe the results
of an exploratory study of first year mathematics undergraduates' criteria and learning
process when validating mathematical arguments or proofs. The study is based on a
series of written tasks and interviews conducted with first year honours mathematics
students at NUI Galway. I presented the whole class with numerous proposed proofs of
mathematical statements, and asked them to evaluate and criticize those. The first year
students' written comments on different and partly incorrect 'proofs' of mathematical
statements revealed some information about their criteria when validating mathematical
arguments. In recently held interviews with eight randomly chosen students I focussed on
the learning experience during the process of proof validation. Considering the observed
learning effect and its large potential extent during the process of proof validation I
propose its practice in the teaching of mathematics.
Keywords: proof, proof validation, transition to University mathematics
In the first part of this article I explain what is meant by proof validation and consider various
aspects of this activity, including how and why mathematicians validate mathematical
arguments. I will then describe our experiments and findings, and consider the
undergraduates' validation skills and practices in relation to previously characterized aspects
of proof validation. I emphasize the learning effect during the process of proof validation and
finally argue for explicit inclusion of its practice in the teaching of mathematics, as the
development of validation skills not only improves the practice of validation itself, but also
the ability to construct proofs, the understanding of mathematical context, the knowledge of
proving strategies and the links between different areas of mathematics.
On proof validation.
Before considering how students validate proofs I discuss the nature of proof validation; I
further distinguish it from other types of reading and from construction of proof. Selden and
Selden (1995) call the readings to determine the correctness of mathematical proofs and the
mental processes associated with them “validations of proof”. I extend this description of
proof validation considering that mathematicians, advanced students or maths teachers
validate not only to determine the correctness of an argument. Discussions with a number of
experienced mathematicians suggest that they also wish to reach an understanding of why a
mathematical statement is true and often to understand the content and the position of the
proved statement in a wider context. Factors that experienced mathematicians might consider
while validating proposed proofs include: whether the argument provides the reader with
understanding,transparency and quality of proof idea and strategy, clarity of the structure,
whether the reasoning is precise, correct and sufficient, and whether the argument is
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 79
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
convincing. In Section 2 below I report on my investigation of the criteria that students
consider as essential for a valuable mathematical argument.
The activity 'proof validation'.
Validation, in comparison to the reading of non-mathematical texts, requires the reader to put
some additional effort into understanding of the reasoning. Validation usually takes more
time, the validator might consider the whole proof or parts of it several times and might be
more inclined to write a few notes checking deductions, verifying justifications, etc.
According to Selden and Selden (2003) the mental process when validating proofs can
include for example asking/answering questions, constructing subproofs or recalling other
theorems and definitions. It is well documented that construction of proof is a major obstacle
for students. Selden and Selden (2003) describe how the ability to validate proofs relates to
the ability to construct them. On the one hand proof construction and proof validation are
different. Proof construction requires 'the right idea' at the 'right time'. The validation process
can usually be managed in a linear order, unlike construction of proof. On the other hand
proof construction and proof validation entail each other as one considers during the process
of proof construction how that proof would be validated, and as validation of a proof is likely
to require the construction of subproofs. I summarize this relationship in the following
diagram.
Figure 1: Construction related to validation of proof
Considering my comments above on the nature of proof validation, I extend this diagram to
highlight the extent of the learning effect through the process of proof validation.
Figure 2: Proof Validation in the process of learning about mathematical proof
I summarize my hypothesis: The ability to validate proofs can improve the ability to
construct proofs, develop deeper understanding of the meaning and significance of the proved
theorem and develop knowledge of proving methods or strategies.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 80
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The experiment.
The study is based on several tests and interviews conducted with first year honours
mathematics students at NUI Galway. The students' way to validate mathematical proofs first
caught my interest when I was analysing their responses to a written exercise that I held in
May 2008, at the end of their first year at University. My research has focussed on that topic
since. Therefore the design of further research instruments concentrated on proof validation.
The test for the new incoming students (D-test08), held in September 2008 with 103
participants, included tasks designed to give me some insights into their proof validation
skills. Based on the findings of the analysis of the written exercises I designed interviews to
be held with a smaller number of students. The aim of these oral exercises was to get a deeper
insight in the students' validation processes. The interviews were held with eight
undergraduates in March 2009. I report below on findings arising to date from all three of
these exercises. Analysis of the data is ongoing. Some of the questions had been adapted
(with permission) from the Longitudinal Proof Project which ran 1999 until 2003 in the U.K.
Considering the previously described aspects of the process of proof validations I investigate
how our first year students validate mathematical proofs, which criteria they use to decide
whether an argument is correct or not and in the learning process during the exercise of
validating and comparing different mathematical arguments.
Written exercises.
The test held in May 2008 was attended by 37 participants. The students were presented with
six attempts to determine, with proof, whether the statement
“When you add any two even numbers, your answer is always even”
is true or false. For each of the attempts, I asked the students to give a mark out of five and a
line of advice.
Observations from written exercises: students' criteria when validating proofs.
The prevalence of some expressions indicates what the students found essential in a good
proof. I list below a number of themes that caught my interest, either because they appeared
quite often or because they surprised me.
The role of examples.
When confronted with a few examples to show the truth of the statement, I found that most
students recognized the necessity of rigour and commented on that. "He has just given
examples, this is not a proof" or "Not a general solution" are typical remarks. Some of the
students' comments indicate that "proof by example" might be seen as another type of proof,
just not as good as one including a general formula: "It's not bad. But it's only proved by
example.", "Although this does prove the statement, it only does so for a few egs...". On the
other hand, students do give examples a surprisingly high if not essential value within a proof.
They often deduct marks on the basis of absence of examples. Some students comment on the
absence of examples to explain the answers ("Give an example to show ...") but most criticize
the lack of examples as an essential part of the reasoning. Some students request examples in
(correct) answers to "back up proof" or to "finish the proof". One student criticizes a correct
answer because the statement is "not proven by numerical example". Without any examples
students don't seem to be satisfied by an argumentation.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 81
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
"Not mathematical enough"
The following are a few selected comments on a correct answer that was expressed purely in
the form of text: "The proof makes sense but she could have used a more mathematical
approach", "Good intuitive answer but needs a mathematical proof”, "Correct answer but
show mathematically", "The proof should be shown mathematically as well as in words". The
argument formulated as text was "not mathematical enough" to most of the undergraduates.
In comparison to another more algebraic looking approach one student comments: "Although
Cathy's answer is true there is too much English and does not mathematically prove it unlike
Aoife". These comments raise the question of what the students associate with the term
"mathematical". Their comments indicate that "mathematical" appears to mean including
formulas ("Try to come up with formula"), algebraic equations ("Give clear equation to
support your answer", "Would like to see this expanded with a general equation") and
mathematical notation ("Use mathematical notation to show this", "Cathy's answer is well
written and ... although she should sum her answer up .. using formal notation").
Consequently another answer, which is fundamentally and irreparably incorrect but includes
algebraic equations and mathematical notation, seemed basically correct to more than half of
the students. Fewer than 30% of the students recognized that this answer was wrong.
A visual approach to prove the statement is generally not accepted by the students.
The proposed answer consisted of a diagram showing how an even number can be represented
by two rows of dots, and how addition of even numbers can be interpreted as concatenation of
two such representations. Most students interpreted the answer as just one example,
visualized in a diagram. "Again Finn's answer only covers 1 solution. He needs to give a
general statement.", "This proves that it works for 12+8. It doesn't prove for all cases.", "Not
a proof, just an example" are a few typical answers. 27% recognized the idea behind the
illustration, but didn't acknowledge a graphical representation of the correct idea as a
mathematical proof: "Good visual proof but use mathematics", "Good visual representation
but needs notational explanation". A proof without numbers and words can't be sufficient:
"There are no words in this proof", "Proof is illustrated using graphics rather than numbers",
"This does not prove anything, words and numbers are needed", "There are no words in this
proof".
"it doesn't explain..."
Some mathematical educators argue that whether or not an argument is accepted as a proof
depends not only on its logical structure, but also on how convincing the argument is. That
aspect seems to play a role in the students' proof evaluation as well. "Nice pictures, you could
have written a line explaining it though", "Intuitively correct but needs to explain why the
answer means the statement is true", "Need more explanation" or "She should explain what
she is doing". The positive comments on the highly marked answers often include a note
about the good explanations: "Aoife has a very clear and straightforward answer", "Well
explained answer" or "Aoife is using clear and simple language to get her answer across..."
are a few comments on the students' favourite answer. Those comments indicate that just
having a good idea to prove a statement is not sufficient for the students. The skills of
convincing and explaining ideas to others matter to them.
I summarize that after their first year in university most students are aware that checking the
truth of a statement for a few examples is not sufficient to prove the statement. On the other
hand examples play an important role in mathematical argumentation to students. Even after
accepting the correctness of an argument they are not convinced until it is shown with a few
examples. Overall the students' picture of proof seems to be vague. To them a valuable proof
should have a certain structure, starting with a definition, followed by some algebraic
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 82
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
equations or formulas, and finishing with some examples. Structure and formalism seem to be
more important to the students than the idea behind the proof. If these requirements are met
most of the students give at least a few marks regardless of the correctness of the particular
steps or whether the overall idea makes sense to them or not. It seems, a good idea to prove a
statement is not being valued as highly as the structure and formalism of a proof. Beside
structure and formalism the quality of explanations played a role in the students' proof
evaluations. It seemed that if an answer didn't show attempts to convince the reader of an
argument, most of the students would deduct marks.
Oral exercises.
In March 2009 I held interviews with eight randomly chosen students who had attended the
written exercise in September as well. The aim was to get a deeper insight into
• students' opinions about valuable proofs. What do students mean when they use the
term “mathematical”? Do they ask for examples in order to understand the given
reasoning or because they consider them as essential part of proofs?
• students' validation process. How do they attempt the task of validation? For example,
do they read the proposed proofs line by line? Do they write notes, verify the
arguments?
• the learning effect during the validation process.
To facilitate comparison of the results one of the statements chosen for consideration in the
interviews was similar to the one in the written exercise, but a bit more difficult.
The squares of even numbers are even, and the squares of odd numbers are odd.
The other statement was different from the exercises the students had performed in the
context of this study so far.
Let f be a quadratic function, f(x) = ax² + bx + c with a,b,c є R and a>0.
Show: f can't have more than two common values with its derivative f'.
Again the students were confronted with five or six different arguments, some incorrect or
partly incorrect, and asked to comment on them and finally rank them. Some of the proposed
proofs were algebraic, some visual, some written in text, others wrong but expressed using
“typical” mathematical notation.
Observations from the oral exercises: the learning process when validating and comparing
different mathematical arguments.
A detailed analysis of the interviews is in progress. In structuring and partly transcribing
them a remarkable pattern caught my attention. The students were very quick in deciding
whether they liked an answer or not and their first opinions and comments were similar to
those in the written exercises. During the meeting though I could observe a process of
understanding when spending more time with the task, comparing ideas with some proofs
they have seen somewhere else, and sometimes even questioning their own criteria.
Especially when ranking the different answers the students reconsidered their opinions and
sometimes changed their minds about certain answers. These findings are in line with
Selden/Selden and Alcock/Weber:
• Selden and Selden describe that students' performance in distinguishing valid from
invalid arguments improved dramatically through the reflection and reconsideration
during the interview. (Selden and Selden, 2003)
• “Our results suggest that many of the students in our study could perform this task
competently, but did not do so without prompting.” (Alcock and Weber, 2004)
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 83
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
The fact that the students were forced by the ranking task to spend some time on their
reflections encouraged a learning process. I conclude that the understanding of mathematical
concepts can improve considerably during the process of careful proof validation.
Conclusion
The findings show on the one hand that undergraduates have a vague yet inflexible picture of
valid proofs. Structure and 'mathematical' looking formalism seem more important to them
than the idea behind its appearance. On the other hand I discovered that reflection during the
process of proof validation encourages a learning process about the nature of mathematical
proofs. Recalling the discussion in the first part of this article about the nature of proof
validation, in particular the relation between construction and validation of mathematical
proof and the attainment of understanding through validation, I conclude that practice of
proof validation can not only improve students' validation skills but can also lead them to a
better understanding of mathematical content and to improved appreciation of deductive
reasoning.
References
Alcock, L. and K. Weber. 2005. Proof validation in real analysis: inferring and checking
warrants. Journal of Mathematical Behaviour 24: 125-134.
Selden, A. and J. Selden. 2003. Validations of proofs written as texts: can undergraduates tell
whether an argument proves a theorem? Journal for Research in Mathematics
Education 34(1): 4-36.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 84
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
An exploration of mathematics students’ distinguishing between function and arbitrary
relation.
Panagiotis Spyrou, Andonis Zagorianakos
Department of Mathematics, University of Athens, Greece
This paper focuses on students’ awareness of the distinction between the concepts of
function and arbitrary relation. This issue is linked to the discrimination between
dependent and independent variables. The research is based on data collected from a
sample of students in the Department of Mathematics at the University of Athens. A
number of factors were anticipated and confirmed, as follows. Firstly, student difficulties
involved vague, obscure or even incorrect beliefs in the asymmetric nature of the
variables involved, and the priority of the dependent variable. Secondly, there were some
difficulties in distinguishing a function from an arbitrary relation. It was also thought that
additional problems occur in the connotations of the Greek word for function, suggesting
the need for additional research into different linguistic environments.
Introduction
The concept of function is essential in the understanding and learning of mathematics. It is
considered to be the most important concept learnt from kindergarten to college or university
(Dubinsky & Harel 1992). The difficulties students experience with this concept can only be
understood in relation to its definition and the appearing of cognitive obstacles. Several
researchers found that in the early stages of function teaching in secondary schools that
natural models dominate using mainly 1-1 (one-to-one) functions. (Evagelidou, Spyrou,
Gagatsis, & Elia 2004; Elia & Spyrou 2006).
