Assessment

Handbooks for the Engl i sh Cl assr oom
Assessment
Mi chael Harri s and Paul McCann
M A C M ILLA N H EIN EM A N N EN G LISH LA N G U A G E TEA C H IN G
A division of M acm illan Publishers Ltd
M acm illan O xford, Betw een Tow ns Road O X4 3PP
ISBN 0 435 28252 2
M ichael H arris, Paul M cC ann 1994, M acm illan Publishers Ltd
First published 1994
A ll rights reserved; no part of this publication m ay be reproduced, stored in a retrieval system ,
transm itted in any form , or by any m eans, electronic, m echanical, photocopying, recording, or
otherw ise, w ithout the prior w ritten perm ission of the publishers.
D esigned by M ike Brain
C over illustration by Jacky Rough
Illustrated by:
N ick H ardcastle
G eoff Jones
The authors w ould like to thank Pauline Rea-D ickens, A lex Teasdale and A nnie M cD onald for helpful
suggestions, C hristina and M aria Jos for support and patience.
Printed and bound in Scotland by Thom son Litho Ltd.
94 95 96 97 98 99 9 8 7 6 5 4 3 2 1
M ichael H arris, Paul M cC ann 1994, M acm illan Publishers Ltd 1
CONTENTS
Introduction 2
Prejudices and problem s 2
Suggestions and solutions 3
C hapter 1 Inform al assessm ent 5
1.1 Introduction 5
1.2 W hen and w hat to assess 7
1.3 Inform al assessm ent of linguistic factors 9
1.4 Inform al assessm ent of non-linguistic factors 21
1.5 Results from inform al assessm ent 24
A ction points 25
C hapter 2 Form al assessm ent 26
2.1 W hy test? 26
2.2 Planning assessm ent program m es 30
2.3 C hoosing test form ats 34
2.4 W riting, adm inistering and m arking tests 49
2.5 Results from form al assessm ent 60
Suggested answ ers 61
A ction points 62
C hapter 3 Self-assessm ent 63
3.1 Introduction 63
3.2 Laying the foundations 68
3.3 A ssessing perform ance 74
3.4 Review ing progress 80
3.5 Results from self-assessm ent 86
Suggested answ ers 88
A ction points 88
G lossary of term s 89
Bibliography 94
2 M ichael H arris, Paul M cC ann 1994, M acm illan Publishers Ltd
INTRODUCTION
What is assessment?
This handbook deals w ith assessment. It is very im portant to m ake a clear distinction
betw een assessment and evaluation. A s teachers, w hen w e carry out assessment,
w e have to m easure the perform ance of our students and the progress they m ake.
W e also need to diagnose the problem s they have and provide our learners w ith
useful feedback. Evaluation on the other hand involves looking at all the factors that
influence the learning process, such as syllabus objectives, course design, m aterials,
m ethodology, teacher perform ance and assessm ent. A ssessm ent and evaluation are
often linked, because assessm ent is one of the m ost valuable sources of inform ation
about w hat is happening in a learning environm ent.
Prejudices and problems
A ssessm ent is generally seen as som ething done to students by teachers. M any
students m ay feel panic and confusion. Tests descend upon them from tim e to tim e
and have to be got through. The m ore able ones m ay even enjoy these
experiences, as they can assert their superiority over the rest of the class. H ow ever,
m any students feel anxious, w orried and inadequate. There is often great pressure
on them to succeed and if they do not, they becom e branded as failures.
U nfortunately this com petition creates m ore losers than w inners. M any teachers feel
little better. Som e feel that tests are only useful as a w ay of m otivating students to
w ork harder and virtually all of us feel insecure and uncom fortable w hen w e have to
pass or fail students.
M any of these negative attitudes tow ards assessm ent com e from the generalised
feeling of a divorce betw een learning and teaching on the one hand, and
assessm ent on the other. The fundam ental reason for this is that assessm ent often
does not feed back into the learning and teaching process.
N Firstly, assessm ent is often seen as synonym ous w ith testing. Testing or formal
assessment, w here test or exam conditions are established, is certainly an
im portant w ay of assessing learners. H ow ever, it is not the only one and both
inform al assessm ent and self-assessm ent are vital. Informal assessment is
assessm ent carried out by the teacher not under special test conditions, but in the
norm al classroom environm ent (for exam ple w ith students helping each other
w hen necessary). Self assessment is that carried out by students them selves of
their ow n progress and problem s.
N Secondly, assessm ent is seen as som ething that happens after learning has
finished, rather than during the learning process. Tests or exam s are often given
at the end of term or of a course and they are often regarded as a rather painful
but necessary prelude to the holidays. For m any learners the inform ation that this
assessm ent gives them about their perform ance and progress com es m uch too
late to be formative, to feed into their ow n learning.
N A nother problem is the kind of feedback that learners are given. O ften inform ation
about how w ell learners are doing is expressed only by a grade or m ark. The grade
classifies students, but does not give any real help by telling students w hat their
specific problem s are and by m aking suggestions to overcom e them .
N A n additional draw back of the w ay assessm ent is carried out in m any classroom s
is that it often concentrates on only one part of w hat has gone on in the
classroom . It is easier to test tangible know ledge of gram m ar for exam ple, than
to assess perform ance at say speaking. A n over-reliance on gram m ar tests gives
students the clear m essage that they have been w asting their tim e trying to
com m unicate in class. W hat m atters is gram m ar.
Introduction
N Yet another reason for negative attitudes to assessm ent am ong students is that,
rather than give them the opportunity to show w hat they have learnt, it tries to
catch students out, to reveal w hat they have not learnt. Learners also feel
alienated by assessm ent because they have no role in it, apart from as passive
participants. For m any learners in this situation, assessm ent m ust seem arbitrary
and at tim es even unfair. Som etim es they get on w ith their teacher, som etim es
they do not. Som etim es they are lucky and revise the right m aterial for a test,
som etim es they are unlucky.
O n the other hand, surprisingly little help is given to teachers to assess effectively
and fairly, to carry out assessm ent that reflects w hat has gone on in the classroom .
The result is that as teachers w e are often on the defensive w hen dealing w ith
learners, parents and school adm inistrators.
A n added problem is often that of the effect of public exam inations, both national
and international. N ot only is the stress factor for learners greatly increased, but the
exam inations can have a profound washback effect (the influence of assessment on
both teaching and learning). In som e cases exam inations can dom inate w hat goes
on in language classroom s for the m onths and even years leading up to them . If all
these exam s w ere forw ard-thinking and com m unicative this w ould be positive.
U nfortunately this is not alw ays the case.
Suggestions and solutions
W hat m odel of assessm ent can be offered as an alternative to this gloom y picture?
Let us look briefly at three of the basic questions that this book w ill try to answ er:
When, who and how should we assess?
When should we assess?
To provide constant feedback w hich itself helps to direct the learning process,
assessm ent has got to be a continuous process going on at different levels. A t an
everyday level it is im portant for som e kind of inform al assessm ent by the teacher to
be done in nearly every lesson, hand in hand w ith learning. M ore in-depth form al
assessm ent should then be done at frequent points throughout the course, giving
feedback to both the learner and the teacher (rather than w aiting for the end of
term ). It is im portant for this to be integrated into the course, not only reflecting
course content, but in m any cases developing and extending topic areas covered. A t
the sam e tim e self-assessm ent needs to be done at regular intervals, so that learners
can be given an opportunity to think about w hat progress they are m aking and
w hat their problem s are. A t the end of a course all the assessm ent that has gone on
should then be put together and final decisions about studentsprogress m ade.
Who should assess?
A ssessm ent has traditionally been the exclusive dom ain of the teacher. W ithout
doubt w e should not abdicate this responsibility for assessing classw ork and
hom ew ork inform ally and for setting and m arking tests that are done under form al
conditions. H ow ever, it is possible to com plem ent and im prove our judgem ents by
asking students to participate in the process them selves.
N ot only do w e gather useful inform ation through self-assessm ent, but involvem ent
of students in the assessm ent process also m eans that their attitudes tow ards their
ow n learning can change significantly. Rather than being m otivated by the threat of
exam inations, students can start to feel m ore responsible for their ow n progress,
thus acquiring greater and m ore intrinsic m otivation for learning. A t the sam e tim e,
self-assessm ent can help students to becom e m ore efficient as learners, to diagnose
their ow n w eaknesses and problem s and then to try to do som ething about them .
Introduction
Self-assessm ent is a relatively new concept. H ow ever, as w e w ill see in Chapter 3,
involving learners in this aspect of our jobs does not threaten our position. O n the
contrary, it helps us by giving us m ore inform ation as w ell as by training learners to
think for them selves.
Finally, in addition to enlisting the help and participation of our learners in
assessm ent, w e should w ork w ith as m any colleagues as possible rather than
w orking on our ow n. A co-operative approach saves both tim e and w ork, and pools
vital know ledge and resources.
How should we assess?
A ssessm ent m ust be done constructively, focusing on achievem ent rather than on
failure. It should allow students to dem onstrate w hat they know , rather than trying
to catch them out. A s classroom teachers w e are assessing our learnersprogress
and m ost of our students should be able to pass. If this is not the case, w e have to
take another look at w hat has gone on during the course.
A ssessm ent m ust also have som e degree of reliability ie that it is consistent and that
under the sam e conditions and w ith the sam e perform ance by students our
assessm ent produces the sam e or at least sim ilar results. If w e do not establish clear
criteria and w ork out clear procedures for assessing beforehand and try to keep to
them , there is a danger that w e w ill discrim inate against som e students. W e should
also inform students of at least the basic outlines of our assessm ent, so that not
only is our assessm ent reliable, but it is seen to be reliable and fair by our students.
It is also very im portant to be clear about w hat w e w ant to assess and to ensure
that w e are assessing that and not som ething else, that our assessm ent has validity.
For exam ple, if w e w ant to assess listening w e m ust only consider understanding
and not assess our studentsability to read or w rite or their ability to produce
accurate language. To reach the goal of validity in our assessm ent it is thus very
im portant that w e have clear assessm ent objectives in the first place and then try to
m ake sure that these objectives are reached.
A nother im portant feature is practicality. A ny approach to assessm ent m ust not be
too tim e-consum ing, in term s of class hours and of our ow n tim e outside the class.
A ssessm ent is only one aspect of our jobs and cannot be allow ed to detract from
teaching or preparation tim e. It should also be practical in term s of physical
resources such as tape-recorders and photocopies.
A final elem ent is accountability. A s professionals, teachers should be able to
provide learners, parents, institutions and society in general, w ith clear indications
of w hat progress has been m ade and if it has not, w hy that is so. W e should be able
to explain the rationale behind the w ay assessm ent takes place and how conclusions
are draw n, rather than hiding behind a sm oke screen of professional secrecy.
In the follow ing pages of this handbook w e aim to discuss som e of the issues in
assessm ent, to look at som e of the options open to us as classroom teachers and to
provide practical ideas for teachers to try out in the classroom .
N First, w e w ill look at informal assessment done by the teacher.
N Then w e w ill exam ine formal assessment or testing, w hich provides us w ith
additional inform ation about our students.
N Finally, w e w ill consider assessm ent done by the students them selves, self-
assessment.
