TOPIC 6 ASSESSING GRAMMAR

6.0 SYNOPSIS
Topic 6 provides input on how to assess grammar. We are going to look at how to test
the students knowledge of grammar. We will also look at a number of types of test
items.
6.1 LEARNING OUTCOMES
By the end of this session, you will be able to:
. demonstrate an understanding of types of tests
!. e"plain the rules for writing tests
#. design items to test students knowledge of grammar
6.2 FRAMEWORK OF TOPICS
63
Types of
Tests
Rules of
writing tests
Developing
test items
Multiple choice Fill-ins Sentence
completion
activities
Transformations
Sentence
writing
Sentence
reordering
Parallel writing
CONTENT
SESSION SIX (3 Hours)
3.0 $ou have taught the grammar. $ou have practised it. $ou have corrected it. But
how do you know if the process has worked% &ow do you test it%
3.1 TYPES OF TEST
We can test the students ability to speak or write. We can test students reading
or listening comprehension skills. 'any tests include all these elements, especially
public e"ams like ()' and '*+T.
)ublic e"ams test how good a students overall command of +nglish is. ,n this
chapter, however, we will look at tests which are given in schools and classes to find out
how well students have done. These are often called achievement tests and are given
after a semester or years work. The aim of such tests is to see if students have learnt
and ac-uired the language they have been studying or have been e"posed to.
3.1.1 WRITING ACHIEVEMENT TESTS
Writing a test is an important .ob that demands skill and patience. /ood tests
show both teacher and students how well they are all doing. They do not fail students
unnecessarily and they give everyone a chance to show how much they have learnt.
Tests can often go wrong, not .ust because of the students lack of knowledge,
but also because of problems in the writing of the tests themselves. When writing tests
teachers should bear in mind the following 0rules.
a. Don! !"#! $%a! &o' %a("n! !a')%!.
The purpose of an achievement test is to find out how well students have achieved what
they have been studying. ,n such a test, then, it is not fair to test things that they havent
been e"posed to.
6
*. Don! !"#! )"n"+a, -no$,".)"
Test writers should remember that they are testing the students knowledge of +nglish,
not their knowledge of the world. 1or e"ample, a test item like this would not be
acceptable.
)icasso was a famous 222222222222222.
The problem is that if students get this item wrong, you dont know if it is because they
dont know about )icasso or because they dont know the word painter 3or artist4.
/. Don! 0n!+o.'/" n"$ !"/%n01'"# 0n !"#!#
5ne thing that confuses students in tests, is the presence of item types and techni-ues
that they have never seen before. ,n other words, if students are given a set of .umbled
words and asked to reorder them to make a sentence, we would e"pect them to have
seen this type of activity before in class. ,f the sentence6ordering activity is completely
new to them, they may have difficulties understanding how to do the -uestion which
have nothing to do with their knowledge of +nglish 3or lack of it4.
.. Don! 2o+)"! !o !"#! !%" !"#!.
,t is e"tremely unwise to write a test and give it straight to the students. 5ften unforseen
problems arise. )erhaps you forgot to write clear instructions. )erhaps there are some
mistakes. )erhaps the test is far too difficult 7 or far too easy. But especially if the test is
important for students 3and most tests are4 you must try to ensure that it works.
$ou should show it to your colleagues who can help to spot problems or misprints. ,f
possible try out the test with a class similar to your own. This is done to spot any
obvious mistakes or problems.
TEST ITEMS
/rammar is typically tested by means of what are called .0#/+"!"30!"4 !"#!#. That is,
individual components of the learners knowledge 3for e"ample, irregular past tense
verb forms such as went, saw, did etc.4 are tested using tasks such as gap6fills:
6!
8et us now look at a number of items that test a students knowledge of grammar.
We will start by looking at more discrete items and end by looking at testing techni-ues
that are slightly more integrative.
506 M',!07," /%o0/"
,n multiple choice items, students have to choose the correct answer from a
number of alternatives. 9t the most simple level, multiple choice can be used to test the
students grammatical knowledge, for e"ample:
:hoose the correct answer 9, B, : or ;.
. :harles 222222222222 to work yesterday.
9. doesnt go B. hasnt gone :. didnt go ;. ,snt going
'ultiple choice items can be made a greater test of all6round comprehension if
they are part of a passage or dialogue, for e"ample:
&ilary : Where are you 3on, in, off, out4 to%
<ane: ,m .ust going to the shops.
&ilary: :ould you 3post, take, bring, buy4 this letter with you%
<ane: $es, of course.
&ilary: 9nd youd better 3take, to take, taking, took4 an umbrella.
<ane: Why%
&ilary: Because its going to rain.
'uliple choice items like this have the great advantage of being easy to mark.
But it is difficult to write 0distractors 3the three wrong answers4, which arent either
absurd 7 or possible 7 as well as the correct answer. (pecial care should be taken to
make sure that there is only one correct answer.
'ultiple choice items like this test students recognition of grammatical items.
They are not tests of the students productive ability.