The reliance on the natural models means that the connection between the dependent
and independent variable is emphasized rather than focusing on the priority of dependent to
the independent variable. Furthermore, the natural models which are offered to the pupils are
idealized, distant from the realities from which they were created and described in analytical
formula, thus making it “difficult for the students to distinguish between relationships
discovered by experience and the mathematical models of these” (Sierpinska 1992, 32). This
approach results in a difficulty in realizing that the dependent variable is a magnitude which is
used to estimate a measurement and that the independent variable is the means for this
particular purpose, with or without an analytical formula.
The etymology of the Greek word for “function” introduced a note of caution. The root
of the Greek word for function (“synartisi”) is different from the origin of its Latin equivalent
which is mainly operational. In colloquial Greek when a person or abstract phenomenon
such as time, speed or measurement has a functional relation (“synartate”) with another
person or abstraction, the effects tend to be symmetrical. A bond is implied, whether active or
inert, which is triggered when “one side” (usually either side) is altered, evoking a change in
the “other side”. Therefore, the common perception of the Greek word for “function” implies
the symmetry of the function variables. This symmetry might create a difficulty in
understanding the difference between the variables in the mathematical definition, i.e. which
is the means and which is the one to measure. Sierpinska (1992) recognized this difficulty as
the obstacle, “regarding the order of variables as irrelevant” (p. 38). This definition of the
obstacle is the starting point of this paper.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 85
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
To conclude, the over reliance on one-to-one correspondence in function teaching and
the common perception of function creates an obstacle which may persist throughout Higher
Education. It was therefore decided to research the persistence of this obstacle in the thinking
of students in Higher Education.
Theoretical framework
The definition of function went through several stages until it reached its present form. This
progressive development has given rise to a number of epistemological obstacles. The first
definitions of the concept of function by Bernoulli, Euler and Cauchy were not complete. This
is because they saw the symmetricality of the dependent and independent variable, in the
context of a relationship. Sierpinska marks this “moment” in the concept’s theory as an
epistemological obstacle: EO(f)–5 (Unconscious scheme of thought) “Regarding the order of
variables as irrelevant” (1992, 38). It was Dirichlet in 1837 that used his study on Fourier
series and the conditions under which a Fourier series converges, and formulated a general
definition of function: "if a variable y is so related to a variable x that whenever a numerical
value is assigned to x there is a rule according to which a unique value of y is determined,
then y is said to be a function of the independent variable x” (Boyer 1968, 600). Dirichlet’s
definition of function is still in use and his main conclusion states the necessity of the
dependent coordinate being uniquely determined and not always the inverse. Therefore, the
Dirichlet definition expressed precisely, for the first time, the notion of a mediated measure
in the concept of function. That is to say: to estimate the dependent variable y, and to achieve
it although there is no immediate access to y, is to measure it through x. Therefore the
independent variable is the mediating variable which gives access to the dependent variable,
resulting in the priority of the latter.
The literature on the study of the epistemological obstacles that occurred through the
development of the definition of function is particularly rich (e.g., Freudenthal, 1983; Sfard,
1992; Dubinsky & Harel, 1992; Sierpinska, 1992; Even and Tirosh, 1995). However, it is
difficult to find any research on the particular obstacle which is the subject of this research.
As discussed in the introduction, the teaching of function during the first years of
school is oriented towards a common perception of function, emphasizing the relation
between the dependent and the independent variable, disregarding the priority of the
dependent variable. Furthermore, the dominant use of one-to-one functions makes it harder
for students to recognize the importance of the dependent variable, which is the target of the
measurement. In addition, the focus on the relational aspect of biunique functions conceals
the richness of the applications which the definition allows.
Finally, the common perception for teaching function in school can be seen as a
collection of habits, whereby “the character of a habit depends on the way in which it can
make us act” (Peirce 1958, Vol. V, par. 18). It is this habitual comprehension of function,
diverging from the formal definition, which results in mathematical obstacles and difficulties.
Thus, we formed the following research questions (accompanied by their relation to the
questions of the questionnaire that was given to the students —the justification for which is
given in the next section):
[1] Can the students recognize the difference between the concepts of function and of an
arbitrary relation? (1st, 2nd question of the questionnaire)
[2] Can the students distinguish the order of the variables x and y, and the asymmetry that
they have? (3rd, 4th question of the questionnaire)
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 86
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Methodology
The methodology arose from the theoretical framework of this paper and sets out to test the
persistence of the mathematical obstacle described above, in students in Higher Education. It
was anticipated that students would encounter problems in recognizing the priority of the
dependent variable and in distinguishing between a function and arbitrary relation.
Within this theoretical framework it was decided to design a survey in two phases.
Phase 1 included the completion of a 4 questions questionnaire by the students (see
Figure 1 below). The format of the questions in the questionnaire included two types of
questions:
1] Crosscheck items corresponding to questions: (key Q=questionnaire)
A) (Q1) Students were given 3 correspondences on graphs and another 3 with table values.
They were asked to find out which represented functions.
B) (Q3) Students were given 4 functions on a graph. They were asked which of the 4 would
still be function if the lines on the graph were rotated by 90°.
2] Open-type questions with short answers such as:
A) (Q1) Students were asked to make the necessary changes to the graphs and table values to
change the arbitrary relations into functions.
B) (Q2) Students were asked to give two examples of (arbitrary) relations which were not
functions, one algebraic and one represented graphically.
C) (Q3) Students were asked to justify whether the 4 functions on the graph remained
functions when turned 90°.
D) (Q4) Students were asked in which case(s) the 90° rotation of a function’s graphical
representation represents a function and to give a general rule.
Figure 1 The Questionnaire given to the students
❶Which of the following relations are function relations? Make the necessary corrections to the rest of them, in order to
transform them into functions.
x
-1
0
1
2
3
y
0
1
2
3
4
x
-3
-2
-1
0
1
y
2
2
2
2
2
x
5
-3
5
0
-3
y
3
2
1
1
6
❷Give two examples of arbitrary relations that are not functions (one described graphically and one analytically (with an
algebraic formula)).
❸What happens to the graphical representations of the following functions if the line on the graph is rotated by 90°? Are they
still functions? Give a short justification in your answer.
❹In which situation(s) does a 90° turn of a function’s graphical representation represents a function? Which general rule
would you use?
The questions in the questionnaire were designed to correspond to the research
questions posed in the theoretical framework, as follows:
A) The 1st question asks students to distinguish which correspondences are functions and
which are arbitrary relations. The 2nd question asks them to give two examples, one graphical
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 87
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
and one with an analytical description, of an arbitrary relation, which does not fit the function
definition. The answers to both of these questions correspond to the 1st research question.
B) The 3rd question asks students to distinguish when a function remains a function on a
graph when the graphical representation is rotated by 90°. In the 4th question they were asked
to interpret this movement and to justify their answers with reference to the application of the
general rule. This 90° turn on the graph occurs as a consequence of the permutation between
the dependent and independent variables. Therefore, the students are tested in their ability to
realize the asymmetry of the x and y variables, i.e., the 2nd research question.
The questionnaires were distributed to 17 students who attended lessons in
“Epistemology of Mathematics”, in April 2009, in the Mathematics Department of the
University of Athens. The course was chosen because of the accessibility to a wide range of
students on different courses, theoretical and applied, and at different stages of study. Student
participation was voluntary; the questions were answered without time constraints, taking
them approximately half an hour to complete.
Phase 2 of the research took place after about 3 weeks (May 2009) and included semistructured interviews, with 13 students out of the 17 who participated in the questionnaire.
Students were fully informed of the objectives of the research and gave their permission for
their interviews to be recorded. In all interviews the interviewer emphasized that the purpose
of the interview was not to examine the students but was to explore what they thought when
they answered the questionnaire, regardless of whether the answers were right or wrong.
Each interview aimed to clarify the answers given on the questionnaire and the
problems the students had encountered completing it. Students were asked additional
questions in order to start a general discussion and explore their understanding of function
with regard to their school and university education. They were asked about (a) their
experience of the concept of function at secondary school, (b) their acquisition of the concept
of function at the university, and (c) what they thought the use of function was outside the
mathematical frame.
All the interviews were audio-recorded and listened to, in order to assess them against
the research questions. This assessment showed a consistency in the interview results. Four
representative interviews were selected, using the following criteria:
(i) Variety of students’ responses to the research questions as shown in both the questionnaire
and the interview.
(ii) The interviews comprised different levels of understanding: 1 high (Georgia), 1 moderate
(Diana), 1 low (Iris). The 4th interview highlighted the findings of the research (Thanos). The
most relevant parts of the interviews for the research were transcribed. These were then
divided into 5 minute intervals or divided according to the research questions, and
accompanied with short comments. Due to the limitations of space this paper contains just a
few but characteristic dialogues from the interviews.
Description of the results — Discussion
Summarizing the main findings of our research, we observed the following:
R1) All students interviewed, except one, gave only examples of 1-1 (biunique) functions.
They said they had used examples they recalled from learning functions at school.
R2) All students interviewed, except one, had difficulties in giving good explanation, or any
explanation, for the “many-to-one” condition in the definition.
R3) The students used mnemonic rules:
A) Seven students out of 17 (41%) gave the same example (the circle example) as a graphical
representation of an arbitrary relation that is not a function (Q2). Moreover, all the
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 88
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
interviewed students except one used inclusively 1-1 examples of functions, an attitude
that shows the strong connection that the students still have with their early function
experience. This connection was admitted extensively in the interviews.
B) Extract from Diana’s interview, typical of the definitions used as mnemonic rules and the
confusion that follows:
D: I try to recall the definitions, as far as I can. When I cannot, I “put my mind to work".
R: Does your “mind” ... agree with the definitions?
D: On this occasion it agrees, not always.
R4) 29,5% of the students (5/17) gave the wrong answers to the 1st as well as the 2nd question
of the questionnaire, where they were asked to recognize the difference between the concepts
of function and arbitrary relation.
R5) 47% of the students (8/17) gave the wrong answer to the 3rd question of the questionnaire,
revealing a difficulty in distinguishing the order of the variables x and y, and their asymmetry.
R6) 64,7% of the students (11/17) gave the wrong answers to the 4th question of the
questionnaire, revealing a difficulty in distinguishing the order of the variables x and y, and
their asymmetry.
R7) Most of the students interviewed have only been concerned with functions in the context
of their school and university education. Nevertheless, their education did not equip them
with the necessary tools for interpreting the concept of function.
Iris is a typical example: although she reported that she always thinks of graphical
representations when she deals with functions, she still had difficulties in giving examples of
arbitrary relations which are not functions. She did not understand what a 90º rotation of the
graph meant despite knowing the formal definition of a function and applying it correctly in
the 1st question.
R8) The interview with Thanos is indicative of the students’ confusion with the “many-toone” condition of the definition. For instance, concerning the use of functions, he reported as
follows:
“Wherever I want to put factors say, the x and y are essentially factors. (For example), x is
able to measure temperatures and y to count days. Or x to count children and y to estimate
the tax return. That is, apart from the fact that we put it in a two-dimensional frame and
take a mathematical perspective; the two dimensions (the 2-axes coordinate system) are
essentially two parameters. The three dimensions are three parameters, and so on.”
It is apparent that he has misunderstandings concerning the priority of the dependent
variable. He also considers x1, x2, x3, as parameters of the function f(x1,x2,x3), which “lead” to
the y variable, thus showing that he is confusing the function of many variables with the
“many-to-one” condition of the definition.
The results from the questionnaires and interviews confirmed the problematic areas
anticipated at the outset of the research. It is evident from the results of the questionnaires and
the interviews that students experienced difficulties in answering all four questions. The most
difficult question is question 4 (R6) where 2 out of every 3 students experienced difficulties.
When incorrect answers for question 4 were combined with the incorrect answers for question
3, where almost half were wrong (R3), they provided evidence of the difficulties students have
in distinguishing the order of variables x and y, and their asymmetry.
Although there is a smaller percentage (29.5%) of wrong answers for the 1st as well as
the 2nd question (R4) their weakness indicates the persistence of the problem. They show
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 89
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
difficulties students have in recognizing the difference between the concepts of function and
of arbitrary relations.
The results from the interviews (see R1, R2, R3, R7 and R8) provided data that
revealed more than the questionnaires. The majority of the students had separate ideas about
the definition and its application. All of the students knew a formal definition. Only 2 students
gave examples of “a single-valued but not uniquely invertible function”. Many students still
experienced difficulties when asked about this specifically in their interview. The dominance
of the biunique functions is further evidence of the dominant influence of the first years of
function teaching.
From the evaluation of the questionnaire data we conclude that there are difficulties in
students’ abilities to recognize the difference between the concepts of function and of
arbitrary relation. However, most difficulties occur in students’ abilities to distinguish the
order of the variables x and y and their asymmetry, confirming Sierpinska’s (2002) claim
(EO(f) – 5).
In our opinion, the etymology of the Greek word has an additional negative impact in
students’ ability to overcome the mathematical obstacle, by encouraging a common
perception in favour of the biunique function. This common perception is handed down
through a school’s teaching methods, often remaining unchallenged. We think it would be
worth testing the same mathematical obstacle in different environments, to isolate the
influence of the etymology of the Greek word and its’ part in the persistence of the obstacle.
References
Peirce, C. S. 1958. Collected papers of Charles Sanders Peirce, Volumes I-VI (C. Hartshorne
& P. Weiss, Eds.). Cambridge: Harvard University Press.
Boyer, C. 1968. A history of mathematics. New York: Wiley & Sons.
Freudenthal, H. 1983. The didactical phenomenology of mathematical structures. Dordrecht:
Reidel.