A t the end of each chapter w e w ill think about how to deal w ith results from each
form of assessm ent.
Inf ormal assessment
1.1 Introduction
The first of the three chapters in this book w ill look at inform al assessm ent in and
outside the classroom . W e w ill look at w hen and how to carry out this kind of
assessm ent of both linguistic and non-linguistic factors (like attitude). Finally w e w ill
look at w hat to do w ith the results from this kind of assessm ent.
Inf ormal assessment is:
Informal assessment is a w ay of collecting inform ation about our students
perform ance in norm al classroom conditions. This is done w ithout establishing test
conditions such as in the case of form al assessm ent. Inform al assessm ent is
som etim es referred to as continuous assessm ent as it is done over a period of tim e
like a term or an academ ic year. H ow ever, form al assessm ent can also be done
continuously as w e w ill see in the next chapter.
W hen w e are in the classroom w ith our students w e intuitively assess their
perform ance w hen speaking, w riting, reading or listening. W e can see w hich
students are perform ing w ell and w hich students are finding difficulties. W e are also
aw are of studentsattitudes, how m uch effort they are m aking and how m uch they
are participating in class activities. H ow ever, to carry out effective inform al
assessm ent w e need to carry out system atic observation.
N Firstly w e need to w ork out w hat w e are going to assess, as it is obviously
im possible to assess all studentsperform ance all the tim e.
N Then w e m ust establish clear criteria for assessing students and not only rely on
rough im pressions.
N Finally it is im portant to link the inform al assessm ent w e do w ith our form al
assessm ent (tests) and w ith self-assessm ent done by the students them selves.
The inform ation w e give students should help them identify areas w hich m ay cause
them difficulties w hen they do form al tests. W e m ust also give our students
feedback and help them to think for them selves about the inform ation, as w ell as
heightening their aw areness of how they assess them selves.
Inf ormal assessment is not:
N Informal assessment is not a replacem ent for other form s of assessm ent such as
form al or self-assessm ent and should not be considered as the only w ay of
obtaining and giving inform ation about our students.
N It is not a w ay of avoiding tests altogether w hich allow s the teacher to m ake
decisions about students based purely on inform al observation.
N It is not a form of assessm ent w ithout criteria on w hich to m ake judgem ents
about studentsprogress.
N It is not a single system w hich can be used by all teachers in all contexts in all
schools or teaching establishm ents each teacher or group of teachers need(s) to
find a system w hich is suitable to their particular situation.
N It is not a form of assessm ent to be considered in isolation from other form s of
assessm ent but rather as part of a larger, overall program m e w hich should be a
balanced system of decision-m aking.
1
C HA P T E R 1 Inform al assessm ent
6 M ichael H arris and Paul M cC ann 1994. This page m ay be photocopied for use in the classroom .
Informal assessment Part of a system
Work with a colleague or colleagues.
Look at the statements below about assessment.
Discuss each statement and then assign a grade for each statement in
the following way:
5 Agree strongly
4 Agree
3 Indifferent
2 Disagree
1 Disagree strongly
1 Informal assessment of language skills is the most important area
for us to assess informally.
5 4 3 2 1
2 Non-linguistic factors should not be assessed this is the job of
psychologists.
5 4 3 2 1
3 Informal assessment is simply the systematic observation of our
students.
5 4 3 2 1
4 Informal assessment could replace all those time-consuming tests
that I have to give.
5 4 3 2 1
5 Informal assessment should be part of an overall system of
assessment.
5 4 3 2 1
6 Informal assessment means what it says : informal I observe
and make a judgement based on my experience rating scales are
a waste of time.
5 4 3 2 1
7 An adequate system of informal assessment already exists in my
school we take in essays and compositions and mark them.
5 4 3 2 1
8 Informal assessment can only take place in the classroom.
5 4 3 2 1
9 Even without set criteria, I always mark written work in the same
way I know what I am looking for.
5 4 3 2 1
10 Students sometimes act differently when I walk round the class
and listen to their conversations.
5 4 3 2 1
1.2 When and what to assess
Traditionally, w e tend to think of this type of assessm ent as consisting solely of
m arking an essaydone for hom ew ork or the teacher w alking round the class
during an activity to get a vague idea of w hether the students are doing all rightor
not. The classroom is obviously the m ost im portant place to assess students
inform ally. It is also w here m ost data is available. Students spend m ore tim e here
than in any other learning environm ent and here w e can m onitor students
perform ance in all four skills speaking, listening, reading and w riting. H ow ever w e
can also assess w ork outside the classroom , eg looking at sam ples of studentsw ork
or collecting in w orkbooks and vocabulary books.
O ne of the first things w e need to do is to w ork out how m uch w e are going to
assess as there are m any other aspects of our job that require attention both in and
outside the classroom . In the classroom w e need to think about classroom
m anagem ent and reaching the aim s of the lesson first. The tim e w e have for
assessm ent is lim ited. Sim ilarly, outside the classroom w e have lots of other things
to do like planning and m aterials creation. W e m ust m ake sure that m arking does
not take up all our tim e. Because of all these pressures, it is useful to think about
w hat things w e are going to assess consciously and w hich things w e are just going
to get an im pression of.
Think about how you carry out informal assessment with your
students. Look at the list of items below. Either on your own or with a
colleague or colleagues, decide which of the items you assess informally
by giving a mark and which you merely get an impression of.
I get an overall I give a specific
Linguistic factors impression mark
a written homework
b written grammar activities
c speaking activities
d projects
e listening tasks
f reading tasks
g writing tasks
h vocabulary activities
Non-linguistic factors
a attitude/effort
b participation in class
c group work
d organisation of work
e presentation of work
f punctuality
Are there any other factors that you consider?
M ichael H arris and Paul M cC ann 1994. This page m ay be photocopied for use in the classroom . 7
In addition to thinking about w hat to assess it can be useful to think about the
weighting that you are going to give each area. W e w ill look at this in m ore detail
w hen w e look at planning assessm ent program m es in the next chapter. H ow ever, it
is w orth thinking about your priorities at this stage; w hich areas you are going to
give m ost m arks for. The priorities that you have w ill be those outlined in the
syllabus plan of the school that you w ork in, but it is w orth establishing clearly the
weighting in your inform al assessm ent at the beginning of a year. To decide w hat is
a pass or a fail you need to balance the four skills, language and non-linguistic
factors. For exam ple you m ay w ant to w eight the final assessm ent in favour of
speaking or listening if you have spent proportionately m ore class tim e on these
skills. In this case you should give a higher percentage of the overall score to these
areas and less for other areas like language and w riting.
A nother key consideration is how m uch inform al assessm ent you are going to do
com pared w ith form al assessm ent or testing. If there are a lot of tests during the
term or year, then inform al assessm ent on a large scale m ay not be adding m uch to
the inform ation gained about a student. H ow ever if there are very few tests,
inform al assessm ent m ay take on a m ore im portant and influential role and m ay
contribute m ore inform ation than in the previous case. Everything depends on
specific institutions and the w ay each institution w orks w ill affect the exact
w eighting of each area of assessm ent.
If you have freedom to decide yourself on the balance betw een inform al and form al
assessm ent in your classroom , it is w orth thinking about it before starting off a
course. It is also im portant to inform your students so that they have a clear idea
about how they are going to be assessed.
Weighting: informal/formal assessment
What is an ideal balance for your classes between informal
assessment and formal assessment or tests?
Which of these comments reflects your classroom situation?
I do nearly all my assessment informally. I have small classes of
about fifteen to twenty students and we have quite a few hours of
English a week. I suppose that I get 80% of my marks from
informal assessment and 20% from tests.
I have about twenty-five or thirty students in my classes and we
only have three hours of English a week. I get 50% of my marks
from informal assessment: the work they do in class, projects and
homework. The other 50% I get from short tests which I give
throughout the term.
I do nearly all of my assessment formally, through tests. I get some
marks from work that students do in class and I always give them
an oral mark. However I have over forty in each class and we only
have two hours of English each week. Because of this I have to get
most of my marks from short tests which I give during the term. I
suppose that I get 70% of my marks from tests and 30% from
informal assessment.
1.3 Inf ormal assessment of linguistic f actors
The m ost com m on w ay of assessing students inform ally has traditionally been by
m arking pieces of w ork that students do either in class or for hom ew ork. These
m arks are then added up at the end of the assessm ent period to w ork out a final
grade. W ithout doubt this m arking of studentsw ork is one of the m ost im portant
sources of inform ation that w e have. H ow ever, this approach has serious draw backs:
N Firstly, it m eans that w e m ight tend to concentrate on w ritten w ork and on
gram m ar exercises rather than focusing on oral skills.
N Secondly, w hen w e m ark w ork w e m ight assign m arks on the basis of the
im pressionthat w e have of the w ork. W e are not thinking in detail about w hat
exactly w e expect the students to achieve.
N Finally, w hen w e add up m arks at the end of term w e tend to m ake judgem ents
in term s of im pressions and w ithout a clear idea of w hat w e are assessing.
Therefore, it is im portant to m ake a conscious effort to assess oral skills during
classes and establish clear criteria to help us assess specific perform ance by students
and com e to decisions about studentsprogress.
A ssessm ent criteria can be described in term s of w hat w e expect our students to be
able to do. This m ay be Yes/No or Pass/Fail ie the student can or cannot satisfy a
certain criterion. A n exam ple m ay be ability to w ork w ithin a group. W ith linguistic
factors, how ever, it is m uch m ore likely that students can satisfy criteria to a greater
or lesser degree ie they w ill be im proving along a continuum on w hich at the
present m om ent they are quite good at som ething and w ith m ore practice they w ill
becom e better, eg oral expression.
In these cases a m ore descriptive system of grading is needed than Yes/No or
Pass/Fail. A n alternative is to grade student perform ance in these areas into a
num ber of bands, eg 0 is a poor perform ance and 5 is an excellent perform ance. W e
can then describe each level or band. The m ore bands w e have the m ore delicate
and descriptive the system w ill be. A t the sam e tim e, the m ore delicate and
descriptive the system is, the finer the distinctions you w ill have to identify in
student perform ances. This m ay prove to be m ore difficult than it sounds
assigning a student a 13 instead of a 14 on a 20 band scale m ay prove to be a very
difficult task due to the fine distinctions betw een the tw o bands. O n the other
hand, the few er the bands w e have, the easier the task is to assign band num bers to
students as the distinction betw een a 3 and a 4 on a 5 band scale w ill be greater
than on a 10 band scale and w ill therefore be m ore easily identified. O f course, the
few er the bands, the m ore rough and ready the system is.
Bearing all this in m ind, w hat w e need is to find the optim um num ber of bands w ith
clear and easily understandable band descripters for our purpose. If w e find this,
then w e w ill have a system w hich is usable and suited to our needs.
W e w ill now look in m ore detail at assessm ent of the fours skills, of language and
assessm ent of non-linguistic factors.
Speaking
Inform al assessm ent is particularly im portant for speaking as m any teachers have
practical difficulties in organising oral tests (lack of a place to do them or lack of
tim e). Because of this, oral assessm ent in m any situations has to be done inform ally.