66
5006 F0,,30n#
1ill6ins are those items where students have to fill a blank with a word or words.
,n the traditional fill6in, students often see five or more separate sentences and have to
fill in a word for each. &owever, a more meaningful test would be to give the pupils a
passage with some selected words removed. 1or e"ample:
8ook at <ennys postcard to <oanne. There are some missing words. Write one
word for each space.
;ear <oanne,
=ome 2222 a beautiful city> The people 2222 very nice. (usan 222222 not like
the city very much, but 2222 likes the meals. 3(he loves shaghetti>4 (usan has
two friends in =ome 2222 )arlo and ?urt. 2222 live 2222 the north6west of
=ome. ?urt is an artist. 22222 pictures are very good. )arlo is ,talian and ?urt is
/erman. &e is 2222 &amburg. , 2222 speak ,talian or /erman, but )arlo and
?urt can speak +nglish very well.
(ee you soon,
8ove,
<enny
,n this e"ample the students have to understand the whole te"t at the same time and
then write the words in the spaces. This is a good test of students comprehension as
well as of their knowledge of individual grammatical items 3such as personal pronouns,
prepositions, and the verb to be, etc4.
1ill6in items are easy to write, although it is sometimes difficult to ensure that
students can put in only one answer> ,t is possible for students to to come up with more
than one word to put in the blank which could be correct.
50006 S"n!"n/" /o47,"!0on
1ill6in items usually ask for only one word. But they can be e"tended to test more of the
students knowledge and use of +nglish. (tudents have to fill in a blank and@or complete
a sentence with more than one word. Test items like this are usually called sentence
completions. 1or e"ample:
6"
:omplete the sentences below so that they make sense.
. 9: 9re you sure youll be all right%
B: ;ont worry 7 ,m 222222222 looking after myself.
!. Both teams were e"hausted. They 2222222222 for three hours.
#. The old lady, who 2222222222, suddenly sat up and asked for some tea.
A. &e was born between BAC and BA!, so he must be in 22222222222222.
(tudents who complete this task successfully show that they have a lot of
grammatical knowledge and that they are able to use the right vocabulary and grammar
to complete the task. 5bviously the writers of this test do not have only one correct
answer in mind, but deciding what to accept makes the test markers task -uite
comple".
50(6 S"n!"n/" +"o+."+0n)
,f students are used to this activity, then part of a test might look like this:
)ut the words in order to make correct sentences.
. he lives @ <ohn is @ and @ in 8ondon. @ a student
22222222222222222222222222222222222222222222222222222222
!. a housewife and @ is @ &is sister @ she is @ secretary. @ a
22222222222222222222222222222222222222222222222222222222
#. at home now, @ isnt @ in :anada. @ &is sister @ she is
222222222222222222222222222222222222222222222222222222222
This type of test item e"plores the students knowledge of synta" and is a useful
addition to a test.
5(6 T+an#2o+4a!0on#
9 test of the students knowledge of synta" and structure is sentence transformation.
&ere students have to rewrite sentences so that they have the same meaning but
different grammatical structures. 1or e"ample:
6#
:omplete the sentences so that they mean the same as the original sentence.
(tart with the words given.
. <ohn is taller than 'ary.
'ary isnt 222222222222222222222222222222222222222222222222.
!. 0, havent seen her for years, he said.
&e said that 222222222222222222222222222222222222222222222.
#. , wont come unless you ring.
,ll come 222222222222222222222222222222222222222222222222.
5nce again, students should have practised transforming sentences before they come
across this e"ercise. But the ability to transform sentences correctly certainly implies
-uite a lot grammatical knowledge 7 although it may take a certain kind of 0intelligence
to be very good at it.
5(06 S"n!"n/" $+0!0n)
(tudents can practise their sentence writing by describing a picture.
This activity certainly tests the students ability to write correct sentences 7 and their
ability to use there with is and are correctly.
There
is
are
5(006 Pa+a,,", $+0!0n)
5ne way of providing a fairly controlled integrative test type is through parallel writing.
&ere you ask the students to use their knowledge of grammar and vocabulary to imitate
a piece they have read 3and understood4.
,n order to complete this successfully, students need to understand a paragraph, its
form and its grammar. 5nce again, the teacher should be sure that pupils have
previously used this techni-ue in class since it might otherwise cause confusion.
S'44a+&
6$
,n this chapter we have looked at a number of e"ercises which test grammar, together
with some 0rules or hints about test design. We have looked at the need for test
e"ercises which encourages the students written production as well as terms which
concentrate as accuracy.
T9(?(:
. Write some multiple choice items to test the students knowledge of the
difference between simple and continuous verb.
!. Write a five item sentence6completion e"ercise for a final test at upper primary
level. Try to make all the items refer to the same conte"t.
#. :hoose a specific language point and write a fill6in e"ercise to practise it.
A. Write a short test for beginner students to use as a parallel writing activity.
D. ;ecide which type of written practice you prefer and say why.
T%" "n. o2 !%0# #"##0on8)oo. ,'/- $0!% &o'+ a/!0(0!0"#9
"%