Dubinsky, E., and G. Harel. 1992. The nature of the process conception of function. In E.
Dubinsky and G. Harel (Eds.), The concept of function: Aspects of epistemology and
pedagogy 85-106. Washington, D.C.: Mathematical Association of America.
Sfard, A. 1992. Operational origins of mathematical objects and the quandary of
reification—The case of function. In E. Dubinsky & G. Harel (Eds.), The concept of
function, Aspects of epistemology and pedagogy 59-84. Washington D.C.: The
Mathematical Association of America.
Sierpinska, A. 1992. On understanding the notion of function. In E. Dubinsky & G. Harel
(Eds.), The concept of function, Aspects of epistemology and pedagogy (pp. 25-28).
Washington D.C.: The Mathematical Association of America.
Even, R., and D. Tirosh. 1995. Subject-matter knowledge and knowledge about students as
sources of teacher presentations and subject matter. Educational Studies in
Mathematics 29: 1-20
Evagelidou, A., P. Spyrou, A. Gagatsis, and I. Elia. 2004. University students’ conception of
function. In M. J. Høines and A. B. Fuglestad (Eds.), Proceedings of the 28th Conference
of the International Group for the Psychology of Mathematics Education 2: 251-258.
Bergen University College, Norway.
Elia, I., and P. Spyrou. 2006. How students conceive function: A triarchic conceptualsemiotic model of the understanding of a complex construct. The Montana Mathematics
Enthusiast 3(2): 256 -272
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 90
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Identifying and developing the mathematical apprehensions of beginning primary
school teachers
Fay Turner
Faculty of Education, University of Cambridge
In this paper I present a summary of a four year study into the development of
mathematical apprehensions in beginning elementary teachers using the Knowledge
Quartet as a framework for reflection on, and discussion about, mathematics teaching.
The term mathematical apprehension is used as an inclusive term to cover both
mathematical content knowledge and conceptions of mathematics teaching. Evidence
from three case studies suggest that focused reflection using the Knowledge Quartet
facilitated the development of mathematical content knowledge and promoted positive
changes in conceptions about mathematics teaching. Experience and working with others
in classrooms and schools were also seen to influence development and change in the
teachers’ apprehensions. However, individual reflection was found to have a mediating
role on the influence of these two social factors.
Introduction
The way in which teachers teach mathematics is influenced both by their mathematical
content knowledge (Ball, 1988) and by their conceptions about mathematics teaching
(Thompson, 1992). The mathematical content knowledge of elementary teachers has been
found to be insufficient for teaching (Brown, Cooney and Jones, 1990; OFSTED, 2000).
Researchers have also found that elementary teachers often have unhelpful conceptions about
mathematics and mathematics teaching (Brown, McNamara, Jones and Hanley, 1999). Initial
teacher education courses alone seem unable to produce necessary developments in
mathematical content knowledge (Carré and Ernest,1993; Williams, 2008) or to promote
sustained positive changes in conceptions of mathematics teaching (Brown et al, 1999) in
beginning teachers.
The aim of this study was to investigate the effectiveness of a sustained approach to
developing the mathematical apprehensions of beginning elementary teachers. The study
began with the conjecture that, supported reflection on the mathematical content of teaching
might promote developments in mathematical content knowledge and changes in conceptions
of mathematics teaching in beginning teachers. The Knowledge Quartet framework
(Rowland, 2008) was used to facilitate such reflection both as a means to, and as a measure
of, professional development. Three theoretical frameworks underpinned the study.
Theoretical frameworks
Mathematical content knowledge for teaching
A theoretical framework for consideration of teachers’ mathematical content knowledge was
derived from the seminal work of Shulman (1986; 1987) and from the work of Deborah Ball
and colleagues in Michigan. Shulman’s three ‘knowledge bases’ which relate specifically to
the content of teaching; subject matter knowledge (SMK), pedagogical content knowledge
(PCK) and curriculum knowledge (CK) provided a foundation. The division of SMK into
substantive and syntactic knowledge (Schwab, 1978) also informed my research as did
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 91
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
refinements of Shulman’s categories by Ball, Thames and Phelps (2008). They identified
common content knowledge (CCK) and specialized content knowledge (SCK) as subdivisions
of SMK, and knowledge of content and learners (KCL) and knowledge of content and
teaching (KCT) as subdivisions of PCK. The Michigan group also include the category of
knowledge on the horizon as an aspect of SMK and knowledge of the curriculum as an aspect
of PCK. The Table 1 below illustrates the relationship between the categories of Shulman
and of the Michigan group and represents the model used in this study as a framework for
investigating the teachers’ mathematical content knowledge.
Table 1
Mathematical Content Knowledge
Subject Matter Knowledge (SMK)
Pedagogical Content Knowledge (PCK)
Common
Specialist
Knowledge onKnowledge of Knowledge of Knowledge of
Content
Content
the Horizon Content and Content and the Curriculum
Knowledge Knowledge
Teaching
Learners
(CCK)
(SCK)
(KCT)
(KCL)
Conceptions of mathematics teaching
The models of teachers’ conceptions of mathematics teaching identified by Kuhs and
Ball (1986) provided the basic framework for this study although other models, such as those
of Ernest (1989) and Askew, Brown, Rhodes, Johnson and Wiliam (1997), were drawn on
where appropriate. Kuhs and Ball identified four dominant models: a classroom-focused
view; a content-focused with an emphasis on performance view; a content-focused with an
emphasis on conceptual understanding view and a learner-focused view. Ernest (1989)
identified six models which were very similar to those of Kuhs and Ball, but included two
extra categories combining characteristics from Kuhs and Ball’s models. This refinement was
found to be generally unnecessary in my research although I drew on it where appropriate. I
also drew on the work of Askew et al (1997) which identified three orientations in teachers’
conceptions about mathematics teaching; a transmission orientation, a connectionist
orientation and a discovery orientation.
An approach to developing mathematical content knowledge and changing conceptions of
mathematics teaching
My approach was based on the model of professional development through reflection
in and on practice (Schön, 1983). However, the role of reflection was investigated in relation
to ideas from socio-cultural theory models of professional development. In social theory
knowledge is seen as situated in social situations and the development of knowledge as
resulting from enculturation or socialisation into the professional culture (Cobb, Yackel and
Wood, 1991; Lave and Wenger, 1991; Wenger, 1998). I recognised the role of socio-cultural
factors in my study and expected teachers’ apprehensions, as revealed through observations
and discussions of their teaching, would be contingent on the context of their teaching. I also
recognised that the individual reflection of participants would be made within the
communities of practice of schools, and would reflect the role and relationships of the
participants within schools. Jaworski (2007) suggested that communities of practice (Wenger,
1998) became communities of inquiry when teachers worked collaboratively to reflect on and
develop their practice. It was hoped that the participants in my research group would become
such a community of inquiry. The rationale for this study was therefore that teachers should
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 92
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
be supported to use reflection on their mathematics teaching within the social contexts of their
teaching and also of the research project.
The Knowledge Quartet framework
The KQ framework was developed by mathematics educators at the University of
Cambridge from observation and videotaping of mathematics teaching (Rowland, 2008).
Analysis of this teaching produced 18 ‘emergent’ codes (Glaser and Strauss, 1967) of
situations in which mathematical content knowledge of teachers was made visible, e.g.
‘concentration on procedures’ and ‘making connections between concepts’. These were later
classified into four ‘superordinate’ categories based on associations between the original
codes. These categories make up the four dimensions of the Knowledge Quartet; foundation,
transformation, connection and contingency. The foundation dimension includes the
propositional knowledge of SMK and PCK that teachers draw on in their practice, as well as
their beliefs about mathematics and mathematics teaching. Transformation encompasses the
ways in which a teacher’s own knowledge is transformed to make it accessible to learners and
connection includes issues of sequencing and connectivity as well as complexity and
conceptual appropriateness. The final dimension of the KQ, contingency, could be described
as ‘thinking on one’s feet’ and is concerned with the way teachers respond to unexpected
student responses.
The study
The study began with 11 student teachers from the 2004-5 cohort of elementary (5-11
years) postgraduate pre-service teacher education course at the University of Cambridge
reducing, as anticipated, to 4 in the fourth and last year of the study. Data came from
observation and analysis of teaching using the KQ as well as from post-lesson reflective
interviews, group and individual interviews and participant written reflective accounts.
Transcripts of interviews and written reflective accounts were all systematically coded using
the qualitative data analysis software NVivo 7. A grounded theory approach (Glaser and
Strauss, 1967) was used which led to the emergence of a hierarchical organisation of codes
into a number of themes.
Case studies were built from the analysis of observed teaching as well as from
analysis of interviews and the participants’ reflective accounts. Six themes in the development
of the participants’ mathematics teaching emerged from the NVivo analysis. These were,
beliefs, confidence, subject knowledge, experience, reflection and working with others. The
KQ analysis of observed teaching provided a ‘spine’ for presenting findings in relation to the
development of participants’ mathematical content knowledge. This was supported by data
from interviews and reflective accounts organised under the themes of subject knowledge and
confidence. Findings in relation to changes in conceptions about mathematics teaching drew
primarily on data organised under the theme of beliefs, and were supported by lesson
observation data. Data from the themes of experience, reflection and working with others
gave insight into the influences on developments in the participants’ mathematical content
knowledge and into influences on changes in their conceptions of mathematics teaching.
Findings
Development of mathematical content knowledge
Looking at the teachers’ content knowledge in relation to the foundation dimension of
the KQ suggested that development in propositional PCK, or knowledge of content and
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 93
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
learners, knowledge of content and teaching and knowledge of curriculum, were greater than
in SMK, or common content knowledge (CCK), specialised content knowledge (SCK) and
knowledge on the horizon. Where development in SMK did take place, it was in relation to
SCK rather than CCK. Reflection on practice helped the teachers identify areas of their SCK
that needed development. Development in these areas was achieved through support from
me, attendance at in-service training or through self-study. It is unlikely that such
developments would have occurred through reflection alone.
The teachers’ active PCK, as revealed through the Transformation, Connection and
Contingency dimensions of the KQ, was also seen to have developed over the study. The
three teachers appeared to focus on different aspects of their Transformation knowledge, and
this focus was reflected in the apparent developments in their practice. All three teachers
made more effective use of demonstrations but this was most apparent in Amy’s teaching. A
more considered use of representations was a strong feature of Kate’s practice and Jess
showed particular development in relation to her use of examples.
All the teachers considered making connections to be important for effective teaching,
and demonstrated this throughout the study. Focusing on connections developed the teachers’
practice in different ways. Amy concentrated on making connections to individual children’s
understanding and interests, Kate became more likely to make connections between
mathematical ideas and Jess increasingly focused on the connections between calculation
operations. The ability to act contingently also became a focus for the teachers, and they all
increasingly considered this to be integral to effective teaching. Responding to children’s
needs and ideas became central to Amy’s early years practice. Kate became particularly
proficient at acting contingently to find alternative representations to address misconceptions
or lack of understanding and Jess became more likely to respond to children’s errors and to
try to understand their mathematical thinking.
Changes in the teachers’ conceptions of mathematics teaching
Each of the teachers held complex views of mathematics teaching, incorporating
elements from all four of Kuhs and Ball’s (1986) views of mathematics teaching. Although
the balance of these elements varied, there appeared to be a pattern in the direction of change
in the three case studies. The teachers moved towards a focus on developing conceptual
understanding and towards a learner-focused view of mathematics teaching. All the teachers
had elements of a classroom-focused view at the beginning of the study but this quickly
diminished. Throughout the study, the three teachers demonstrated content-focused
conceptions of mathematics teaching with differences between them in the balance between
an emphasis on performance and an emphasis on conceptual understanding. However, for all
three, the direction of change over the four years was towards a greater emphasis on
conceptual understanding. Amy demonstrated some emphasis on performance at the
beginning but moved to a strong emphasis on conceptual understanding. Kate and Jess
increasingly appeared to emphasise conceptual understanding although both retained strong
elements of emphasis on performance.
Amy began with a learner-focused view of teaching and this appeared to be
strengthened over the study. Kate and Jess also appeared to focus increasingly on the needs of
their pupils. However, the way in which they interpreted these needs differed. Amy’s
interpretation was the most consistent with constructivist ideas inherent in a learner-focused
view of mathematics teaching. Kate also increasingly tried to understand children’s
mathematical thinking but remained focused on helping them achieve pre-determined
products or processes. Jess appeared to interpret the needs of her pupils in terms of their
‘ability’ and the need for them to achieve success in order to develop confidence. Amy
increasingly took an inquiry or problem solving approach to her teaching, consistent with a
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 94
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
learner-focused view of mathematics teaching.
approach, but found it more difficult to adopt.
Influences on development and change
Kate and Jess also moved towards this
Experience of teaching mathematics, and working with colleagues, were found to be
important influences on the teachers’ mathematical apprehensions. However, reflection
mediated the effects of these two influences. It was reflection on experience which catalysed
developments in mathematical content knowledge, and the focus of the teachers’ reflection on
experience which reinforced or changed their conceptions about mathematics teaching.
Working with colleagues was also found to be an important influence in developing
mathematical content knowledge and conceptions of mathematics teaching. This influence
was quite different for each teacher. Amy’s learner-focused view of mathematics teaching
was shared by her colleagues and reinforced this view. However, Kate questioned some of
the practices in her school and Jess was uncomfortable with the way in which mathematics
was taught in her first post. All three teachers reflected on the principles and practices of their
schools facilitating alignment (Wenger, 1998) for Amy, leading to critical alignment
(Jaworski, 2006) for Kate and to non-alignment with her first school for Jess.