In addition, inform al assessm ent of speaking can have an im portant effect on
learners. If they see that speaking English and participating in class is rew arded they
w ill be m ore m otivated to participate actively and try to use English in the class.
Inform al assessm ent of speaking is done by observing studentsoral perform ance in
class, by m onitoring speaking activities as they happen. In a busy classroom w hen
you have lots of other things to do this is not alw ays easy. If you are w alking around
the class m onitoring a class oral activity (for exam ple all students are w orking in
sm all groups discussing a particular issue or solving a problem ) you m ay w ell obtain
a fairly good idea of class perform ance as a w hole and individual perform ance in
particular areas. You m ight com e to the conclusion that a particular class needs
rem edial w ork in pronunciation, or that a particular students intonation is not as
acceptable as that of other m em bers of the sam e class.
A s you w ill probably spend a lot of tim e w ith your classes you w ill gain m any insights into
these areas as a rapport builds up betw een you and the students and you begin to
function as one w orking unit w ith m utual confidence. It m ust be pointed out here that
your intuitive judgem ents are often very useful. O ne w ay in w hich you can system atise
this gathering of im pressions is by giving students points based on pre-defined criteria
w hen you see that they are perform ing w ell, either in groups or in front of the w hole
class. A t the end of the assessm ent period you can then add up the points that students
have to get an idea of how w ell they are doing.
H ow ever, the question of unreliability and inaccuracy of judgem ent should be
considered in the follow ing w ays. Firstly w e m ay get a distorted view of overall class
ability if w e cannot listen to all students during all activities as so often happens
w ith large num bers of students. Secondly, w e m ay get a distorted view of particular
students w ho m ight try harder w hen they know w e are listening to them ; on the
other hand, students m ay perform w orse in the know ledge that the teacher is
actively m onitoring them .
The first thing w e need to do is to produce a system w hich attem pts to give us the
possibility of m aking reliable and objective judgem ents about our students. The
criteria w e choose and the descriptors for each criteria, eg on a 5 band scale w ill
provide a degree of reliability w e w ould not have if w e used no criteria at all except
our intuition. The criteria m ight focus on a particular aspect of speaking, eg fluency,
intonation, pronunciation, self-correction w here necessary etc. W hat our criteria w ill
do for us is allow us to place a student on a scale according to his/her perform ance
at any one given m om ent in tim e.
W e can establish our ow n criteria, but if other teachers in the school are using
different criteria there can be a considerable am ount of unreliability in assessm ent in
the school. Therefore it is im portant firstly to agree on criteria for assessing students
w ith our colleagues. The follow ing step is to m ake sure that w e all agree on exactly
w hat the assessm ent scales m ean. This standardisation of perceptions can be
achieved through m eetings, discussions and w here possible listening to recorded
perform ances and using these for standardisation purposes.
O ur bands m ay have m ore than one aspect of speaking and m ay describe the
speaking skill as one or tw o general criteria. Look at the follow ing band scale:
5 Speaks fluently alm ost no errors
4 Speaks quite fluently som e errors
3 Som e difficulty in speaking m any errors
2 D ifficulty w ith speaking alm ost incom prehensible
1 U nable to use language incom prehensible
W hile this scale m ay serve as a rough guide to the fluency and num ber of errors our
students are m aking and w hile it m ay be possible to place our students on this scale
as a result of inform al observation, there are certain questions and problem s w e
w ould have to clear up before using it; What does fluently actually mean? What or
how much is some in band scale 4? Do the bands move from one to the other in a
uniform way or is there a big jump between one particular band scale and another?
and m ost im portantly Is this exactly what we want to assess?
Oral assessment criteria
Look at the scale on page 10 and, if at all possible with your
colleagues, write a fuller description of the bands. Think about
including descriptions of these things:
fluency (speed/amount of hesitation)
message (relevance and appropriacy)
accuracy (grammatical and lexical errors)
pronunciation (sounds/intonation/stress)
Compare your descriptions with those on page 44
O nce you and your colleagues have decided on the assessm ent criteria and the type
of scale to use, you w ill w ant to put the system into practice. H ow can you
im plem ent the system ? W hat follow s is a list of suggested w ays of carrying out
system atic observation and recording the results. H ow ever, you should adapt these
m ethods to your ow n particular situation and w here necessary think of other w ays
of system atically observing and recording your findings w hich m ay be m ore suited to
your needs. N o single m ethod is suitable in all teaching situations or all institutions.
N The first step is to produce band scales to assess oral perform ance. You can either
do this on your ow n or (m uch better) w ork them out w ith your colleagues.
H aving done this you can then use them to assess your students com petence at
any given tim e, eg at the end of each m onth or at the end of each term . Your
judgem ents w ill be based on your observation of students in the preceding
period. It is also possible to give your students the sam e bands and ask them to
assess them selves. Your assessm ent and that done by learners can then be
com pared. Furtherm ore, assessm ent criteria should be m ade available to
everybody concerned including institutions and parents of students. (For further
discussion, see self-assessment page 63).
N A nother alternative is to assess studentsperform ance at a particular tim e. The
best w ay of doing this is perhaps to tell students that you w ill be assessing them
over the next few lessons. Then you can m ake sure that in the lessons there are a
series of speaking tasks. W ith large classes it is probably a good idea to choose
five or six students to assess in each lesson (though do not tell them as w hen they
know they have been assessed they m ay stop m aking an effort!). D uring the tasks
you can go around the class and w rite dow n your assessm ents of students
perform ance. The best tasks to use are those that students norm ally do in class
and in general group or pairw ork activities are the m ost suitable: pairw ork
interview s, group surveys, roleplays, inform ation gap activities etc. You can
though use som e w hole class activities, w here students give a presentation
(individually or in groups) or w here they act out roleplays that they have prepared.
Both of these activities can be a good conclusion to project and task based w ork
and also provide good opportunities for peer and self-assessm ent w ith
assessm ent bands given to the students. The first m ethod suggested m easures
potential perform ance this m ethod m easures actual perform ance in tasks. A s w ell
as assessing speaking by oral bands alone, they can be used in conjunction w ith a
set of pre-defined criteria w hich are objectives to be aim ed at and w hich are
Yes/No type criteria, eg participates in discussions. They can also be used along
w ith m ore system atic im pression m arking, for exam ple giving students points for
good perform ance throughout the term .
N The last question to ask is When we should carry out informal assessment of
speaking? This w ill depend on the tim e w hich is available and the num ber of
hours of class you have each w eek. It is obvious that the m ore hours you have in
the classroom w ith the students, the greater the opportunity to assess. A nother
factor to consider is the frequency w ith w hich you have to give reports on your
students. If results of assessm ents are to be reported each term , then the inform al
assessm ent program m e w ill have to take this into account and w ill have to tie in
w ith the other form s of assessm ent being used to m ake up the overall assessm ent
of each student. If results are to be reported each academ ic year, the inform al
assessm ent program m e w ill have to take this into account in the sam e w ay.
Writing
A ssessing our studentsw ritten w ork can be very tim e consum ing and unless you
are careful you can find yourself w ith m any hours of m arking, w hich takes tim e
aw ay from other aspects of teaching like lesson planning. It is thus very im portant to
choose the m ost im portant pieces of w riting that students do and not try to assess
every piece of w ritten w ork. Thus w e need to decide how m any and w hat sort of
com positions w e are going to assess. Younger secondary learners m ay not need to
do as m uch w riting as older secondary students w ho have to do w riting in public
exam inations. The degree of im portance that w riting has should be reflected in your
syllabus plans. A nother point to consider is the am ount of w riting w hich should be
done individually and that w hich is done in groups. G roup w riting activities are not
only good for developing w riting skills, particularly in a m ixed ability situation, they
can also m ake your m arking load m ore viable and enable you to give m ore feedback
to m ore of your students. For exam ple, in a large class of over forty it is m ore
practical to m ark ten group com positions every w eek than forty individual ones.
A nother point to consider is giving feedback. W e need to m ake very sure that the
feedback w e give to students is used by them . It can be very frustrating to spend a
lot of tim e m arking w hen students just look at the m ark and nothing else. W hen w e
give feedback to our students it is im portant that they think about it them selves and
w ork out how they can do better in the future. Therefore the link betw een inform al
assessm ent of w riting and self-assessm ent is vital.
A nother crucial factor is the need to m ake sure that the assessm ent w e do of our
students is as reliable as possible. For exam ple, if you are correcting or assessing an
essay or com position, you w ill often give an im pressionor im pressionistic m ark.
This m eans that you w ill read through the piece of w ork once, tw ice or perhaps
m ore tim es, and give it a m ark. This could be som e point on a scale of 110 or a
letter from A E. The chances are that if you are assessing the w ork from a class like
this you w ill apply som e sort of criteria w hich you m ay have in your head, w ithout
consciously listing them . Even if you are w orking on your ow n it w ould probably be
m uch better if you actually listed these criteria, and it w ould certainly be very useful
for students to know exactly w hat they are.
H ow ever, w e need to think about the question of unreliability and inaccuracy of
judgem ent in several w ays. Firstly, w hen w e are m arking by im pression it m ay
happen that w e give a different m ark to the sam e piece of w ork on a different day
or even part of a day. For exam ple, w hen w e are tired at the end of the day w e
could give a com position a low er m ark than in the m orning. C onversely w e m ight
be m ore generous on a Friday than on a M onday. Secondly, as w e are all different
w e m ight have different sets of criteria. Thus a student could get a different m ark
for the sam e piece of w ork if it w ere m arked by different teachers.
A s w ith speaking w e need to think very carefully about w hat criteria w e use.
Look at the follow ing scale:
5 Excellent w riter
4 G ood w riter
3 M odest w riter
2 M arginal w riter
1 Poor w riter
If w e use this scale, w e need to define previously w hat the m eaning of the
descriptors is.
For example:
W hat m akes a student an excellentw riter rather than a goodw riter?
N Is it com prehensibility?
N Is it gram m atical accuracy?
N Is it spelling?
N Is it the w ay they organise the text?
W hen w e are w orking out rating scales and descriptors to use there are tw o
possibilities. W e can either produce overall descriptions of w riting ability or
concentrate on different aspects of w riting and separate these aspects out into
individual scales. G eneral overall descriptions, ie those containing different features
of ability, are called holistic scales. Those w hich separate out different aspects of
ability are called analytic scales.
Example of holistic scale for writing ability:
5 C onstructs gram m atically correct sentences and phrases and show s full
m astery of appropriate vocabulary. Text is organised coherently. C orrect spelling
at all tim es. Excellent content and presentation. M essage w holly relevant
4 In general, gram m atically correct sentences and phrases but som e errors w hich
do not affect understanding. D oes not use appropriate vocabulary at all tim es.
Som e difficulties w ith organisation of text. Som e errors in spelling. G ood
content and presentation. M essage m ostly relevant.
3 G ram m atical errors and use of vocabulary affects understanding as does
organisation of text. M any errors in spelling. Satisfactory content and
presentation. M essage not alw ays relevant.
2 Text understood w ith difficulty due to inaccurate gram m ar and inappropriate
use of vocabulary. Poor content and presentation. M essage generally lacks
relevancy.
1 Im possible to understand text due to frequency of gram m atical errors and
incoherence of organisation. Poor spelling M essage irrelevant.
Informal assessment Number of bands
Look at the 1 5 band scale in the box below.
Work alone or with a colleague or colleagues and expand the five
bands into a ten band scale by writing more detailed descriptions.
5 Constructs grammatically correct sentences and phrases and shows
full mastery of appropriate vocabulary.
Text is organised coherently and is well structured.
Correct spelling at all times.
4 In general, grammatically correct sentences and phrases but some
errors which do not affect understanding.
Does not use appropriate vocabulary at all times.
Some difficulties with organisation of text.
Some errors in spelling.
3 Grammatical errors and use of vocabulary affects understanding as
does organisation of text. Many errors in spelling.
2 Text understood with difficulty due to inaccurate grammar and
inappropriate use of vocabulary.
1 Impossible to understand text due to frequency of grammatical
errors and incoherence of organisation. Poor spelling.
When you have finished, answer the following questions:
What difficulties did you encounter when writing the expanded
scale?
Do you think that with your expanded scale teachers would find it
easy to differentiate between a 6 and a 7 or a 3 and a 4?
Does each activity, (eg spelling) appear in each band?
Do you think that your holistic band scale could or should be
broken up into analytic scales?
By looking at your ten point scale, would it be obvious what is an
acceptable level, ie the cut-off point or pass mark?
Analytic scales separate the aspects or activities given in the box above and thus
cater for the student w ho cuts across the bands and their descriptors. For exam ple, it
is possible that a student m ay be a 3 in the scale above for gram m atical accuracy but
a 2 for spelling. H ow ever, both analytic and holistic scales can be used successfully to
assess ability if the descriptors are clear enough and if perceptions are standardised.
If a student falls into m ore than one band on a holistic scale, the teacher should be
able to choose the m ost appropriate or best fit for that students w ork.
W hen m arking studentscom positions you m ay find that rather than trying to
m easure perform ance com prehensively, you can focus on a couple of aspects. For
exam ple, w hen assessing a story you could focus on text organisation and linking,
not m arking gram m atical accuracy or spelling. For a project poster giving
inform ation about a country you could focus on content and presentation. W hile
this approach does not m easure all aspects of w riting it is a good w ay of giving
students m anageable feedback.
W hatever set of criteria you use it is im portant to tell students w hat they are
beforehand so they w ill be aw are of w hat is expected of them and w ill hopefully
think m ore closely about your assessm ent w hen they receive it. The clear
establishm ent of criteria enables them to assess their ow n w ork as w e w ill see in the
chapter on self-assessm ent. A nother thing to consider is how you are actually going
to give feedback to your students. If you correct everything and w rite the
corrections on their w ork students can be very de-m otivated. There is nothing w orse
than receiving a piece of w ork back w hich is covered w ith red ink! O ne w ay in
w hich you can do this and at the sam e tim e get students to think about their
m istakes is by using a correction code. Through a code you can signal problem s that
students are having. W hen you give w ork back students can w ork either individually
or in groups to see if they can correct their ow n w ork. W hile they are doing this you
can go around the class and answ er any problem s that they m ight have.
Correction codes
Look at this correction code. Work out your own code to use with
your students.
S spelling V vocabulary
P punctuation G grammar
WO word order Pa paragraphs
L linking Pr prepositions
When you use a code with your students, go over it on the board and
to explain give them examples of mistakes and problems.
W hen you m ake judgem ents about your learnersprogress in w riting over a period
of tim e you can add up the results that they have had in their com positions or use
the w ork they have done to assess them globally, using the rating scales looked at
earlier here. A n im portant point to rem em ber is that if you have been doing a lot of
group and project w ork it w ill be necessary to com plem ent this w ith som e form al
assessm ent, so that you can check on individual perform ance.
Listening
W e norm ally develop listening skills in the classroom in lockstep fashion. This m eans
that all the students listen to one text at the sam e tim e. M ost text books or
coursebooks include listening passages for use in the classroom . Typical student
tasks involve filling in boxes w ith inform ation, ticking true or false options, ordering
events in chronological order and others designed to develop listening skills.
W e can inform ally assess studentslistening proficiency by getting an im pression of
w hat they have understood or by sim ply looking at the answ ers given by any one or
m ore student(s).W e can m onitor and assess studentslistening proficiency w hilst
m onitoring activities such as pair w ork activities or reactions to instructions from you.
O ne sim ple w ay of checking on answ ers m ay be asking for a show of hands to find
out how m any students, eg answ ered 10 correctly, 9 correctly, 7, 6, 5 and so on.
Perhaps there are students w ho m ay put up their hand at 8 w hen they actually
answ ered 3 item s correctly this can give you false results and m ay also im ply an
attitude problem on the part of one or m ore student(s). Extralinguistic clues can
often be a basis on w hich to assess a students listening proficiency, eg facial
gestures, especially eyes I could tell he didnt understand a word because he just
looked blank. O r the converse situation I could tell she understood everything
because she nodded in agreement at everything I said. O bviously, neither m ay be
an accurate assessm ent of these studentsproficiency in listening. In the first case
the student m ay have understood m ore than credited for but w as sim ply not
interested. In the second case the student m ay have understood nothing but w anted
to give the opposite im pression by nodding.
A n alternative w ould be to go through the answ ers one by one and ask the class
after each question w ho answ ered correctly. In this w ay, those w ho scored highly
are not being overtly praised and those w ho achieved low scores are not being
asked to adm it this in front of the rest of the class and are thus not placed in a
potentially threatening situation. You should find that this m ethod of eliciting scores
w ill provide m ore honest and open reporting from students as the elem ent of
com petition and perceived threat has been rem oved. You should also m ake it clear
to students w hy you are asking for inform ation. You can m onitor class responses to
each question and assess:
N the class as a w hole in relation to particular questions in the listening task
N any students w ho do not raise their hands very often and w ho have therefore
gained low scores in the task, thereby identifying quickly and efficiently problem
areas and students w ho m ay be having problem s w ith listening.
A nother w ay of assessing understanding of texts is to ask students to recycle w hat
they have heard. This can be in the form of recycling orally or in w ritten form . For
exam ple, follow ing a section of listening text, students can be asked to sum m arise
w hat they have heard. In large classes, this can be inefficient in term s of tim e and if
individual students are asked in front of the rest of the class it m ay have a negative
effect, by producing a threatening situation for the shy or less able student. A w ay
around this is to ask students to w ork in groups or pairs and to report to their
partners the im portant points of a text. A lternatively, another w ay of checking is to
refocus the question by saying that you w ant to know the results in order to
establish the difficulty of the task.
A further m ethod of assessm ent is to use a recorded text as the basis for a speaking
activity. For exam ple, a recorded text about the econom y of an island m ay be used
as a prologue to a speaking task in w hich students are asked to discuss the
problem s of the island and the solutions to them . D uring the speaking phase you
could go around the class and listen to the discussions, focusing on how pairs are
using the inform ation that they have been given in the listening text. A s w ell as
using listening texts you can assess students ability to understand you or other
students in the class. You can gauge ability to understand language by observing
student reaction to instructions. H ere you w ill be able to observe those students
w ho m ay be having problem s w ith listening sim ply because they are not able to
follow your instructions. A lso during speaking activities you can go around the class
and observe studentsability to understand each other.
You m ay w ant or need to m ake a judgem ent of listening ability. You can collect
m arks from listening activities that students have done and that you have taken in.
O ne point to bear in m ind here is that listening can be easily tested in this w ay and
that it m ay not be w orth doing the sam e things tw ice. W hen you com e to m ake an
overall judgem ent of listening ability, you can use band scales as for speaking and
w riting. O nce again, you need bands w hich reflect the kind of listening that you
have been doing. For exam ple, the band scale show n on page 17 focuses on aspects
such as sim ple/com plex m essages, different form s of text presentation (eg radio, TV,
song etc.) and differentiation betw een im plicit and explicit inform ation.
Example of band scales for listening ability:
5 C an understand com plex m essages
C an understand different oral presentations
C an distinguish betw een explicit and im plicit inform ation
4 Som etim es has difficulty w ith som e com plex m essages
C an understand m ost oral presentations
In general, can distinguish explicit and im plicit inform ation
3 H as difficulty w ith com plex m essages
H as difficulty w ith som e oral presentations
C annot, in general, distinguish betw een explicit and im plicit inform ation
2 C annot understand com plex m essages
H as difficulty w ith m ost types of oral presentation
C annot distinguish betw een explicit and im plicit inform ation
1 C annot understand sim ple m essages
C annot understand any type of oral presentation
Choose five or six students from your class.
During a listening activity in normal class hours, use the above rating
scale for listening to assess your students listening ability.
Reading
W e often do reading tasks in class in lockstep fashion: the w hole class reading one
text and answ ering questions on it. Typical exam ples are the skim m ing and scanning
activities so w idely used or the com prehension questionsat the end of a text.
There are various w ays of assessing this kind of reading in the classroom . The first is
by going around the class w hile students are doing a reading activity and observing
w hich students seem to be understanding it and w hich are having difficulties. The
im pressions that you get from this can be m isleading as som e students w ill be
reluctant to adm it difficulties. A nother w ay is by checking class understanding of
four or five points from a given reading passage. You can go through the answ ers
w ith the w hole class and afterw ards ask for a show of hands of w hat questions they
got right. A s m entioned before w ith listening, this is not a very reliable source of
inform ation. A n alternative is to collect in answ ers to reading texts and m ark them
yourself. Because this is som ething you can easily do as a test you m ay prefer to do
this kind of activity as part of form al rather than inform al assessm ent.
W hen assessing reading in this w ay a factor to bear in m ind is subject know ledge.
Som e students m ay know a lot about the subject of the text and thus have a clear
advantage over those w ho know little or nothing about it. For exam ple if you are
looking at a scientific text non-science based students are at a disadvantage,
w hereas the science students are considerably helped. In this situation your
judgem ent of these studentsability to read w ould probably not be very accurate.
A nsw ering a few questions about a text does not give reliable inform ation about
overall understanding of it. O ne solution to the problem is sim ilar to that identified
in the section on listening ie recycling. This m ay take the form of discussion of
reading passages or w riting about reading texts. For exam ple students can read a
letter and then w rite a reply to it or read a text and then roleplay a situation from it.
Jigsaw reading can be useful here, w hen you divide a text up and different groups
have to tell the others about their part of the text. O ne w ay of checking
understanding is by asking students to tell you about a text in their ow n language.
Students can also w ork out their ow n tasks on texts for other students to do.
By looking at understanding of specific texts w e are not really catering for students
w ho have different reading styles, speeds, interests and ability. In m ixed ability
classes it can be particularly interesting to assess individual reading. Som e schools
have a library and in these cases you m ay be able to use it for this kind of reading. If
you do not have one, a m ini-library could be set up in the classroom . There are
com m ercially available series of graded readers w hich give clear indications of the
num ber of w ords used and approxim ately how m any hours of English a student
should have studied for in order to be able to cope w ith the texts in the reader. If
you can set up an individual reading schem e you can m onitor the activities, talk to
individual students, helping out w here necessary and thus be able to m ake
judgem ents about reading ability for different individuals. O ne-to-one conversation
is a key concept here. You m ay w ish to discuss a text, reasons for its choice and its
difficulty level w ith individual students w hile the class as a w hole is perform ing a
task or com pleting a w orksheet. A s w ell as assessm ent through oral feedback, a
quicker w ay is to get students to w rite about w hat they have read.