Reflection using the KQ to focus on the mathematical content of teaching helped the
teachers to focus on their own and on their pupils’ understanding of mathematical concepts,
and this supported developments in their SCK. Their reflection also informed the teachers’
thinking about how to transform their own understanding in order to make it accessible to
their pupils, and this supported developments in their PCK. Where the teachers focused on
the effectiveness of their teaching their KCT was enhanced, and where they focused on the
understanding of learners, their KCL was enhanced. The KQ focused the teachers’ reflection
on the content of the mathematics and on the children’s engagement with, and learning of,
that content. This supported movement towards conceptions of mathematics teaching that
emphasised conceptual understanding, and which were learner focused.
References
Askew, M., M. Brown, V. Rhodes, D. Johnson, and D. Wiliam. 1997. Effective Teachers of
Numeracy. Report of a study carried out for the Teacher Training Agency 1995-96 by
the School of Education King’s College London.
Ball, D. L. 1988. Unlearning to Teach Mathematics. For the Learning of Mathematics, 8(1):
40-48.
Ball, D. L., M. H. Thames, and G. Phelps. 2008. Content knowledge for teaching: What
makes it special? Journal of Teacher Education, 59(5): 389-407
Brown, S. I., T. J. Cooney, and D. Jones. 1990. Mathematics teacher education. In
Handbook of Research on Teacher Education , ed. W.R. Houston, 639-656. New
York: Macmillan.
Brown, T., O. McNamara, L. Jones, and U. Hanley. 1999. Primary student teachers’
understanding of mathematics and its teaching. British educational Research Journal,
25(3): 299-322.
Carré, C. and P. Ernest. 1993. Performance in subject-matter knowledge in mathematics. In
Learning to Teach, eds. N. Bennett and C. Carré, 36-50. London: Routledge.
Cobb, P., E. Yackel, and T. Wood. 1991. Curriculum and teacher development: Psychological
and anthropological perspectives. In Integrating research on teaching and learning
mathematics, eds. E. Fennema, T. P. Carpenter and S. J. Lamon, 83- 120. Albany, NY:
SUNY Press.
Ernest, P. 1989. The knowledge, beliefs and attitudes of the mathematics teacher: a model’,
Journal of Education for Teaching 15: 13-33.
Glaser, B. G. and A. L. Strauss. 1967. The discovery of grounded theory: Strategies for
qualitative research. New York: Aldine de Gruyter.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 95
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Jaworski, B. 2006. Theory and practice in mathematics teaching development: Critical
inquiry as a mode of learning teaching. Journal of Mathematics Teacher Education 9:
187-211.
Kuhs, T. M. and B. L. Ball. 1986. Approaches to mathematics: Mapping the domains of
knowledge, skills and dispositions. East Lancing:Michigan State University, Center
on Teacher Education.
Lave, J. and E. Wenger. 1991. Situated learning: Legitimate peripheral participation. New
York: Cambridge University Press.
Office for Standards in Education (OFSTED). 2000. The National Numeracy Strategy: The
First Year: A Report from HM Chief Inspector of Schools. London: HMSO.
Rowland, T. 2008. Researching teachers’ mathematics disciplinary knowledge. In
International handbook of mathematics teacher education, eds. P. Sullivan and T.
Wood, 1: 273-298. Rotterdam: Sense
Schön, D.A. 1983. The Reflective Practitioner, New York, Basic Books.
Shulman, L. S. 1986. Those who understand, knowledge growth in teaching. Educational
Researcher 15 (2): 4-14.
Shulman, L. S. 1987. Knowledge and teaching: Foundations of the new reform. Harvard
Educational Review, 57(1): 1-22.
Schwab, J. J. 1978. Education and the structure of the disciplines. In Science, Curriculum
and Liberal Education, eds. I. Westbury and N. J. Wilkof, 229-272. Chicago:
University of Chicago press.
Thompson, A. G. 1984. The Relationship of Teachers’ Conceptions of Mathematics and
Mathematics Teaching to Instructional Practice. Educational Studies in Mathematics,
25(2): 105-127.
Wenger, E. 1998. Communities of practice: Learning meaning and identity. Cambridge:
Cambridge University Press
Williams, P. 2008. Independent review of mathematics teaching in early years settings and
primary schools, Department for Children, Schools and Families (DfCSF).
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 96
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
What Might We Learn From the Prodigals? Exploring the Decisions and Experiences of
Adults Returning to Mathematics
Robert Ward-Penny
Institute of Education, University of Warwick
This paper reports on a research project which explored the decision making process and
the experiences of adults who had returned to mathematics after a significant period of
time away. Data was gathered using a combination of a questionnaire and follow-up
interviews with selected participants. This paper presents some of the key findings
together with some examples from the stories of these learners. Finally, it argues that
these ‘prodigals’ offer a vivid reminder of the role of mathematics as cultural capital, and
an additional perspective on many issues of current interest in mathematics education.
Introduction and Background
At the time of writing, the twin issues of disaffection and underachievement appear
prominently in many pieces of research (for example, Nardi and Steward 2003) and published
reviews of teaching and learning (for example, Smith 2004). Against this rather pessimistic
background there exist a significant number of individuals who voluntarily return to learning
mathematics in one form or another. Having met a number of these individuals personally, I
questioned how we might best consider their actions against the wider background; do they
contradict our conceptualisations of disaffection and underachievement, or are they the
‘exceptions that prove the rule’?
In order to focus this research around this apparent contradiction, this research was
designed to include only a subset of adult learners: those whose initial schooling had taken
place within the United Kingdom, and who had spent some time away from formal education
before beginning their current study of mathematics. This last criterion gave rise to the
collective moniker ‘prodigals’, after the biblical story of the prodigal son, and excluded
learners who had begun adult learning courses directly from school.
The three key research questions can therefore be posed using this new term: firstly,
who are the prodigals? What are their demographic characteristics, and is there sufficient
diversity within the group to suggest that there are different types of prodigal learners in
mathematics? Secondly, what motivates the prodigals to return to the study of mathematics?
How might we conceptualise their decision-making process? Finally, how do the prodigals’
experiences of learning mathematics as an adult compare to, and contrast with, their
experiences of learning mathematics at school? By combining the answers to these three
questions, we can begin to consider what we might learn from the prodigals.
Construction of the Sample
Constructing a sample that accurately reflects the prodigal community is problematic; in the
words of Coben (2003, 73), “experience tells anyone who has ever worked with adults that
there is no such thing as a generic adult learner of numeracy.” Furthermore, there is
considerable diversity in the prodigals’ experiences, and some sub-groups were difficult to
isolate, such as those undertaking informal, community-based courses, as well as learners
who were studying using independent tutors and entering exams as private candidates. The
sample for this initial exploratory research was drawn from learners in official programmes,
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 97
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
connected to a school or further education institution. Although this was primarily due to
matters of convenience, there is some evidence to support this approach: the 2006 report into
the Skills for Life programme reported that “more than two million of the 2.4 million people
taking up courses by July 2004 undertook them in further education.” (House of Commons
Committee of Public Accounts 2006, 6)
In an attempt to increase the range of backgrounds included in the study, participant
groups were selected from two contrasting regions, one centred on a major city, and another
which was largely non-metropolitan. In fact, the eventual results demonstrated little or no
differences between regions. The groups for inclusion were selected through a process of
negotiation with staff working in adult education and a consideration of each group’s
background; groups where the majority had been educated outside of the United Kingdom
were excluded for the reasons discussed above. Participant groups were also chosen so as to
encompass both adult numeracy courses and GCSE (General Certificate of Secondary
Education) or GCSE-equivalent courses.
Methodology
The research was conducted in two stages. The first stage consisted of a questionnaire. After a
promising pilot, this was distributed to participants (n=66). It contained a combination of
open and closed questions, which are discussed below, together with the results. After some
preliminary analysis, the questionnaire was followed up by six face-to-face semi-structured
interviews, and one interview conducted by e-mail. The questions in these interviews were
designed not only to help support the generation of a narrative of each individual’s experience
of learning mathematics, but also to support the development of a convergent validity
regarding common responses. This concern also influenced the choice of participants – for
example, since a number of questionnaires had mentioned that ‘memory’ was a concern when
learning mathematics as an adult, at least one participant who had included this response on
their questionnaire was selected for interview.
The focus on the individual places this research firmly within an interpretive
paradigm, drawing on some of the ideas of grounded theory. It was conceived and conducted
as an exploratory study, and an attempt was made to minimise dependence on any one
theoretical framework during the study; this is due in part to the nature of the research, but it
is also consequent of the objects of study; Coben (2003, 110) comments that “explicit
reference to a theoretical frame is constrained by the under-theorised state of the field.” The
analysis of the interview data drew on the phenomenographic tradition, managing the
transcripts by sorting features qualitatively into broad categories which allowed for
subsequent analysis.
All participants volunteered to take part in the research. Its aims and purposes were
explained to them, anonymity was ensured and the option to withdraw was offered to
participants at each stage of the research.
Questionnaire Results: Demographic Characteristics
The first section of the questionnaire concerned the participant’s gender, age, and previous
educational history. The majority of the sample (80%) was female, a tendency which has been
observed elsewhere in adult education contexts (for example, Benn and Burton 1994; Coben
et al. 2007). This gender bias was statistically independent of course type (χ2 = 0.310, df=1,
p=0.578).
The age distribution of the participants was roughly normal, centred at about 35.
However, there was a highly statistically significant relationship between age and course type
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 98
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
(χ2 = 12.810, df=3, p=0.005). The age profile of the adult numeracy learners demonstrated a
negative skew, and the age profile of the GCSE learners demonstrated a positive skew. One
contributing factor towards this skew could be the presence of learners who were retaking
their GCSE within a decade of leaving school. This suggestion is supported by the
information gathered regarding participants’ previous educational history. Of 39 GCSE-level
participants, 17 had already attained a GCSE grade in mathematics and were seeking to better
it. (This proportion increases to more than half if CSE and O-level grades are included.) This
high proportion could be interpreted as a consequence of the role of the GCSE qualification as
a gatekeeper.
Interestingly, 12 of the 26 participants on adult numeracy courses had also previously
achieved a GCSE or CSE grade in mathematics. This belies the existence of a uniform
pathway through adult qualifications in mathematics. Although some learners might begin
with a numeracy qualification and then move onto GCSE, this model does not fit all learners.
Another possible stereotype challenged by the results of this section was that of adult learners
of mathematics possessing a low general level of education. There was a huge range of
responses in terms of previous educational history, ranging from no qualifications to a degree
in Fine Art. Finally in this section, 84% of those taking an adult numeracy course indicated
that they had undertaken a previous adult education course, compared to 25% of the GCSE
level learners (χ2 = 21.467, df=1, p<0.001).
Questionnaire Results: The Decision to Return to Mathematics
The second section of the questionnaire concerned the participant’s decision to return to
studying mathematics. Issues of motivation are, of course, difficult to assess and summarise;
writers such as Hamilton and Hillier (2006) suggest that the decision to return to mathematics
should be considered as both gradual and sudden, and this dual perspective was followed up
in the interviews. Despite its basic approach, however, this section of the questionnaire
yielded interesting results.
Firstly, a question about the timescale of the decision suggested that most learners had
been considering taking the course for a substantial period of time; the modal response was
‘for significantly longer than a year’. (This tendency seemed much more pronounced in the
females than the males, but the difference was not statistically significant.)
Next, participants had to indicate their motives for returning to the course. Initially
they were required to tick which motives from a list of twelve (drawn from the literature, and
together with a thirteenth ‘other’ option) they felt were relevant to them. They then had to
indicate which motive was the single most important one. Whilst space constraints prevent a
full summary of the results here, intrinsic motives such as confidence and personal
development scored highly with the numeracy learners, whilst extrinsic motives focused on
the GCSE qualification itself scored more highly with the GCSE learners. This became
especially obvious when the most important reason was selected – 22 out of 31 GCSE level
learners selected “I need a qualification to help me get onto another course.” No other reason
scored more than 3 votes.
Questionnaire Results: Learning Mathematics as an Adult
The third and final section of the questionnaire encouraged the participants to compare their
experiences of learning mathematics as an adult with their school experiences. This was done
through a combination of paired Likert scales and open-ended questions.
Typically, the adult experience was portrayed much more positively, with ‘as an adult’
responses tending to be about two grades higher on a five-point Likert scale than their
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 99
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
corresponding ‘at school’ responses. The validity of these increases was supported by the
accompanying open-ended responses. These comments related to a number of clear themes;
for example, the role of the teacher in the learning process, and the use of contexts and
applications.
“I didn’t have things explained at school, work was put in front of you and you were
expected to do it.” (GCSE Learner)
“Hated my school maths teacher. He didn’t make maths fun. My current maths tutor is
fantastic, if you don’t understand she will explain it in as many different ways as possible
until you do. She also relates all our maths to everyday life.” (GCSE Learner)
Some of these were inevitably shaped by the structure of the questionnaire, but in
other places issues which I had presumed to be relevant (such as the use of ICT) were
ignored, and others became apparent instead. These themes continued to be present in the
interviews. Although some participants were less enthused about learning mathematics than
others, (and one openly resented being forced to take the course in order to progress in their
wider education,) the overwhelming tone of the comments was very positive, both in terms of
the participants’ attitudes towards mathematics and their own self-confidence. One participant
wrote that “it has been rewarding to change past negative messages”; another rejoiced in “the
fact that after 30 years of thinking I can’t do maths I can!”
Summary of the Interview Dialogues
As discussed above, the interviews served to explore issues such as the decision making
process in a finer level of detail, as well as offering me an opportunity to clarify meaning.