Informal assessment Reading records
Look at the Reading Record Form below:
READING RECORD FORM
Name
Class
Title of book
Author
Summary of plot
Personal opinion
The reading record form on page 18 w ould allow you to:
N m onitor w hat your students are reading
N assess w hether they have understood the m ain points of the text they have read
N assess how students have reacted to the text by giving them the opportunity to
m ake a brief criticism
N inform ally assess their w ritten expression
Photocopy the Reading Record Form and try it out w ith a group of students.
Find out studentsreaction to the form . D oes the form provide you w ith the
inform ation you require? H ow w ould you use the inform ation you have collected?
W hat of the criteria for m aking overall judgem ents about reading ability? A s w ith all
other skills, the kind of judgem ent you m ake should reflect the kind of reading that
you have been doing. For exam ple, if students need to be able to read new spapers,
m agazines and inform al letters, then you should expose them to these types of texts
and assess their perform ance only on these texts. It w ould be unw ise, unfair and
unreliable to expose them only to m agazines and then expect them to m ake a
judgem ent about how w ell they read letters.
O ne sort of band you can use can be used in different situations as it does not
specify the kinds of texts. You can use the band together w ith your ow n syllabus
plan to assess students.
For example:
5 C an understand all necessary text types w ith no difficulty
4 H as m inor difficulty w ith different text types
3 H as considerable difficulty w ith different text types
2 C annot understand different text types
1 C annot understand any text type
A nother approach is to specify text types and use a Yes/No criteria.
For example:
1 C an understand factual articles in m agazines
YES
N O
2 C an understand inform al letters
YES
N O
Choose five or six students from your class.
During a reading activity in normal class hours, use the above rating
scale for reading to assess your students reading ability. If possible,
use both the holistic scale and the Yes/No scales. Which do you prefer?
Grammar and vocabulary
A s w ell as assessing studentsperform ance and progress in term s of the four skills, it
can also be useful to m easure their know ledge and ability to use specific structures
and vocabulary. This is an area w hich can probably be best assessed through
periodical short progress tests but inform al assessm ent can also be used.
H aving presented a structure it is im portant to assess learnersunderstanding of it
and ability to use it. This should be done at the practice stage of the lesson. D uring
controlled practice activities you can go around the class and identify problem s that
students are having and observe w hich students are having particular difficulties.
H aving done this you m ay need to w ork out extra practice activities for all or som e
of the students in the class. To avoid dividing the class and establishing a special
rem edial group you can do further activities in groups, w here the stronger students
can help the w eaker ones.
In addition to assessing studentsknow ledge and use of specific structures directly
after presentation (in the language practice stages of a lesson), you m ay find it
useful to assess studentsgrasp of structure and lexis in general w hile they are
doing free w riting and speaking activities. Rather than assess com m unication as a
w hole, w hich w e have looked at in previous sections, you can focus solely on
language. For exam ple, in a speaking activity you could go around the class and
w rite dow n the m ost im portant m istakes that the students are m aking. You can
then go through the m istakes w ith the w hole class, or alternatively plan rem edial
activities to deal w ith the problem s.
A s w ell as focusing on studentsuse of language you can also check their know ledge
of it in an inform al w ay. This is particularly useful before they have form al tests so
that you can give them extra practice if necessary and they get an idea of w hat
language they need to study m ore. O ne w ay of doing this is through class quizzes.
You can divide the class up into groups and organise a com petitive quiz betw een the
groups, m aking sure first that each group has a balance betw een m ore and less able
students. You can w ork out a series of questions to ask about both gram m ar and
lexis and you can run the quiz in an enjoyable w ay, giving students bonus points,
prizes etc. The sort of questions you can ask the groups are as follow s:
What is wrong with this sentence? My fathers are called John and Mary. Make a
sentence from these words. What does bicycle mean in Spanish, Italian etc?
In this inform al w ay you can get an idea of your learnersknow ledge and ability to
use specific structures and lexis. A ny problem s that you diagnose through this sort
of assessm ent can then be dealt w ith by rem edial presentation or further practice.
W hat you cannot get from the kind of inform al assessm ent of language suggested
here is an idea of individual studentsknow ledge and ability to use language. This is
best done through short language tests spread out over the course.
1.4 Inf ormal assessment of non-linguistic f actors
W e also need to think about our studentsoverall educational developm ent. It is
im portant for learners to develop in term s of a language and in term s of attitudes
tow ards learning, tow ards language, different cultures and other people. W e also
need to consider studentsability to take responsibility for and organise their ow n
learning. W e need to consider how m uch im portance w e are going to give to non-
linguistic factors. If w e discount non-linguistic factors and only assess language, w e
can find ourselves passing those students w ho have done no w ork and have even
been disruptive in class but w ho started w ith a higher level than m ost of the class.
The converse could also be true.
G iving non-linguistic factors greater im portance w ill encourage personal effort and
achievem ent im portant in a m ixed ability situation. H ow ever, w e could be open to
accusations that w e are failing to reflect our studentslinguistic perform ance and
that w e m ust assess students in relation to pre-established syllabus goals. The
balance is a difficult one and how w e solve it w ill depend on the particular teaching
situation that w e are in.
A ssessing non-linguistic areas can also be a very difficult area to deal w ith in term s of
reliability and fairness. If w e say that a student has a poor attitude w ithout having
concrete evidence for it, w e can lay ourselves open to accusations of partiality and
unfairness. W e need to be clear about the criteria w e are using and provide proof to
back up our decisions about non-linguistic factors.
W e need to carry out system atic observation and to keep records of it, (eg
attendance, participation in class etc). Studentsw ork, eg com positions vocabulary
books etc can be good indicators of attitudes and responsibility.
Informal assessment Non-linguistic factors
Either on your own or with a colleague, decide which of the items
below you think are important in terms of assessment. If you think that
an item should be assessed, how could you assess it? If not, why not?
a Attitude
is interested in class activities
is willing to offer opinions
is co-operative with teacher/peers
is willing to respond to the opinions of others
b Co-operativeness
is able to work in pairs
is able to work in groups
is able to work as a member of the whole class
is able to share ideas and knowledge
c Independence
is able to plan and organise own work
is able to self-correct where necessary
is able to use sources of information
d Creativity and presentation
shows original thought, initiative, inventiveness
presents work neatly and in an ordered manner
W e w ill introduce the factors first and then after initial discussion w e w ill suggest a
w ay of m apping the factors discussed onto a reporting system .
Attitude
W e can look at attitudes tow ards learning, attitudes tow ards the language itself,
attitudes tow ards other cultures and attitudes tow ards other people. The m ost
im portant one is the first one, as w ithout a positive attitude to learning, learning
itself is not very likely to take place.
O ne w ay of assessing our studentsattitude is by w riting profiles.
Look at the tw o student profiles. W hich one is the best student?
Student A
This student appears in the classroom to be a passive learner. H e offers no
opinions nor adds anything to any language discussion. Student A appears to be
uninterested in everything that is going on around him . H e seem s bored. H e does
not co-operate w ith other m em bers of the class. If involved in tw o-w ay
com m unication, he interrupts his com panions constantly and often dism isses their
opinions, insisting that his opinions are right.
Student B
This student appears alw ays to be actively involved in class activities. She alw ays
offers an opinion and enriches discussions w ith her view s. Student B alw ays
appears to be interested in w hat is happening in the classroom . She appears to be
stim ulated by class activities. She co-operates w ith you the teacher and the rest of
the class. She is a good listener and appears to accept her com panions
view points w hilst at the sam e tim e giving her ow n at the appropriate m om ent in
a conversation.
A judgem ent or rating could be m ade each m onth or term or academ ic year.
Student A and B profiles identified the follow ing areas as possible criteria for
assessm ent:
N is passive/active
N offers opinions/does not offer opinions
N show s interest/does not show interest
N co-operates/does not co-operate
N accepts opinions/does not accept opinions
O ther areas w e could consider in term s of attitude are effort and interest. H ere w e
can look at concrete indicators of effort and interest, eg w hether a student hands in
hom ew ork regularly w hen asked, w hether a student is punctual, w hether his/her
attendance is regular and w hether he/she reads in English outside the class.
Group work
In language classroom s the ability to w ork w ith other people in a group is crucial. It
should also be one of our general educational objectives to develop the ability to
co-operate w ith other people. A ssessm ent of group w ork can be carried out in the
sam e w ay as that of attitude and the follow ing areas as assessm ent can be
identified:
N has difficulty/does not have difficulty in groups
N co-operates/does not co-operate w ith group m em bers
N is able to w ork in groups/only able to w ork alone
N accepts/rejects group organisation
N accepts/rejects the w ork of others
W e also need to consider w hen to observe and collect our data project w ork, role-
plays and gam es all provide us w ith opportunities to observe and assess. Perhaps the
m ost useful of these is project w ork w hen students need to w ork together to produce
som ething. C o-operation w ill thus involve m aking decisions in groups, sharing out
w ork, helping each other and positively criticising the w ork of others. A s w e w ill
m ention in the chapter on self-assessm ent, peer assessm ent can be a very good
source of inform ation about this. W hile you can observe students w orking together in
the classroom you w ill not be aw are of the dynam ics of each group in the class.
Organisation of work
This is the ability of students to organise their ow n w ork, a very im portant learning
skill w hich w e need to help our students develop. A ssessm ent can be carried out in
the sam e w ay as the previous tw o factors and the follow ing areas as assessm ent
criteria can be identified:
N is able/unable to organise w ork system atically
N is able/unable to plan w ork
N is able/unable to produce presentable w ork
N is conscientious/not conscientious
N self-corrects/does not self-correct
O ther things w hich w e could take into consideration w hen looking at students
ability to organise their ow n w ork are the follow ing: their notebook, their gram m ar
notes and their vocabulary books. A ll of these are im portant indicators and one w ay
of helping you assess this area could be to take them in and look at them . It is fairly
easy to see their ability to organise their w ork from the w ay students w rite their
notes and store lexis and gram m ar. It is w orth pointing out how ever that if you do
this you should give them som e help to start w ith. Learner training activities at the
beginning of the year should include looking at w ays of organising learning such as
establishing vocabulary books.
Independence
This is the ability of students to w ork on their ow n as w ell as part of a group (see
under G roup W ork). Students should be encouraged and trained to w ork alone using
resources available to them such as dictionaries, text books and other reference
w orks. The follow ing areas as assessm ent criteria can be identified:
N is able/unable to use dictionaries
N is able/unable to use the course text book as a resource
N is able/unable to refer back to previous w ork as an aid
N is able/unable to produce pieces of w ork w ithout help
A ll the factors m entioned need to be m apped onto a rating scale if they are to be
assessed w ith any degree of reliability. O n the follow ing page is a suggested m odel,
taking as its criteria a YES/NO response to the areas identified in each non-linguistic
factor above. The different criteria w ould be best separated out into four different
scales w ith a decision m ade for each item .