Many of the earlier findings were confirmed or further exemplified in the individual
narratives that emerged. For example, the role of mathematics as a gatekeeper qualification
was again evident, with many participants needing a qualification for a promotion or for
entrance to another course. The role of the teacher also continued to be prominent in many
stories, both at school and as an adult:
“It was far more informal – there was no sort of, ‘yes sir, no sir, three bags full sir’, you
know, if I had a question I could ask *****. It was far more informal, almost on a sort of
friendly level – you know, just a friend who could do maths, whereas before it was sort of,
you know, Mr. Such-a-body…” (Adult Numeracy Learner)
A related concern was the issue of fault and the apportioning of blame. It was
interesting to see how different participants interpreted their experiences and allocated
responsibility for their perceived ‘failure’, both onto themselves and to others, and this might
offer some insight into the reliability of prodigals as narrators.
Other themes that re-emerged included the issue of explanation and context, and the
role of mathematics as a functional toolkit:
“There’s still things I can’t do, and don’t understand, but in general I’m a lot happier… for
instance, the course is sort of designed around real life, so if I went into a shop and it said
that there was seventy-five percent off, I’m now able to stand there and work out how much
I am actually saving. So yeah, there are lots of situations in life where I am now using
number, where, I feel a lot happier… yeah, definitely.” (Adult Numeracy Learner)
As discussed above, some issues became prominent at the interview stage that had not
been directly examined in the questionnaire. One of these was the effect of ability grouping
on achievement and motivation. Interestingly, negative experiences were reported by
prodigals who had been placed in bottom sets, middle sets and top sets:
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 100
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
“Yeah, because I think the highest I could get was a ‘D’ anyway, and I didn’t even get that,
I got an ‘F’, so it was almost pointless really… I think because I couldn’t do it I just lost
interest.” (GCSE Learner – Bottom Set)
“It was like they were constantly pushing you. And the way they’d split the classes up, so
you had bottom maths, middle maths, higher maths, it was like – the person in higher maths
was… ‘oh yeah, but I’m better at maths than you, I’m in a higher group than you’, and all
that sort of stuff, and we never had that in college, because you’re all in the same group.”
(GCSE Learner – Middle Set)
“I was put in the wrong group really, because I was put in the top group, and I had no idea
what he (the teacher) was on about most of the time. So I switched off, really. I think if I’d
started in a lower group I would have found my feet and then maybe been able to progress
up. But at the top group there was all the really bright kids, and I had no idea what I was
doing.” (GCSE Learner – Top Set)
Issues of space prevent a full discussion of the accounts that made up the interviews,
but other key issues raised included the role that some of the participants’ desire to help their
children played in their decisions, and the influence that learning mathematics and conquering
certain fears has had in developing a wider academic self-confidence.
Discussion
Each of the three research questions identified above has been answered in part, but each also
lends itself to further study. Preliminary demographic features of the prodigals have been
identified, and these have some resonance with pre-existing research into the general adult
education community. However, both the demographic results and the differences revealed in
the decision making process point towards there being at least two different types of
prodigals, which are connected to the two different types of course considered. It would be
interesting to extend this research to prodigals undertaking A-level courses, or distance
learning degrees in mathematics, and see how their responses compared to the two groups
explored above. This study has also produced some preliminary conclusions about the
motivations behind the prodigals’ decisions, and also explored how their experiences of
learning mathematics as an adult compare (generally favourably) to their experiences at
school. However, the question still remains: what might we learn from the prodigals?
Perhaps the most obvious, and encouraging thing that we can take from this group of
learners is proof at the level of the individual that negative attitudes towards mathematics can
be changed, and negative experiences can be overcome. Moreover, there is some evidence
that whilst an improvement of one’s general confidence and academic self-concept is not
always an explicit motive for returning to study mathematics, it is often a consequence. It is
possible for learners to improve what has been termed as a ‘mathematical trajectory’ (Noyes
2007), often with great benefits to the individual concerned, and sometimes with a refreshing
enthusiasm. “For the first time in my life,” one learner wrote, “I actually enjoyed solving
maths problems. Weird or what?”
Beyond this, the results outlined above form a vivid reminder of the role of
mathematics as cultural capital in the sense of Bourdieu (1973). A proper deconstruction of
the meaning of this term as it applies to mathematics, and the relevance this has to the stories
detailed above lies outside the scope of this summary, but the role that mathematics plays as a
gatekeeper is indisputably clear throughout both the questionnaire and interview results.
Finally, it is also striking how often, and how strongly, the narratives gathered touch
on issues that relate to current discussions in mathematics education, such as the
consequences of ability grouping in the classroom, or the place of contexts in demonstrating
relevance. Whilst this is undoubtedly partly an artefact of the methodology, the frequency and
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 101
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
nature of the occurrences suggests a genuine phenomenon. Possibly, then, the prodigals offer
an additional perspective on these issues. Although their viewpoint in undoubtedly biased, it
could be challenging in a healthy way; perhaps they are the ‘exception that proves the rule’.
Robert Ward-Penny holds an ESRC sponsored scholarship at the University of Warwick.
References
Bourdieu, P. 1973. Cultural Reproduction and Social Reproduction. In Knowledge, Education
and Cultural Change, ed. R. Brown. London: Tavistock.
Benn, R. and R. Burton 1994. Participation and the Mathematics Deterrent. Studies in the
Education of Adults 26(2): 236-249.
Coben, D., ed. 2003. Adult Numeracy: A Review of Research and Related Literature.
London: NRDC.
Coben, D., M. Brown, V. Rhodes, J. Swain, K. Ananiadou, P. Brown, J. Ashton, D. Holder,
S. Lowe, C. Magee, S. Nieduszynska and V. Storey 2007. Effective Teaching and
Learning: Numeracy. London: NRDC.
Hamilton, M. and Y. Hillier 2006. Changing Faces of Adult Literacy, Language and
Numeracy – A Critical History. Stoke on Trent: Trentham Books.
House of Commons Committee of Public Accounts 2006. Skills for Life: Improving Adult
Literacy and Numeracy – Twenty-First Report of Session 2005-06. London: HMSO.
Nardi, E. and S. Steward 2003. Is Maths TIRED? A Profile of Quiet Disaffection in the
Secondary Mathematics Classroom. British Education Research Journal 29(3): 345367.
Noyes, A. 2007. Rethinking School Mathematics. London: Paul Chapman Publishing.
Smith, A. 2004. Making Mathematics Count. London: HMSO.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 102
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Design Decisions: A Microworld for Mathematical Generalisation
Eirini Geraniou, Manolis Mavrikis, Celia Hoyles, Richard Noss
Institute of Education, London Knowledge Lab
This paper provides the preliminary analysis of a study in which year 7 students
interacted with eXpresser; a microworld designed to support students’ transition
from the ‘specific’ to the ‘general’ by constructing figural patterns of square tiles
and finding rules to describe their model constructions. We present evidence that
support three key design decisions of eXpresser and discuss how these features
facilitate students’ expression of generalisation.
Introduction
Utilising young students’ natural algebraic ideas and developing them through carefully
designed digital media appears to be a fruitful avenue to explore in teaching of algebra.
Students can verbalise algebraic rules in natural language but struggle to use mathematical
language (Warren and Cooper 2008). They often fail to see the rationale, let alone the power,
of generalisation. The MiGen project4 is tackling this problem by supporting 11-14 year old
students in their problem-solving during generalisation tasks, and providing them with a
rationale for finding and checking general constructions and rules. The core of the MiGen
system is a microworld, named eXpresser, in which students can build figural patterns of
square coloured tiles and express the rules underlying them. The eXpresser seeks to provide
students with a model for generalisation that could be used as a precursor to introducing
algebra, to help them develop an algebraic ‘habit of mind’ (Cuoco et al. 1997). In this paper,
we will focus on student interactions with specific functionalities of expresser arising from
three of the key design decisions (all of the design decisions are extensively described in Noss
et al. 2009, Geraniou et al. 2009).
The Microworld, the eXpresser
This section provides a short description of expresser (for a detailed
description the reader is referred to Noss et al. 2009, Geraniou et al.
2009). Students are presented with tasks such as the one shown in
Figure 1. The pattern is animated and the figure number changes
accordingly. Students then build constructions for the patterns by
expressing what they ‘see’ as the structure of the pattern, making
explicit any of their rules, and finally using the relationships to
obtain the number of tiles needed in the pattern using the metaphor
of colouring the right number.
Figure 2 shows a snapshot of the microworld. The student Figure 1. Train-track activity
has just finished constructing the given pattern using a C-shaped
building block (shown in A). She associated the Figure Number of the task with the number
of ‘holes’ in the pattern and used a variable with this name. To create the pattern, the building
block is repeated as many times as the value of the variable ‘holes’ (B), in this case 3. In
4
The project is funded through the Technology Enhanced Learning Phase (ESRC/EPSRC-TLRP- RES-139-250381).
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 103
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
every repetition the block is placed two squares across (C) and zero places down (D). To
complete the pattern the student needed a ‘line’ of three tiles at the end.
Finally the student has
to colour the pattern by
allocating to it the exact
number of coloured tiles. In
the case of the sub-pattern
made of C-shapes, this
required 15 tiles, i.e. 5 * holes
(E).
As students build their
constructions in ‘My World’,
a ‘General World’ can be seen
alongside it (right window).
This world exactly mirrors
‘My World’ until the student
has unlocked a number (i.e.
created a variable), at which
point eXpresser randomly
Figure 2. Constructing and describing with rules a pattern in the
changes its value in the
eXpresser. Letters highlight the main features: A) base shape to be
‘General World’. The idea of
repeated to make a component of the pattern, B) number of repetitions
‘locked’
and
‘unlocked’
(in this case the value of the variable `holes’) C) Number of squares to
move to the right after each repetition (in this case 2) D) Number of
numbers was introduced to
squares to move down after each repetition E) Units of colours required
allow students to specify
to paint the pattern component F) Any variable used in the construction
whether a number should stay
takes a random value in the ‘General World’ (G) A rule for the total
the same (locked, i.e.
number of units of colours required to paint the whole pattern in a
constants) or could change
general way. (H) Patterns can be animated when the system changes the
value of the variables and are coloured if the rule is correct.
(unlocked, i.e. variables). In
‘My World’ students can edit the unlocked numbers whereas in the ‘General World’,
eXpresser chooses random values for the ‘unlocked’ numbers. For example, in the snapshot
the value of `holes’ is 5 resulting in a different instance of the pattern (F). The pattern in the
‘General World’ is coloured only when students express correct general rules for the pattern.
Students cannot interact with the ‘General World’ except by clicking the play button (H) to
animate their patterns and test its generality.
Methodology
Throughout the development of eXpresser, we have followed an iterative design process,
interleaving software development phases with pilot studies with students of our target age
(11-14 years old). We have also integrated feedback from teachers and teacher educators as
well as the students who participated in the studies. The study presented here comprised three
activities: the first introduced the functionalities of the microworld through a series of eight
video-tutorials, the second where students used eXpresser to build a simple pattern, and the
third, called Train-track (as presented in Figure 1) in which students were asked to find a rule
that gave the number of green tiles for any figure number. Sixteen students who participated
in this study were asked questions throughout their interactions designed to reveal their
comprehension of the system and the particular design features under study.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 104
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Students’ Interactions with eXpresser
This section focuses on three key design features and student reactions to them.
Animated task presentation to provide a rationale for generality
All patterns were presented to students animated changing at fixed time intervals
showing a different instance of the pattern each time. This made it hard for students to count
the number of green tiles (e.g. Figure 1), while allowing them to see the variant and invariant
parts of the pattern. This presentation provided a rationale for deriving a rule that output the
number of green tiles for any instance of the pattern, i.e. a ‘general’ rule giving concrete
instantiation to the meaning of ‘any’. In addition, they were given the chance to see their
constructions animated by clicking the play button in the General World (Figure 2H) that
allowed them to validate their model’s generality.
One student when asked to describe the way she saw the task in Figure 1 replied:
Ann: ‘These are flashing green squares and it [referring to the figure number] changes
number’.
At the end of the session, she was asked:
Why do you think we presented the task like this
(animated)?
Ann: because you can’t make it move on paper. […] it’s not just one number and it
doesn’t stay the same.
Besides the dynamic potential the technology offers in comparison to conventional
paper presentation, Ann seemed to have gained a more ‘general’ perspective after interacting
with eXpresser. Ann expressed the notion of a variable in her own words as ‘it doesn’t stay
the same’. We could also see how she switched to paying attention to the value of the number
of tiles rather than the changing shape – a rather clear instance of differentiating (or at least
seeing the importance) between the object and its value- and in so doing sees the connection
between the changing pattern and the figure number.
Working on a specific case ‘with an eye’ on the general
The most crucial, yet difficult, step for students is to distinguish between what stays the same
and what changes between different instances of a pattern. Students tend to work with a
particular example and struggle to find the solution for an infinite number of invisible and
unspecified cases. We therefore designed the system with two separate, yet linked, windows
as described earlier. If students make two or more sub-patterns (e.g. see figure 3) to build
their pattern, they have to express the necessary relationship(s) between the unlocked
numbers if their construction is not to be messed-up in the General World (random numbers
are chosen for each unlocked number). This happened with Nancy who made the train-track
pattern using two sub-patterns of vertical lines and vertical with a gap but any relationship
between them. She was surprised when she noticed the messed-up pattern in the General
World, but this helped her realise that she had to express that the number of repetitions of the
building block A1 as one more than the number of the building block B1.
Researcher:
what happened there [pointing at the General World]?
Nancy:
It is not joined up ... because you only have 1,2,3 of them and it’s got to be ,
one more extra. [...] If there is 1,2,3,4,5 of these, that means that you always have to add one
more. You might have 5 then you need to add an extra line, so that’s 6.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 105
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
By using the word ‘always’, Nancy revealed her ability to generalise in natural
language and eXpresser allowed her to express this relationship by using locked and unlocked
numbers. She was then able to construct the general rule for her model (as mentioned later in
this section).
Figure 3. Nancy’s construction from two sub-patterns. The first (A1) is made
by repeating a building block of 3 tiles five times. Since the number of
repetitions was unlocked, it was changed randomly to 3 in the General World
(A2). The second sub-pattern (B1) is made by repeating a block of 2 tiles
four times. In the General World (B2) the unlocked number was changed to
6. This helped her realise that she had to express the number of repetitions of
the building block A1 in terms of the number of repetitions of B1.
A powerful description of the General World and its overall purpose was given by
Kathy:
Kathy: in My World is like a plan and in the General World is what comes to life and actually
moves. It [General World] makes it live and animates it. […] The only difference is that on My
World you put the figure number and it doesn’t change, whereas in the General World, the figure
number changes.
Kathy succeeded in seeing the rationale of the General World as a window that
provided different instances of her pattern and therefore a window on to her own
generalisations. She also had a rather powerful metaphor of a ‘plan’ for the construction of a
rule via the specific case, one which could be ‘implemented’ (our term) in the general case.
Mutually supportive model construction and rule construction
In eXpresser, colouring a pattern requires an expression (a rule) that allocates the correct
number of tiles in all cases. Based on the constructed pattern, students need to find the exact
number of coloured tiles for their pattern. For local patterns (figure 2E) this can be expressed
as a multiplication of the number of repetitions and the number of tiles in the building block.
We were convinced that this design decision would give a direct experience with the number
of tiles needed to construct the building block and encourage students to look at the structure
of their pattern. It would further support them to construct a rule that calculates the number of
coloured tiles using the number of tiles of the building block as a coefficient. For their
complete construction and colouring of the pattern, they need to provide a combination of
local rules.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 106
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Nancy, for example, when constructing the model in Figure 3, derived two rules for
her two sub-patterns. Once she had coloured the individual patterns, it only required a small
step to understand that the general rule can be constructed by adding the two together (see
Figure 4). It seemed, therefore, that for most
students the process of model construction
supported the construction of a final rule for the
pattern. In addition, when asked to show how they
worked out their rule, they usually employ
elements of their model to describe it (e.g. see figures 5 and 6).
Figure 4. Nancy’s rule
Figure 5. Henry’s rule for train-track task
Figure 6. Henry’s demonstration of his
rule based on his model.
Conclusion
Our attempt to build a system that gives students the opportunity to work with the particular
and general at the same time is still ongoing. The approach we chose through the three key
design ideas is not the only existing one, but has advantages compared to the conventional
methods with paper-and-pencil. We have therefore some provisional evidence interaction
with eXpresser as a model of generalisation engages students, provokes them to think about
generalisable structures and helps them to make the transition from numbers to variables in a
way that is meaningful. In the interviews, all students used their rules to give the right number
of green tiles for different figure numbers using structural reasoning and not pattern spotting.
Kathy, for example, having found a similar rule to Henry, said:
Kathy:
my rule was 5 green times the figure number add 3.
Researcher:
so for figure number 6, what would it be? How many would there be?
Kathy:
it would be… 5 times 6…30…it would be 33.
Researcher:
for 12?
Kathy:
it would be 5 times 12…60…add 3 ….63.
Researcher:
for 600?
Kathy:
5 times 6…300…and then 3…
It seems that interaction with eXpresser discouraged students from calculating and
spotting patterns at the expense of expressing structure (see, for example, with reference to
this type of task, Noss et al. 1997, Healy and Hoyles 2000, Küchemann and Hoyles 2009).
Paper and pencil approaches tend to lead to the referents of the relevant variables becoming
obscured, thus limiting students’ propensity to conceptualise relationships between variables,
to justify and use them in a meaningful way. With eXpresser, students have to construct
models employing the structures they see and find rules with generality in mind.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 107
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
REFERENCES
Cuoco, A., E. P. Goldenberg and J. Mark. 1997. Habits of Mind: an organizing principle for
mathematics curriculum. Journal of Mathematical Behavior 15(4): 375-402.
Geraniou, E., M. Mavrikis, K. Kahn, C. Hoyles, and R. Noss. 2009. Developing a Microworld
to Support Mathematical Generalisation. In PME 33: International Group for the
Psychology of Mathematics Education, 49-56. Thessaloniki.
Healy, L., and C. Hoyles. 2000. A Study of Proof Conceptions in Algebra. Journal for
Research in Mathematics Education 31(4): 396-428.
Küchemann, D. and C. Hoyles. 2009. From computational to structural reasoning: tracking
changes over time In Teaching and Learning Proof Across the Grades K-16
Perspective, ed. D.A. Stylianou, M.L. Blanton and E.J. Knuth. Lawrence Erlbaum
Associates.
Noss, R., L. Healy, and C. Hoyles. 1997. The Construction of Mathematical Meanings:
Connecting the Visual with the Symbolic. Educational Studies in Mathematics 33(2):
203-233.
Noss, R., C. Hoyles, M. Mavrikis, E. Geraniou, S. Santos and D. Pearce. 2009. Broadening
the sense of `dynamic': a microworld to support students' mathematical generalisation.
Special Issue of The International Journal on Mathematics Education (ZDM):
Transforming Mathematics Education through the Use of Dynamic Mathematics
Technologies 41(5): 493-503.
Warren, E. and T. Cooper. 2008. The effect of different representations on year 3 to 5
students’ ability to generalise. ZDM Mathematics Education 40: 23-37.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 108
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Do students’ beliefs relating to the teaching of primary mathematics match their
practices in school?
Caroline Rickard
Initial Teacher Education, University of Chichester, UK
This paper reports research findings from the second year of a small-scale, longitudinal case
study undertaken with undergraduate students at the University of Chichester. This four year
project seeks to explore the impact of the beliefs of students in Initial Teacher Education
(ITE) upon their teaching of primary mathematics, noting placement constraints. Data
collection involved observations and interviews, and took place in the latter half of a six week
block of school experience.
Keywords: Mathematics; Beliefs; Primary; Teacher Education; School Experience
Aims and introduction
Working in ITE since 2001, I seek to engage students in mathematics in a way which
reflects my beliefs about how mathematics should be taught to primary age children, noting
that as teachers we “convey messages about the nature of mathematics by the way we teach
it” (Nickson, 2004: 43). Interviewing ten first year undergraduates after a six week
introductory mathematics module established, unsurprisingly, that their views about how
primary mathematics should be taught broadly reflected my own (Rickard, 2008). A clear
question remained however:
• Would clearly articulated beliefs relating to the teaching of primary mathematics
match the students’ practices on school experience?
The purpose of this next phase was thus to observe some of the students’ mathematics
teaching in order to investigate the transferability of their beliefs into the school context; in
particular whether their beliefs were resilient enough to withstand the demands made upon
student teachers, and whether a particular school placement might constrain or shape the
student teacher’s views. Various sources (see for example Brown and Borko, 1992) caution
that classroom pressures may mitigate against application of perceived ‘best practice’
however “early and continued reflection about mathematics beliefs and practices, beginning
in teacher preparation, may be the key to improving the quality of mathematics instruction
and minimizing inconsistency between beliefs and practice” (Raymond, 1997: 574). Williams
supports this view suggesting that involving students in “articulating and discussing beliefs
and practices associated with mathematics” (2001: 447) is likely to result in more effective
practices.
Methodology
Data was gathered through observation and interview in the spring of 2009 with a
triangulation of methods chosen because, as Elton-Chalcraft et al (2008: 79) point out,
“observations can be useful for overcoming the difference that can exist between what people
say and what they do in practice”. Holding interviews directly after the observations also gave
an opportunity to discuss any additional beliefs, noting that not all beliefs were necessarily
going to be explicitly incorporated in any single observation. Three opening questions were
identified in advance:
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 109
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
1. What would you have done differently in that lesson given the chance? (Incorporating
self-evaluation and leading to exploration of whether anything prevented students
from doing what they wanted, asking them for example to reflect on pedagogical
beliefs they felt they had to put ‘on hold’ during the placement.)
2. Since the initial interview asking you about your beliefs as to how primary
mathematics should be taught, are you conscious of any changes to those beliefs?
(Including describing the new beliefs and what students felt had caused them to alter.)
3. Are there any other examples you want to give of mathematics lessons you’ve planned
and taught which you feel would help to exemplify your beliefs? (An opportunity to
explore beliefs and practices over the slightly longer term.)
Four of the original research group were observed; students who knew me and were
willing to participate in the interviews in their first year and then available in year 2, and in
this respect a climate of trust (Hopkins, 2008) was easily established. One participant, Jenny,
was taking a secondary rather than a primary placement (names have been changed to
preserve the anonymity of the students). Observations were undertaken on a non-participant
basis.
Figure 1 shows the responses relating to the most commonly expressed beliefs in the
original interviews in the first year of the research (n=10) and these features were developed
into a simple observation schedule for use in the second phase.
ETHOS... there should be a positive atmosphere in the mathematics classroom; the 70%
word ‘fun’ was mentioned a lot, and a desire to avoid mathematics being
perceived as boring.
DOING... an emphasis on practical mathematics lessons using a variety of
70%
resources, with the phrase ‘hands-on’ used several times.
CONTEXT... referred to in various ways including mathematics linked to themes, 60%
cross-curricular opportunities and real world mathematics.
DISCUSSION... opportunities for children to talk about their mathematics, often
50%
linked to implications for working in groups.
Figure 1.
Interval sampling helped to focus my attention on how frequently (if at all) I was
seeing the sorts of practices that reflected my students’ original beliefs. Having predetermined categories for observation may have helped to address issues of reliability to some
extent, but personal expectations may have been a potential source of bias: “Put simply, you
are more likely to see the things you expect to see, and hear the things you want to hear”
(O’Leary, 2004: 176). The schedule was accompanied by a page for notes about the context
of the observation, such as the number and age of the children, organisation of the classroom,
and of course detail of the mathematical focus. I also recorded information about the setting
and/ or ability grouping of the children.
As a result of piloting the observation schedule, systematic recording in the ‘ethos’
column was abandoned, value judgements being too subjective. It was however pleasing to
note that students praised children regularly, congratulating them on their effort and
achievement.
Findings
Four observations took place in May of 2009 in different schools in West Sussex and
the background information is summarised in figure 2.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 110
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Janey
Y1
(n=27)
Amelia
Y1
(n=27)
Y3
(n=31)
Jill
Jenny
Y10
(n=14)
Mixed ability
class organised
into three ability
groups.
Lower ability set.
Counting on, including crossing a tens boundary.
Lower ability set
organised into
ability groups.
Average ability
set n an all boys
school.
Multiplication relating to multiples of certain creepy
crawlies (e.g. spiders) with certain numbers of legs.
Collecting data from peers including tallying.
Revision of geometrical constructions.
Figure 2.
In the interviews which took place after the observations, three of the four students
reported that they would have liked to have done things differently given more freedom.
Interestingly the fourth student, Amelia, would herself have made minor changes but none of
these related to a lack of freedom, or in fact to the improvements her mentor felt were
required. Whereas the others spoke a lot about the children’s learning in reflection on their
lessons, Amelia’s focus was associated more closely with things she would do differently, for
example not talking over the children (waiting for quiet) and making greater use of a puppet.
Several themes emerged from the analysis of the data; these will be reported under the
same broad headings as those taken from the original research, ignoring ‘Ethos’ however as
mentioned above.
Doing
This category encompassed the idea of practical activity and the use of resources, and
issues associated with equipment use came to light even before the visits took place! In emails
prior to my visit Jill wrote about being asked to teach without using resources, something
which the class teacher had suggested would help the Y3 children to better develop their
mental skills. Jill felt she was in an awkward position as she privately disagreed, feeling that a
number of the children were struggling and failing to make progress as a result. She was
however ‘allowed’ to use resources for her observation, which suggests that this was a
‘special’ lesson for my visit, something which Bryman refers to as ‘reactive effects’ (2008:
266). O’Leary (2004) warns that any person who knows they are being observed is likely to
alter their resulting behaviour and my students’ awareness of the research focus was
unavoidable having already asked them to articulate their beliefs about the teaching of
primary mathematics in year 1. As the schools were oblivious to my research focus however,
I hadn’t, perhaps naively, anticipated any alteration to school practices on account of my visit.
Whilst Jill felt that this lesson had been more successful as a result of the use of resources, she
went further to say that she would still have done things differently had the class teacher been
in agreement, for example providing photocopies of the minibeasts on the children’s tables so
that to start with they could count the numbers of legs if they needed to. Interestingly, she was
the only one of the four research participants who had not explicitly mentioned practical
activity through use of resources in her original interview, and yet it is clearly something she
now feels strongly about. Jill’s lesson also provided an explicit attempt to have children
‘doing’ as she had given out mini whiteboards which were used for 17% of the lesson.
Janey felt that I would have seen a greater level of interaction in other lessons, for
example when she made a broom handle into a physical/ visual numberline and through her
regular use of stories and songs. She reaffirmed her belief that mathematics should be taught
using a creative approach; she said that she would have liked to take the children outside or
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 111
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
into the hall for more of their mathematics lessons, but was concerned that she would find it
difficult to control her “bubbly” class. In this lesson examples of practical activity were fairly
minimal such as inviting children to hold items or stick things up on the board.
Typical to any secondary placement, Jenny was working across different classes and
with different members of staff, and she cited a particular member of staff, an ex-student of
Chichester, whose classroom and teaching style she felt were noticeably different to the other
practitioners, incorporating lots of engaging resources. In Jenny’s original interview she had
talked about the benefits of practical activity using resources, and whilst this lesson
necessarily met this condition (geometrical constructions using a ruler and pair of compasses)
she would like to have made greater use of resources in her other lessons. General
expectations of silent working were also mentioned by Jenny in relation to constraints on
doing more “interesting things”.