A ttitude Is an active m em ber of the class. Yes/N o
Is able and w illing to offer ow n opinions on subjects. Yes/N o
Show s interest in class activities. Yes/N o
Is generally co-operative w ith other people in class. Yes/N o
Is capable of m odifying opinions in the light of those
of others Yes/N o
Use the criteria suggested about group work, organisation of work and
independence to write your own scales like those above for attitude.
1.5 Results f rom inf ormal assessment
O nce you have im plem ented your inform al assessm ent program m e you w ill have
results in the form of inform ation about your students. This inform ation m ay be a
set of num bers such as ratings for speaking, reading, listening and w riting or a set
of criteria w hich the student can or cannot m eet such as w orking in groups, ability
to produce w ork independently etc. In order to use these results, they w ill need to
be recorded against each students nam e, perhaps in a register or other notebook
used for this purpose and should be accessible to your students, your institution and
yourself.
A s inform al assessm ent form s just one part of the overall assessm ent program m e,
results could be periodically com pared to those obtained from form al and self-
assessm ent. If results from all three form s of assessm ent correlate highly, ie in
general each part seem s to give sim ilar results, you m ight assum e that a fairly
reliable picture of a students ability is being built up and that any decisions you
m ake about that student w ill be based on solid foundations. If a particular students
results do not correlate, eg his/her results on inform al assessm ent are greatly
different to results on form al assessm ent, you w ill probably w ant to find out the
reasons. It m ay be that this student is nervous w hen taking form al tests. If this
pattern is repeated across all studentsresults, it m ay be that you need to take a
look at the validity and reliability of your form al tests. O f course, if results from tests
seem m ore consistent than those from your inform al testing, you w ill need to think
again about how you are assessing inform ally.
You m ay find that results from inform al assessm ent of non-linguistic factors such as
attitude or group w ork give you im portant inform ation about a student or students.
There m ay be an attitude problem against w hich m easures need to be taken in your
classroom teaching m ethods. In any case, inform ation about students should be
looked at in each individual case and should at all tim es feed back into the teaching
and learning process.
A s teachers involved in language learning, w e are prim arily concerned w ith linguistic
aspects of our studentsperform ance and com petence. H ow ever, w e should not lose
sight of the fact that other non-linguistic factors can and do affect the strictly
linguistic in a positive or negative m anner. O ur job as teachers m ust be to assess and
through our assessm ent m axim ise the learning potential of our students in all
aspects of their learning experience. U nderestim ating the extra-linguistic factors
w ould be to ignore an im portant aspect of that experience. A ssessing these non-
linguistic factors and discussing these w ith your students should heighten their
aw areness of the im portance of such factors and should thus aid the m axim isation
of that potential. It should not be forgotten that w e are not solely concerned w ith
language per se but in education in the broadest sense of the w ord.
Action points: inf ormal assessment
1 A s a result of reading this chapter are you going to m ake any changes in the
w ays you inform ally assess your students? If so, w hat are they?
2 H ow are you going to w eight different areas?
(For example, listening 25% , reading 20% etc.)
3 H ow are you going to inform ally assess your studentsreceptive
perform ance ie listening and reading?
N by giving an im pression m ark
N by collecting in m arks from tasks
N by using a rating scale
4 H ow are you going to inform ally assess your studentsproductive
perform ance ie speaking and w riting?
N by giving an im pression m ark
N by collecting and adding together a series of im pression m arks
N by collecting and adding together a series of m arks based on rating scales
N by using rating scales to assess overall perform ance
5 A re you going to take into consideration non-linguistic factors? W hy/w hy
not?
6 If yes to the previous question, how are you going to assess non-linguistic
factors?
N by observing students in class and giving an im pression m ark
N by observing students in class and rating them using scales
N by collecting in vocabulary/notebooks and m arking them
N by using peer assessm ent of group w ork
Formal assessment
2.1 Why test?
In this chapter w e shall look at the reasons for testing our students and w e shall
exam ine som e of the possible test types w hich w e could use. The list of types of test
is not intended to be exhaustive, but intends to give you a breakdow n of the m ain
types of test com m only used, their advantages and disadvantages and the contexts
in w hich they are best put to use. The type of test w e use w ill depend on the
situation, the need and the purpose of testing.
Objections
Som e com m on argum ents against classroom testing (as opposed to inform al
assessm ent) are the follow ing:
a Testing takes up valuable time which I could better use for teaching.
b I know how my students are doing anyway. Why do I need to test them as well?
c Some of the students in my class, who have worked hard all year, fail tests.
Others who have done less work pass them, just because they have been to an
English speaking country or because they have private classes.
d Tests do not motivate my students at all. The good ones feel complacent and the
bad ones just give up.
Rationale
H ere are som e answ ers to the objections raised in a to d above:
a Testing takes up tim e, but it should be seen as part of the teaching/learning
process rather than som ething divorced from it.
b This inform ation is not alw ays accurate, especially in the case of large classes.
W hen students are w orking together in the classroom situation, individual
w eaknesses can be obscured. Tests therefore enable us to m easure progress in a
m ore individualised w ay.
c Tests are only part of the assessm ent process and should be seen as a
com plem ent to, eg self and inform al assessm ent.
d M otivation w ill depend on how w e use the results of the test. If tests are seen as
com petitive, they m ay dem otivate students If they are seen as opportunities for
students to find out how they are progressing, tests can m otivate students to
think about their problem s and do som ething about them . Thus it is im portant for
tests to be linked to self-assessm ent so they can add to other areas of form ative
assessm ent, assessm ent w hich helps our learners to develop.
Reasons f or testing
Let us look in detail at w hy w e should test our students. Firstly, w henever a test is
adm inistered, there is a decision to be m ade:
N W e m ay w ant to find out about a candidates suitability to follow a course of
study, although this is not the case in state education.
N W e m ay need to find out how a student is progressing during a course of study
and possibly identify problem areas before a course ends.
N W e m ay w ant to com pare a students perform ance w ith that of other students.
N W e m ay w ant to find out how m uch a student has learned during the course or
academ ic year ie com pare w hat students can do at the end of the course
com pared w ith the beginning of the course.
2
C HA P T E R 2 Form al assessm ent
In all these cases, decisions are m ade about the student. If no decision is to be m ade
as a result of the test, there seem s little point in adm inistering the test. If you have a
clear idea of the kind of decision w hich needs to be m ade about your learners, then
you w ill be closer to identifying the m ost appropriate kind of test for you.
A com m on m isconception, held by students and teachers alike, is that a test is
som ething w hich is done to or at people rather than som ething w hich is done by
them and for them . W e should view the decisions w hich are m ade about students as
decisions w hich are taken for them . Form al testing should be seen as a com plem ent
to other form s of assessm ent, eg self-assessm ent and inform al assessm ent. The
basic differences are that if w e have a w ell designed, reliable and valid test, then the
test w ill m easure studentsability in a m ore objective w ay than m ore subjective
form s of assessm ent such as inform al observation and self-assessm ent. This is not to
say that all form al testing is objective. N or should w e take the view that subjective is
necessarily bad and that objective is necessarily good. For exam ple all oral tests are
subjectively m arked and all m ultiple choice type tests are objectively m arked.
The influence of tests on teaching and learning is called the washback effect. If your
students have to do a test or m aybe a public exam ination at the end of the course
this w ill affect the syllabus. If w e have a good test, this should affect teaching in a
positive m anner. If w e have a bad test, this m ight affect teaching in a negative
m anner. W hat is a good test and w hat is a bad test? A test can have a positive
influence if it contains authentic, real-life exam ples of the type of tasks w hich your
learners w ill need to perform in the future. Tests can have a negative influence if
they contain artificial tasks not linked to real future needs. Teaching m ethods w ill
probably reflect these tasks and the learning process could end up revolving around
w hat w e m ight term exam practice. Your ow n tests w ill also have an effect on your
studentslearning. If you test m ainly gram m ar, your students w ill assum e that this is
the m ost im portant thing to learn and m ay m ake less effort during other m ore
com m unicative activities.
Washback effect
Match these examples of assessment with their washback effects:
1 A progress test which concentrates
on one out of four chapters
covered in the textbook.
2 An end of year test which
concentrates on grammar and
vocabulary, even though you have
done lots of speaking and listening.
3 A diagnostic test which has shown
that your class is very weak on
speaking skills.
4 An end of year test of reading,
listening and speaking, covering
a wide range of material.
Check your answers on page 61
Bad washback. Students will
feel that luck was more
important than hard work.
Bad washback. Students will
feel that classwork has been a
waste of time.
Good washback. Students will
feel they have been tested fairly.
Good washback. The
information will help you plan
your course to suit the needs of
your students.
W hen w e are testing w e also need to ask ourselves w ho the test is for and w ho w ill
look at the results. W e therefore need to think about w ho w e are accountable to.
Firstly w e m ust think about our learners w ho need to receive feedback about
progress and perform ance. They w ill com pare the test and the test m ethods to the
teaching m ethods you have used. Therefore it is im portant for us to tell them how
w e are going to test them at the beginning of a course and perhaps tell them w hy
w e are going to test them like this.
In m ost schools parents w ill use test results as a w ay of determ ining students
progress, interest and application to the course of study. Parents w ill also consider
results, especially those from public exam inations, as evidence of the effectiveness
of the teacher, teaching m ethods and school. Test results should therefore be easily
accessible and com prehensible to parents of students. W e should be able to explain
to parents how w e got the results and our rationale for testing in this w ay.
Types of test
Before, during and at the end of courses, different types of test are adm inistered.
The key question is the purpose of the test. D ifferent tests are adm inistered for
different purposes and are used at different stages of the course as a m eans of
gathering inform ation about students. The m ain types of test are discussed below
and the pros and cons of each type are described. H ow ever, you, as language
teachers, should decide on w hat the best option is in each instance for your
particular group(s) of learners and your particular teaching situation.
Progress tests
This type of test is adm inistered during courses. Progress tests m ay be adm inistered
after certain blocks of study, eg after x num ber of units, at the end of each w eek,
each term etc. The test aim s to find out inform ation about how w ell classes as a
w hole and individual students have grasped the learning objectives, how w ell the
course content is functioning w ithin the specified aim s and objectives and future
course design. Teachers can easily identify how w ell students are progressing in a
very short period of tim e, eg a progress test of half an hour can give a great deal of
inform ation about the class if the test is w ell designed. Progress tests can provide a
great deal of inform ation if the test sam ples w idely from the course content.
Progress tests can perform a very im portant form ative function in that they do not
only give inform ation to the teacher but can provide im portant feedback to the
student. W hen linked w ith self-assessm ent, feedback can help learners to identify
their ow n problem s and to set their ow n goals for the future.
Summative tests
These tests are adm inistered at the end of courses and their objective is to see if
students have achieved the objectives set out in the syllabus. W hile these tests can
be used effectively as a w ay of deciding w hether students m ove on to a higher level,
this can be done better at secondary school level by progress tests throughout the
year in com bination w ith inform al assessm ent by the teacher. A lot of inform ation
gained from a sum m ative test is often w asted because it does not feed back into
the learning process as does form ative assessm ent. A lso, end of year tests can put a
lot of stress on both teachers and students.