The opportunity for children to engage in practical activity was most extensive in the
lesson taught by Amelia as the children were circulating the room to gather data. In many
ways, however, the fact that all four lessons were on different themes makes it difficult to
compare them effectively.
Context
The value of context related to the teaching of mathematics was mentioned by Janey,
Jill and Amelia in the original interviews and all three lessons included some reference to a
context. The strongest link was in Amelia’s lesson but as mentioned above, comparison is
affected by differences in focus; this was a data handling lesson and data has to be about
something! Jill’s lesson made effective use of bugs (the focus of about 35% of the lesson) to
exemplify multiplicative structure (numbers of legs) and Janey told a short story about a
character who was scared of crossing the tens boundary, but links made to this focus were
only briefly maintained. She did however speak articulately about making greater use of links
to real life in other lessons.
Jill stated that she had become far more conscious of the importance of context as a
result of this placement; she reported that the focus had been predominantly upon pure
calculation, and that the children had then struggled with some complicated word problems
which she had been given to use with them.
Jenny didn’t mention context in her original interview and the observed lesson wasn’t
linked to any aspect of real-life or similar.
Discussion
Talk opportunities were mentioned as a feature of best practice by Jenny and Amelia
in the original interviews. Classroom talk in all four observations was predominantly
interactions between the teacher and the children, and only Amelia explicitly directed children
to talk to their peers about the mathematics, although Jenny did invite pupils to the board to
explain their work. Directed discussion occurred in 24% of Amelia’s lesson; 4 minutes being
allocated to the discussion of something specific with a partner, and 9 minutes of collecting
data from peers, an activity which necessitated discussion. In reflecting on her lesson Jill
mentioned that she thought her teaching was dominated by teacher talk, and this was indeed
borne out in practice. The emphasis in all four lessons was upon the teacher asking and the
children answering questions, and the giving of instructions for work, behaviour reminders
etc. There was also self-initiated discussion between pupils in all four lessons; many of Jill’s
Y3 children discussed the mathematics between themselves whilst they were working, and
this focus on mathematical talk continued until nearly the end of the lesson. With both Y1
classes however, the talk more often seemed to relate to off-task topics of discussion.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 112
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
In addition to the three themes explored above, two new themes also emerged: issues
associated with lesson structure; and opportunities for children to think for themselves.
Lesson structure
Volunteering reflections on lesson structure both Janey and Jill stated clearly that they
would have liked a better link between their mental and oral starters and the rest of their
lessons, but were constrained by school planning. Jill stated that hers was “nonsense in
relation to lesson” and she could suggest alternatives that would have better supported the
children in preparation for the lesson ahead.
Lesson structure can also be linked to notions of pace; in another email
communication from Jill, also prior to the visit, she expressed considerable frustration. She
was concerned that the pace of the teaching was leaving children behind, writing “I have a
feeling that the children are sometimes lost with the concepts and pace”. In her follow-up
interview Jill returned to the lesson structure theme saying that she felt logical steps were
sometimes omitted to the detriment of the children’s understanding.
Thinking
Valuing thinking was clearly a concern shared by both Jenny and Jill; the idea that
children should be given ample opportunities and encouragement to think for themselves in
mathematics lessons. When comparing her primary and secondary experiences Jenny stated
that she really wished “the boys to think for themselves” and to be more independent and Jill
made a similar point about her lowest ability group in the set. As these children never worked
alone she felt they were not getting “a chance to think for themselves and play around with
numbers”. Jenny stated that whilst last year the “kids were willing to try anything”, this year
the “children are out of the habit”.
Conclusion and future research
Returning to the original themes, first identified in 2008, it’s particularly interesting to
note that the notion of ‘fun’ in relation to mathematics lessons was not mentioned in the same
explicit way in this, the second year of the research. One possibility however is that one
doesn’t necessarily think to mention everything, however important, in the course of a single
interview. This may particularly be the case with fairly fundamental and deeply ingrained
beliefs. Jill failed to mention the use of resources in her initial interview and yet having
limited access to them in this year’s placement was quite an issue for her, suggesting that their
use may well have been an integral part of her original belief system. Linked to general use of
resources is a question over the effectiveness of their use: Jill for example chose to refer to
minibeasts, an excellent way of exemplifying multiplicative structures.
With three of the four participants articulating very clearly what they would have
wanted to do differently, given the chance, it is possible to conclude that they felt somewhat
constrained by their particular school placement. This is good news as it suggests that they
will be able to think for themselves and hopefully make some healthy decisions about the way
in which they teach primary mathematics in the future. One concern however is the effect of a
school experience which is at odds with your own belief system with both Jill and Jenny
speaking of doubts; Jill in relation to her own ability to teach, and Jenny the desire to pursue a
career in teaching.
Reflective awareness of one’s own practice allows us to move forwards; Jill was
clearly aware that she was dominating classroom talk and is therefore in a position to try to
address this. In fact, a whole new project on how we encourage student teachers to afford the
children lots of talk-time in the mathematics classroom would be a very suitable line of
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 113
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
enquiry! This could be linked to Jill and Jenny’s opinion that the children they were working
with this year were not being offered nearly enough opportunities to think for themselves.
The intention is now to continue to ‘follow’ this small group of students into their
third and final year (when they will be placed in different schools), and where possible into
their first year of teaching, noting the issues associated with comparing lessons involving
different branches of mathematics and classes of children of different ages. Clearly the most
useful focus in the long term relates to how we might support students and newly qualified
teachers in remaining consciously aware of their own beliefs, continuing to reflect on their
own and others’ approaches to the teaching of mathematics, and where necessary to strive for
change for the better.
Acknowledgements
Grateful thanks are owed to the small band of participants who have so willingly allowed me
to invade their classrooms.
References
Brown, C.A. and Borko, H. (1992) Becoming a Mathematics Teacher, in Grouws, D. A. (Editor) Handbook of
Research on Mathematics Teaching and Learning, Oxford: Macmillan
Publishing Company
Bryman, A. (2008) Social Research Methods (3rd Edition) Oxford: University Press
Elton-Chalcraft, S., Hansen, A. and Twiselton, S. (Editors) (2008) Doing Classroom Research: A Stepby-Step Guide for Student Teachers, Maidenhead: McGraw Hill Open University Press
Hopkins, D. (2008) A Teacher’s Guide to Classroom Research, Maidenhead: McGraw Hill Open
University Press
Nickson, M. (2004) Teaching and Learning Mathematics: A Guide to Recent Research and its
Applications (2nd Edition) London: Continuum
O’Leary, Z. (2004) The Essential Guide to Doing Research, London: Sage Publications
Raymond, A.M. (Nov 1997) Inconsistency between a Beginning Elementary School Teacher’s
Mathematics
Beliefs and Teaching Practice, Journal for Research in Mathematics Education,
Vol.28, No. 5, pp. 550576
Rickard, C. (2008 unpublished) Developing Beliefs about the Teaching of Primary Mathematics,
University of Chichester
Williams, H. (2001) Preparation of Primary and Secondary Mathematics Teachers, in Holton, D. (Editor) The
Teaching and Leaning of Mathematics at University Level: an ICMI Study,
London: Kluwer Academic Publishers
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 114
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
BSRLM Geometry working group: tasks that support the development of geometric
reasoning at KS3
Sue Forsythe; Keith Jones
University of Leicester; University of Southampton
Students at Key Stage 3 (ie aged 11-14) in English schools are expected to learn the
definitions of the properties of triangles, quadrilaterals and other polygons and to be able
to use these definitions to solve problems (including being able to explain and justify
their solutions). This paper focuses on a pair of Year 8 students (aged 12-13) working on
a task using dynamic geomtry software. In the research, the children investigated
triangles and quadrilaterals by dragging two lines within a shape (ie the diagonals of a
quadrilateral, or base and height of a triangle) and noting the position and orientation of
the lines which gave rise to specific shapes. Following this, the students were asked to
use what they had found in order to construct specific triangles and quadrilaterals when
starting with a blank screen. While the research is currently ongoing, and is using a
design research methodology, the evidence to date is that the task has the potential to
scaffold students’ thinking around the properties of 2D shapes and hence support the
development of geometric reasoning.
Keywords: dynamic geometry, task, design-based research
Introduction
The Framework for Secondary Mathematics in England (DCSF, 2008) indicates that Year 8
students (aged 12-13 years) are expected to know and understand the properties of triangles
and quadrilaterals, to be able to solve problems using these properties and to classify
quadrilaterals according to geometric properties. Yet simply expecting students to memorise
such shapes and their properties is likely to be insufficient support for students developing
their own meaningful concepts in geometry (Battista, 2002). This paper describes an attempt
to devise a task which would encourage students to develop a deeper understanding of how
shapes can be defined by considering the properties of two internal perpendicular lines. These
lines are the diagonals in the case of certain quadrilaterals and the height and base in the case
of triangles. The students explored these shapes in a dynamic geometry environment,
specifically The Geometers Sketchpad (GSP) version 4 (Jackiw, 2001).
Using tasks to support learning in geometry
Open problems in geometry have been shown to encourage children to develop meaningful
concepts (eg: Mogetta et al 1999 a). Open problems usually consist of a short statement
where students are asked to explore connections between elements of a figure. Open problems
do not lend themselves to solution solely through the use of learned procedures; students have
to decide how to explore the problem and there may be a number of results that could be
reasonable solutions to the problem. The benefit to the students of working through open
problems is that the outcomes are meaningful to them and the opportunity to explain their
results may be a pre-cursor to being able to prove in geometry (eg: Jones, 2000).
Working on a problem in a computer environment also has benefits. Papert (1993)
argued that computers encourage concrete thinking and ways of solving problems that involve
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 115
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
finding approximate solutions and then tweaking them until the optimal solution is found.
This approach seems to suit children.
It is helpful to provide problems which the students find engaging. Ainley, Pratt and
Hansen (2006) consider that tasks which involve programming computers can provide
purpose and utility, which enriches students’ learning of mathematical concepts; de Villiers
(1994) said something similar when he described ‘functional understanding’ ie the
understanding of the usefulness or value of doing a specific bit of mathematics, as being just
as important as relational and logical understanding which was described by Skemp (1973).
The task, described in this paper, was designed to take all of the above into account. It
was intended that the task would stimulate the students into thinking more deeply about the
methods of constructing specific triangles and quadrilaterals and that they would be able to
explain why their methods worked. As Mogetta et al (1999 b) say, switching between what
the student notices on the screen (empirical evidence) and the geometrical theory may
stimulate the students’ capability to prove as they have to explain the reasons behind what
they are observing. Problems can only really be solved when what is observed on the screen is
explained geometrically.
Dragging and measuring in a dynamic geometry environment
With DGS software (such as GSP, Cabri, GeoGebra), an important function is the dragging
mode which allows the user to drag geometric objects on the computer screen. The computer
interface allows direct manipulation of the drawing on the screen via the drag mode whilst, at
the same time, preserving all the geometric properties used to construct the figure (Laborde,
1993).
Another function in DGS is the measure menu. Students can measure lines and angles
on the diagram and, as the diagram is dragged, the measurements given on the screen are
updated continuously. Olivero and Robutti (2007) describe students using the drag mode to
adjust a sketch on the screen until the measurements indicate that they have obtained a
particular figure from a generic one and they called this an example of ‘guided measuring’.
When students check their constructions through measuring and dragging this is called
‘validation measuring’.
Design-based research
The methodology used in this study was that of design-based research (Brown, 1992; Designbased research collective, 2003). In a design-based research experiment, the researchers aim
to study how learners learn by designing tasks and learning situations through which they
hope to see improved learning outcomes. Design-based research uses the design experiment
to study learning and develop theories about learning in a specific context but which can be
extrapolated to theorise about learning in a broader context (Barab and Squires, 2004).
In a learning situation, even a simple one where there are two students and one
instructor / researcher as in this case, cognition is not separate from the thinker, the task or the
environment - these all need to be treated as one complex system (Design-based research
collective, 2003). The learning environment is a complex system of inter-relating aspects
where one aspect cannot be changed without it affecting all other aspects. Design-based
research methodology accepts this as the case and works with it rather than against it (Brown,
1992). Design-based research experiments thus need to take account of all the aspects of the
learning situation and how these all work together. Testing, scrutinising and revising a design
results in an iterative process over several cycles of the research (Cobb et al, 2003).
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 116
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Clearly there are some challenges in design-based research which need to be
considered. The role of the researcher(s) in a design-based experiment means that they have a
large influence in how the experiment proceeds (Barab and Squire, 2004). Design-based
research experiments can also generate a large amount of data and the researcher must choose
which data to focus on in the analysis. The researcher must be aware of the need to be
objective and not select data that backs up their own preconceived ideas (Brown, 1992). This
shows how important it is for the researcher to try to be objective and for all their activities to
be transparent.
The experiment
Two boys and two girls worked in pairs in the summer term of Year 8 (students aged 13) for
two 50 minute sessions. Data was collected from these sessions in the form of an audio tape
and a recording of the computer screen using image capture software. The students were
given a task which was loosely related to a ‘real world’ situation in that they were asked to
imagine a toy kite made of two ‘sticks’ which provide the scaffolding for the kite. The fabric
which makes up the kite is imagined to be elastic so that the ‘sticks’ can be moved around to
create different shaped ‘kites’. The students were asked to investigate the different shapes that
can be made in this way and to describe the orientation of the ‘sticks’ inside each shape. In a
later session the students were asked to construct ‘drag proof’ shapes starting with a blank
screen.