Entry/placement tests
This type of test w ill indicate at w hich level a learner w ill learn m ost effectively in a
situation w here there are different levels or stream s. The aim is to produce groups
w hich are hom ogeneous in level that w ill use institutional and teacher tim e m ost
effectively. The larger the groups of learners to be designed, the m ore
hom ogeneous the groups w ill need to be and therefore the m ore reliable the entry
or placem ent test w ill need to be. This type of test is less useful w here students are
grouped alphabetically or according to age rather than ability. Entry or placem ent
tests are not very com m on in state run institutions.
Diagnostic test
A s the nam e suggests, this type of test is used to find out problem areas. W here
other types of tests are based on success, diagnostic tests are based on failure. W e
w ant to know in w hich areas a student or group of students are having problem s,
w hich parts of a course or learning objectives those students cannot cope w ith. O ne
w ay of looking at this type of test is to consider it as a technique based on eliciting
errors rather than correct answ ers or language.
D iagnostic inform ation is vital for teachers in order to design further course
activities and w ork out rem edial activities. The inform ation can also be useful for
learners, as they can analyze their ow n problem s. D iagnostic testing is present in
m any progress tests for the sim ple reason that the progress tests identify problem
areas. H ow ever, a reliable diagnostic test is difficult to design.
Prof iciency tests
This type of test aim s to describe w hat students are capable of doing in a foreign
language and are usually set by external bodies such as exam ination boards.
Proficiency tests enable students to have som e proof of their ability in a language.
They also provide potential em ployers w ith som e guarantee of proficiency in a
language because exam ination boards are seen as bodies w hich set standards in an
im partial w ay and boardsexam inations are generally considered to be reliable and
valid. Som e proficiency tests, w hile claim ing to be com m unicative, often have a
large language com ponent such as gram m ar or vocabulary. This can have a negative
w ashback effect on teaching in term s of exam ination preparation.
Your testing
Which of these questions do you need to answer about your students?
What levels they should go into?
How are they getting on at the moment?
What are their problems and weaknesses?
How much have they learnt over the course?
What can my students do in English?
What types of test do you have to do? Match these types with the
questions:
summative / diagnostic / proficiency / placement / progress
Check your answers on page 61
2.2 Planning assessment programmes
H aving taken the decision to assess students form ally, it is vital to do it system atically.
It is not enough to test students in a piecem eal w ay, as the need arises. W e need to
plan a clear assessm ent program m e at the beginning of the academ ic year. First, w e
need to think about our syllabus objectives and w ork out a program m e, thinking
about w hat w e are going to test and w hen. Then w e need to decide how w e are
going to test and choose the m ost suitable test form ats for our purpose. H aving done
this w e can actually start to w rite tests and adm inister them . Finally, w e need to think
about the results and use them to com e to conclusions about our learnersprogress.
The stages in progress testing
1 Planning syllabus objectives
Assessment programme
weighting (skills/language) timetable for assessment
2 Construction test formats
Test construction
writing/editing tests
3 Administration finished test
Administration
test conditions answer sheets marking schemes/criteria
4 Results students scores
Conclusions
students progress information for evaluation of course
When are we going to test?
Traditionally, exam s are at the end of term or at the end of year. Students are under
great pressure and often do not do their best; teachers are snow ed under w ith
preparation of exam s and then vast am ounts of m arking. The crucial decisions of
pass and fail are often taken during, or in the afterm ath of this period. The end of
term exam thus has several m ajor disadvantages:
N Students and teachers are under stress. For the teacher a large am ount of w ork is
concentrated into a short period of tim e. For the student, if he or she has a bad
day, the w ork done throughout the year is questioned. Thus the end of term
exam is often not a representative sam ple of studentsperform ance over the
course. U sed as the only instrum ent it can result in unreliable assessm ent.
N The diagnostic inform ation from the exam s is usually discarded. W hat w e learn
about studentsw eaknesses and problem s com es very late and is often forgotten
in the end of term rush. A t the start of the follow ing term this inform ation is
hardly ever checked up on and used by the next teacher of the class.
N Students them selves are m uch too w orried about their results to think about the
difficulties they had, or reflect on the progress they m ade.
N In the eyes of the students, assessm ent is divorced from learning. Exam
technique, luck on the day and the ability to w rite quickly are qualities prized
m ore than hard w ork, interest and enthusiasm .
N The period after exam s is a kind of academ ic no-m ans-land betw een school and
the holidays; the teacher has to occupy large groups of distracted and potentially
disruptive adolescents w hose m otivation to study is low .
To overcom e som e of the problem s related to end of term exam s it is possible to
balance your form al assessm ent w ith the inform al assessm ent you have been doing
throughout the term . This helps to take into account the fact that students could
have had a bad day w hen they did the test. You can also m ake sure that you go
through the exam s thoroughly w ith your students, m aking sure that they receive
adequate feedback about their problem s. Students can reflect on the problem s they
had and you can set rem edial w ork for them (perhaps to do over the holidays). It is
also w orth considering an alternative to the end of term exam : a series of short
assessm ent tasksto take place throughout the term . The tim e spent assessing
form ally is the sam e, but this tim e is a great deal m ore productive.
H ere is an exam ple of an assessm ent task program m e:
Assessment tasks: Autumn term
Week 1: Wri ti ng task (20 mi n) To find out w hat students have rem em bered after
the holidays or find out about new students and their w eaknesses.
Week 3: Li steni ng task (15 mi n) Som ething fairly easy to build confidence. A
listening task w ell w ithin studentsreach, (eg short dialogue).
Week 4: Readi ng task A text related to one of the topic areas being looked at,
(eg picture story/cartoon).
Week 6: Language assessment task To check on gram m ar and vocabulary
studied so far.
Week 8: Wri ti ng task To check one of the w riting areas looked at so far, (eg
inform al letters).
Week 9: Li steni ng task M ore challenging task and text than the first one, (eg
story/longer dialogue).
Week 11: Readi ng task M ore challenging task and text than the first one, (eg
extract from childrens non-fiction).
Week 12: Language assessment task To check on the gram m ar and vocabulary
studied since the previous language task.
Total ti me spent 150 mi nutes
This system has considerable advantages over the traditional end of term exam .
Firstly, there is less stress for both students and teachers. If a student has a bad day,
he/she has a chance to do better another day. Thus, the assessm ent tasksystem
should provide a m uch m ore accurate picture of studentsability. Secondly, the w ork
involved in preparing tests and correcting them is spread over the term , instead of
being concentrated into one hectic period. Reliability in m arking should be
correspondingly enhanced. Thirdly, the assessm ent tasks provide vital diagnostic
data for the teacher, w hich can then feed back into the course. In addition, students
get very useful inform ation about how they are doing throughout the course.
A ssessm ent tasks avoid the divorce betw een testing and teaching that end of term
exam s often produce. A s w ell as taking place during the learning period, tasks can
be related to the topic areas being studied. A t the sam e tim e as m easuring progress,
tasks can be interesting and fun. If the them atic area being covered is A nim als, a
reading task on polar bears could extend studentsknow ledge. Finally, w ith a system
of assessm ent tasks, there is no post-exam period. It is how ever an idea to give a
task right near the end w hich students have to do, to avoid students m issing classes
to study for other exam s, or discuss w ith teachers of other subjects the idea of
giving up end of term exam s com pletely.
Planning an assessment programme
Look at the example of the assessment task programme on page 31.
Choose one of your classes and plan your own programme of
assessment tasks for a term. Write out the number of weeks in the
term and then space formal assessment tasks as you see fit.
What are we going to test?
A n im portant thing is to decide on w eighting betw een different elem ents in the
course. This w ill depend on the kind of class you have and the syllabuses you are
follow ing. If you pay a lot of attention to w riting and gram m ar, your form al
assessm ent should reflect this and you should test in the sam e w ay. If, on the other
hand, you concentrate largely on speaking and listening, you should test m ainly these.
W hile this m ay seem obvious, it is surprising how often com m unicativeclasses
have tests w hich are gram m ar-based. This has a very negative w ashback effect on
students. They quite naturally com e to feel that w hile speaking and listening are
good fun, w hat really m atters is gram m ar.
W hen planning form al assessm ent at classroom level it is useful to m ake a clear
breakdow n of w hat you are going to test and how m uch w eight each area carries.
This breakdow n should clearly reflect the overall syllabus balance of your course.
For example:
reading
35%
writing
40%
grammar
25%
speaking
20%
reading
20%
Listening
20%
language
20%
writing
20%
Assessment weighting
Look at the example of weighting on page 32. Draw a pie graph
illustrating how you weight your formal assessment for one of your
classes. Does it reflect your syllabus objectives? If it does not, change
the weighting for the next time that you organise your programme.
H aving decided on w eighting, w e need to establish priorities. W e cannot test
everything that students have done throughout the course. W e m ust therefore look
at our syllabuses and choose a sam ple of areas to assess form ally.
Syllabus priorities
Look at the example of syllabus priorities below, for 13/14 year old
post-elementary students. Then list your own priorities for your
classes. Think about the following areas:
speaking writing
listening grammar
reading vocabulary
Syllabus objectives
Speaking general fluency/ability to talk about own life
(likes/dislikes/homes/food/past)
Listening extensive listening (listening for gist and for specific
information) stories/dialogues/radio programmes
Reading extensive reading skimming/scanning/dictionary use
magazines/comics/childrens fiction/non-fiction
Writing writing about own lives letters/stories/postcards
Grammar revision and introduction of: present simple /continuous,
past simple, present perfect, future: going to, countables
Lexis lexical areas: classroom/animals/homes/travel
2.3 Choosing test f ormats
H aving w orked out a general assessm ent program m e, the next step is to decide how
w e are going to test. A t this stage it is im portant to refer once m ore to som e of the
fundam ental concepts in assessm ent.
N Firstly, w e w ant our assessm ent tasks to be reliable. This m eans that they should
be consistent m easures of studentsperform ance. If the sam e test w ere given
under the sam e conditions on a different day, the results should be the sam e.
N Secondly, w e m ust ensure that our assessm ent tasks have validity, they test w hat
w e w ant them to, rather than testing som ething else. For exam ple, a reading test
w here answ ers are m arked for gram m atical accuracy is not only a test of reading.
It is a test of w riting and gram m ar as w ell.
N Thirdly, our assessm ent tasks need to be practical. They m ust not take too long to
do or m ark, be too difficult to organise or involve equipm ent and resources that
w e do not possess.
N Lastly, it should be clear to our students that our tests reflect w hat w e are doing
in class, thus hopefully causing a positive washback effect on learning. If tasks
assess com m unicative ability, our students w ill be encouraged to take part in
com m unicative activities. If they test gram m ar, our students w ill see skills w ork as
peripheral to the m ain job of learning gram m ar.