The findings
The first session allowed the researcher to assess the students’ prior knowledge of shapes and
their properties. The students were presented with a GSP file which contained a 6 cm
horizontal bar and 8 cm vertical bar. The students completed the shape by joining the ends of
the bars (see figure 1) and then constructed the interior of the shape. Even though the bars
could be dragged anywhere on the screen, the students preferred shapes that had vertical
symmetry. By dragging the bars they were able to make kites, a rhombus, an isosceles
triangle, right angled triangles in different orientations, and concave kites. The students were
then asked to measure objects in the shape in order to be absolutely sure that they had made
the shape. They did this using the measure facility of GSP, which shows the measurements as
text on the screen.
Next the students were provided with a GSP file with perpendicular bars of adjustable
lengths. With two equal perpendicular bars the students were able to make a square as well as
various kites, an isosceles triangle and a right angled isosceles triangle. Each time the students
made a shape they were asked to describe the orientation of the bars inside the shape. The
students realised that the measures changed as they dragged the bars inside the shapes. The
students moved the bars inside the shape whilst checking the measurement given on the
screen until the measurements were as close as they could get them. This is an example of
‘guided measuring’ as described by Olivero and Robutti (2007).
The students were happy to accept measurements which were close, but not exact, in
order to feel confident that they had generated a particular shape. This mirrors the use of
measurement in the pencil and paper environment. They also noticed, when prompted, the
orientation of the bars inside the shape which would be useful in order to carry out the task in
the second session when they would be asked to construct shapes starting with a blank screen.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 117
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Figure 1: the Kite task in GSP
For the second session the students were provided with ‘instruction cards’ which
explained how to do various operations from the Construct and Transform menus (such as
mid-point of a line, perpendicular to a line). These cards were placed on the desk so that the
students could choose which ones they needed. The instruction cards served to give hints to
the students as to what might be a useful construction.
For the first part of the session, the students were asked to make an isosceles triangle.
It was clear from observing their attempts that they had remembered the orientation of the
bars from the previous session. At first they drew lines to represent the bars ‘by eye’ and then
completed the outside edges of the shape. When they discovered that the result could be
dragged out of shape they realised they needed to make shapes that were ‘drag proof.’ It was
at this point that they looked at the help cards on the desk.
The best constructions that the students made were when they chose what they would
like to do to their diagram and found ways to get the GSP to achieve that. For example when
making the square they first drew a line, then constructed a mid-point and then constructed a
perpendicular line through this mid-point. They were perplexed when the perpendicular was
an infinite line and they did not know where to place the opposite corners of the square along
the infinite line. The following conversation ensued with the girls.
Researcher: What would you like to do? If you don’t know how to do it, what would you
like to do? And I might be able to tell you how to do it.
Girl 1: Find a point here which is, like, the same distance and will make a right angle.
Researcher: Well we know if it’s on that line it’ll be a right angle. So what do you think
you could do?
Girl 2: Is there some way you could, almost, spin it round?
Researcher: If you spin it round, what’s that called?
Girl 2: Rotate.
The students then rotated the first line 90 degrees onto the infinite perpendicular line
to find out where the corners of the square would go. As they did in session 1, the students
used the measure facility to check that they had made the shapes correctly. An example of a
square with measurements is shown in figure 2.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 118
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Figure 2: constructing a square in GSP
Discussion
The levels of geometrical reasoning devised by van Hiele (1986) are often used to
characterise the level of students’ development in geometrical reasoning. If the task used in
this research is useful for developing the students’ reasoning then we should see some
progress within the levels. The students showed, in the first session, that they were able to
understand shapes as being collections of properties, which is evidence of reasoning at van
Hiele level 2 (ie shapes as being collections of properties). In the second session the students
were able to construct specific shapes starting with a blank screen, namely isosceles triangle,
kite, square, equilateral triangle. This indicates that they had learnt something about the
properties of the diagonals of the quadrilaterals, or the base and height of triangles, in order to
do this. Being able to solve problems using these properties, such as constructing the shapes
in a dynamic geometry environment, would indicate progression towards van Hiele level 3. In
addition, the provision of ‘instruction cards’ served to support the development of the
students’ geometrical vocabulary. The students started to talk about what they were doing
using the language on the cards, which is also the language used by the software.
Conclusion
What has been described in this paper is the first iteration of the design experiment process.
The evidence to date is that the task has the potential to scaffold students’ thinking around the
properties of 2D shapes but that the task needs to be developed further in order to consolidate
and build on what the students have learned. For example, a third session would be useful,
where the students might be asked to build a macro which would generate a shape such as a
square. This would also encourage the students to find efficient ways to generate shapes leading them, perhaps, to a realisation of the minimum properties required to render a specific
shape such as a square. This activity could also suggest to students that they explain why their
macro works - which may lead them onto more formal proof and would provide more solid
evidence of development towards van Hiele level 3.
The use of geometrical language to support learning is an issue which has received
only modest attention in this study so far. Future iterations of the work with students could
usefully consider the potential that the task and the dynamic geometry software have to
develop students’ use of geometrical language.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 119
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Acknowledgements
The authors would like to thank the colleagues who attended the BSRLM conference session
on 14 November 2009 and especially for their comments and suggestions.
References
Ainley, J., Pratt, D., and Hansen, A. 2006. Connecting engagement and focus in pedagogic
task design. British Educational Research Journal, 32(1), 23-38
Barab, S., and Squire, K. 2004. Design-based research: Putting a stake in the ground. Journal
of the Learning Sciences, 13(1), 1-14
Battista, M. T. 2002. Learning geometry in a dynamic computer environment. Teaching
Children mathematics, 88(6), 333-339
Brown, A. L. 1992. Design experiments: Theoretical and methodological challenges in
creating complex interventions in classroom settings. Journal of the Learning
sciences, 2(2), 141-178
Cobb, P., Confrey, J., diSessa, A., Lehrer. R. and Schauble, L. 2003. Design experiments in
educational research. Educational Researcher, 32 (1), 9-13.
DCSF 2008. Framework for Teaching Mathematics.
http://nationalstrategies.standards.dcsf.gov.uk (accessed Dec 2009)
Design-based Research Collective 2003. Design-based research: An emerging paradigm for
educational enquiry. Educational Researcher, 32 (1), 5-8
Jackiw, N. 2001. The Geometers’ Sketchpad (version 4). Key Curriculum Press.
Jones, K. 2000. Providing a foundation for deductive reasoning. Educational Studies in
Mathematics, 44(1-3), 55–85.
Laborde, C. 1993. Do the pupils learn and what do they learn in a computer based
environment? The case of Cabri-géomètre. Proceedings of ‘Technology in
Mathematics Teaching: A bridge between teaching and learning’. Birmingham :
University of Birmingham.
Mogetta, C., Olivero, F. and Jones, K. 1999. Providing the motivation to prove in a dynamic
geometry environment. Proceedings of the British Society for Research into Learning
Mathematics, 19(2), 91-96.
Mogetta, C., Olivero, F. and Jones, K. 1999. Designing dynamic geometry tasks that support
the proving process. Proceedings of the British Society for Research into Learning
Mathematics, 19(3), 97-102.
Olivero, F. and Robutti, O. 2007. Measuring in dynamic geometry environments as a tool for
conjecturing and proving. International journal of Computers for Mathematical
Learning, 12, 135-156
Papert, S. 1993. The Children's Machine: Rethinking School in the Age of the Computer.
Harvester Wheatsheaf.
Skemp, R. 1973. Relational versus instrumental understanding. Mathematics Teaching, 77,
20-26
Van Hiele, P. M. 1986. Structure and Insight: A theory of mathematics education. Orlando,
Fla: Academic Press
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 120
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
Working group on trigonometry: meeting 4
Notes by Anne Watson
Department of Education, University of Oxford
These notes record the discussion at the fourth meeting of this working group. The focus
was on the history of trigonometry, and discussing three different approaches to teaching
it which have appeared in recent readings.
Keywords: congruence, similarity, astronomy, triangle, trigonometry
Historical development
Leo Rogers had supplied us with a brief overview of the history of trigonometry.
Unfortunately he was unable to join us, being occupied with a Working Group on History of
Mathematics. There appear to be two strands, the astronomical strand which used circles and
arcs as the basic tool to track the position and movement of stars and planets, and the
surveying strand which used ratios of sides of right-angled triangles. The sundial uses both.
In an earlier meeting we had circulated the suggested teaching approach of Thompson,
Carlson and Silverman (2007) for pre-service teachers. They use arcs of circles in order to
present sine etc. not as ratios which have to be imagined, but as measures of arcs which can
be seen. We referred to this briefly, and the fact that for small angles one can approximate a
circle segment as a triangle. Historically there seems to be a progression from approximating
a slim segment as one triangle to bisecting the relevant chord to approximate it as a rightangled triangle on the half-chord.
We thought that the history of trigonometry includes the history of our exploration of
their properties.
Two future actions arise from this discussion:
• Learn more about the triangle roots of trig, as illustrated in the Nine Chapters
to find heights of tall things (Liu Hui); find out what the engineers were doing,
as well as the astronomers
• Return to reading the Thompson paper to look for analogies.
Comparing two approaches
We then compared two recent articles on teaching trigonometry (Kemp, 2009; Steer, da Silva
and Easton, 2009). Both had appeared in the same issue of Mathematics Teaching without
editorial comment. We thought it might be helpful to compare them in the light of some of the
analysis we have been doing in the working group.
Similarities
•
•
•
•
both took an approach which focused on properties of triangles
both delayed the introduction of technical terms until the relations had been
established empirically
both depended on learners noticing what stays the same when certain features
of triangles are varied
both aimed at complex understanding rather than technical procedures
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 121
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
•
•
•
both emphasised the significance of similarity
both mentioned ‘SOHCAHTOA’, but we were not sure if this mnemonic was
being used to communicate to the reader that formalisation and application of
the ideas was the endgoal, or whether it had been introduced to learners in that
form – the mnemonic itself being the endgoal
both appeared to have resolution of triangles as the goal, rather than a
functional understanding.
Differences
One approach (Steer et al., 2009) depended on dynamic geometry, and was reported a way
which emphasised the technology, while the other (Kemp, 2009) used lowtech materials. The
paper by Steer et al. described the implementation of ideas developed by Jeremy Burke
(2006). It was the third of a series but summarised the overall approach. Burke’s ideas had
been circulated to this group earlier, and this article omitted relating trigonometric ideas to the
unit circle, as he had suggested. We wondered why this might be so, and concluded that the
realities of pressures on teaching may have led them to truncate the proposed sequence. Some
readers felt that it dwelled in detail on technology use, rather than relegating software to the
position of a tool.
Another difference was that, while both used similarity as the central idea, this was
approached as a ‘constant binary relation between sides of these triangles’ rather than through
‘preservation of proportion by scaling’ being a central idea that makes trigonometry possible.
The scaling idea was more prevalent in Steer et al’s approach than Kemp’s. In the former,
similarity arises after consideration of congruence, so the emphasis is on types of sameness,
and scaling (multiplication) when triangles share the (a,a,a) characteristic is central. In the
latter, similarity arises as a relation among sides of triangles that ‘look the same’ and
multiplication is a choice from four binary operations to find one that is constant.
Neither approach explains why only right-angled triangles are chosen for this
exploration, since all sets of similar triangles have common ratios. Use of a unit circle could
have made this clear.
Embodiment of trigonometric ideas
Finally we read a passage from Lakoff and Nunez (2000) about how trigonometric ideas can
arise from blended fundamental conceptual metaphors which relate to how humans are in the
world. They claim that there has to be a metaphor to enable us to relate angles to numbers, for
which we have some fundamental understandings. They draw on the idea of the unit circle as
the appropriate metaphor, and describe this as a blend of circles in the Euclidean plane with
the two-dimensional numerical metaphor of the Cartesian plane, and the angle in the
Euclidean plane having two legs that delineate the angle. The final blend gives the familiar
diagram of the right angled triangle generated by one rotating leg of the angle and the vertical
dropped from it. They show that trig ratios and the functions are both represented and
generated by this diagram.
What they offer is logical, in that these ideas are related mathematically in the way
they describe, but incompatible with the historical development we had thought about earlier.
We want to think more about how ‘surveying the earth’ and ‘measuring the heavens’ might
have been perceived over centuries, and what role the unit circle might have played.
Trigonometrical ideas were used long before Descartes offered the ‘metaphorical blend’ for
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 122
Joubert, M. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 29(3) November 2009
number and space. The chronological development does not necessarily match the logical or
metaphoric perspective, and does not support the order in which they relate the ‘metaphors’.
We observed that what is ‘natural’ is culturally loaded, since for some cultures
distance and direction combined is a fundamental way of seeing the world. Further, some
relevant static relations, such as our understanding of near and far objects, are also naturally
embodied in our ways of being in the world and also contribute to trigonometric
understanding. In summary, we could not understand why it is helpful to see these ideas as
metaphors, nor could we agree that these were the appropriate metaphors for trigonometric
understanding.
Future plans
We shall next meet at the BSRLM meeting in Summer 2010. This is an open group and all
are welcome to join. If you would like copies of earlier readings please contact
anne.watson@education.ox.ac.uk.
References
Burke, J. 2006. Trigonometry: an introduction using dynamic geometry. London, King’s
College.
Kemp, A. 2009. Trigonometry from first principles. Mathematics Teaching. 215: 40-41.
Lakoff, G. and Nunez, R. 2000. Where mathematics comes from: how the embodied mind
brings mathematics into being. New York, Basic Books.
Steer, J., de Vila, M. and Eaton, J. 2009. Trigonometry with year 8: Part 3. Mathematics
Teaching. 215: 6-8.
Thompson, P., Carlson, M. and Silverman, J. 2007. The design of tasks in support of teachers’
development of coherent mathematical meanings. Journal of Mathematics Teacher
Education. 10: 415-432.
From Informal Proceedings 29-3 (BSRLM) available at bsrlm.org.uk © the author - 123