In an ideal w orld, our assessm ent tasks w ould be extrem ely practical and reliable,
w ith high validity and a beneficial washback effect. In practice how ever, this is not
so easy to achieve. If w e w ant high reliability, discrete itemtest form ats are the m ost
suitable. These are form ats w ith m any item s requiring short answ ers, (eg m ultiple
choice questions). Because there are m ore item s, w e have m ore inform ation available
to us and reliability is increased. These test form ats are also very practical, usually
being quick and easy to m ark. U nfortunately, such form ats can have low validity,
because doing a test such as a m ultiple choice is not a test of real com m unication.
Such discrete item tasks can also have a negative washback effect, w ith the test-
w ise student often doing best. O n the other hand, if w e w ant high validity,
integrative and open-ended test form ats can be better. Integrative form ats involve
com m unication and interaction (for exam ple reading a letter and replying to it).
O pen-ended form ats are those w here responses are open, (eg w ritten com positions
such as stories).
Both of these form ats can have beneficial washback effects, as students becom e
aw are of the im portance of com m unication. H ow ever, it can be difficult to interpret
results from such open-ended tests. Firstly, it can be hard to decide w hy students had
problem s (w as it the reading or the w riting?). Secondly, these tests can be im practical,
taking a long tim e to m ark and adm inister. Thirdly, they can be unreliable, as they are
m uch m ore difficult to m ark consistently than discrete item tests.Thus, w e have a
certain dilem m a as to how w e are going to test.
1 Discrete item formats
2 Integrative/open-ended formats
A s both discrete item and integrative test form ats have their advantages and
disadvantages, perhaps the solution is to m ix the kinds of test form ats w e use. W e
could em ploy som e integrative tasks, especially for productive skills like w riting and
speaking. For the receptive skills and for testing language, w e could use som e
discrete item tasks. O ur discrete item form ats m ake it easier for us to assess w ith
greater reliability and w ith less tim e for m arking. O ur m ore open-ended form ats
enable us to give our students tests of real com m unication. H ow ever, w e need to
look in m ore detail at w hat kind of test form at best fits our specific requirem ents.
Reviewing test formats
1 Read through the list of test formats. List those that you use.
2 Look through it again and make a list of other activities that you
use in class and that might be interesting for you to use in your tests
in the future. Ask these questions:
a Will it help me test reliably what I want to test?
b Have I got enough time and resources to use it?
c If I use it will it have a good effect on my students and reinforce my
syllabus objectives?
HIGH RELIABILITY
PRACTICAL
LOW VALIDITY
NEGATIVE
WASHBACK
HIGH VALIDITY
POSITIVE
WASHBACK
LOWER RELIABILITY
LESS
PRACTICAL
Survey of test f ormats
A bbreviations:
IQ intelligence quotient: a com m only used m easurem ent of intelligence
ss students info inform ation dk dont know
Reading and listening
TEST FORMAT PROS CONS
Short answ er questions
eg H ow tall are elephants?
Three m etres.
Table com pletion
eg C om plete the table w ith
inform ation.(age / fam ily etc).
D iagram s/m aps/pictures
eg Label the places on the m ap
Listing
eg List the kinds of food
m entioned in the text.
True, false, dont know
eg M ark the sentences t/f or dk:
Lions are cats.
M ultiple choice
eg C hoose the correct answ er:
John goes out.- a som etim es
b rarely c never.
Sequencing (texts/ pictures)
eg Listen and put the
paragraphs in order.
Text C om pletion
eg Listen and com plete the
inform ation about the film .
Problem solving
eg From the follow ing info
w ork out the peoples nam es
W ord attack
eg W ork out the m eaning of
these w ords from the text.
Identify topic (text/paragraph)
eg M atch the title w ith the
text.
Linking
eg W hat does the underlined
w ord refer to? It arrived late.
Identify linking w ords in a text
eg after/next etc
D iscrepancies
eg Read the text then listen
and list the differences.
Easy to write and mark. Very
good for checking gist or
intensive understanding of
texts.
Easy to construct and mark.
Good for checking specific info
or data form a text.
Quite realistic, motivating tasks.
Good for checking specific info.
Realistic. Easy to write and
mark. Better for listening than
reading.
Easy to write. Quite realistic.
Tests gist or intensive
understanding well.
Some writing. Important not to
mark for accuracy and to decide
what answers are correct. Need
to ensure qs do not test ss
knowledge of the world.
Some writing. Need to decide
on what answers accepted as
correct.
Can be difficult to draw
pictures. Could involve non-
linguistic skills (eg map reading).
Tests recognition of words.
Does not test understanding or
meaning.
High guessing element:
50% for t/f, 33% for t/f/dk.
Very easy to mark thus good for
very large classes. Good for
checking gist or intensive
understanding.
Easy to construct. Good for
stories (listening) and for linking
of discourse.
Quite realistic. Good for
listening for specific info.
Realistic and good fun. Tests
global understanding.
Tests ability to infer meaning
from context (for reading).
Good for gist reading. Easy to
construct and mark.
Good for testing intensive
understanding / linking within
text. (cohesion)
Good for working out how a
text holds together. (cohesion)
Realistic, integrative test of
both reading and listening.
Very difficult to construct.
Wrong options (distracters) can
distract better students.
Guessing element.
Very difficult to mark. If one
answer is wrong, others are
too. Impractical unless marking
scheme is adapted.
For reading it tests knowledge
of language. (see close and
gap-fill)
Can test general IQ not ability
to read or listen.
NOT suitable for listening (when
it can work as a vocabulary
test).
Need to think of suitable
answers when questions are
open (eg giving titles to texts).
Need to underline words and
give line numbers to make task
easier to do and mark.
Can be more a test of word
recognition than of
understanding.
Difficult to see if problems are
due to listening or reading.
Writing
Essay questions
eg W rite about a day w hen
everything w ent w rong.
G uided w riting
U sing pictures, notes, diagram s
(giving ss som e input of info)
Punctuation
(punctuating texts)
Sum m ary
eg Read the text and sum m arise
it in 20 lines.
N ote taking
eg Read the text and w rite
notes.
D ictation
eg Listen and w rite dow n the
text.
C om bined
eg Read the letter and w rite a
reply.
Speaking
Free interview s (chat to ss in
groups or as individuals)
Picture description (using photo
or draw ing)
Inform ation transfer
(inform ation gap through notes
or pictures)
Roleplay
Ss assum e roles (w ith or w ithout
cued info)
O ral presentations
Ss prepare and give short talks.
Very easy to set. Better for
higher levels.
More realistic than essays,
because input can create reason
for communication. Gives
students help, thus good for
lower levels. Easier to mark
than free writing.
Good for testing specific
knowledge of punctuation.
A realistic, integrative task.
Tests both reading and writing.
A realistic and useful task for ss
to practise.
Realistic. A good integrative
test of listening and writing
(spelling).
Realistic and very good for
writing.
Very unrealistic and often tests
imagination or content
knowledge. Difficult and time
consuming to mark.
Input can be a test of reading
if ss do not understand input
this will affect their writing.
Restricted and not easy to mark
punctuation can be subjective
(related to style).
Difficult to mark. (What is
important in a text can be
subjective)
Can be difficult to mark unless
it is guided.
Very difficult to mark, unless a
very clear scheme is established.
Difficult to mark-to distinguish
between problems in reading or
writing
Realistic and can reduce stress
for ss.
Gives tester time to listen and
ss something concrete.
Realistic need for
communication. Tests key
interactive strategies.
Excellent for testing interaction
and commonly used task in
most materials.
Realistic and gives the tester
time to assess performance.
Very difficult to rate
performance (personality factor
shy /outgoing). Need to
maintain conversation at same
time as rating.
Artificial task and there is no
interaction.
Can be problems when one
student is a lot weaker than the
other. (doesnt work)
Can test the ability to act.
No interaction and can have a
high stress factor not suitable
for younger ss.
Language
C loze Blank out every nth w ord
(5th/6th) in a reading text.
Ss com plete the blanks
G ap-fill Selected w ords in a text
are blanked out. Ss have to fill
in the blanks.
W ord sequencing, eg O rder
the w ords below to m ake
sentences.
Editing Ss correct m istakes.
C ued sentences (from pictures
or w ords)
Personalisation, eg W rite five
sentences about your fam ily.
W riting short texts
eg W rite a dialogue w ith these
w ords.
Identifying structures
eg tenses/parts of speech
Translation
eg Translate these sentences
into English.
Table com pletion
eg C om plete the table w ith
these adjectives.
Sentence transform ation
eg Finish the sentences so that
they m ean the sam e.
Lexis classification
eg M atch the w ords w ith the
topics.
M atching sentence halves
(one half of a sentence w ith
the other)
M atching w ords / definitions
Identifying test formats
Look at the examples of the assessment tasks on the next few pages.
Identify the formats used in the tasks. Check your answers on page
62.
Example Task 1 Combined (reading/writing)
Easy to construct (a text and a
bottle of liquid paper).
Good for testing
different structures, provides
clear contexts. Easy to write
and mark.
Excellent for testing structures
with problems of word order
(eg-questions).
Realistic. Good for testing L1
interference.
Easy to write and mark.
Easy to write. Good for some
functions (eg likes /dislikes)
Tests usage of structures
beyond sentence level.
Tests knowledge of grammatical
system and of metalanguage
Tests problems of form and
usage caused by L1
interference.
Good for testing knowledge of
irregulars and word-building
Good for testing some
structures (eg-passives
/conditionals)
Good for testing lexical sets.
Good for testing some
structures (eg conditionals /
gerunds and infinitives).
Good for specific words and
link with dictionary skills.
Can test IQ more than language
(puzzle element).
Can be irritating for ss.
Need to think about possible
answers.
Can be difficult to mark (some
mistakes worse than others).
Contains puzzle element.
Can be too difficult if mistakes
are not signalled.
Can be mechanical repeated
use of same structure.
Can be mechanical.
Can be difficult to mark (open-
ended). Need for clear criteria.
Does not test usage of
structures.
Many ss and teachers feel that
the target language only should
be used.
Mechanical does not test
usage.
Very artificial and can test IQ as
well as language.
Does not test usage of
vocabulary.
Need numbers and letters to
make it easier to mark.Eg 2 c
Does not involve usage of
vocabulary.
Examples of test f ormats
Assessment task 1 Writing (elementary)
Read this letter from a penfriend and reply to it. Include this
information about yourself: Name Age Family Where you live
Hobbies Favourite sports Music
15 Newmarket Rd,
Scunthorpe,
Great Britain.
10th June.
Dear......,
Hi! My names Rodney and Im your new penfriend. I am
thirteen and Ive got four sisters!!! (Sammy, Karen, Angie and
Kim) My mums called Doris and she works in a bank. My
dad, Rodney, is a mechanic. We live in a town called
Scunthorpe. It is really boring and I dont like it very much.
Before we lived in Manchester.
Ive got lots of hobbies. I collect pens (now Ive got 450 of
them!). Also I make model aeroplanes. I dont like sport very
much, but I like table tennis. I am really good at it!! My
favourite music is heavy metal and my favourite group is The
Electric Chameleons.
What about you? Write to me and tell me about your
hobbies and sports.
Bye!
Rodney
Marking criteria
Give an impression mark out of ten. Use this rating scale to help you:
9/10 Communicates all the relevant information and is interesting to
read. Hardly any mistakes of grammar, lexis, spelling or
punctuation.
7/8 Communicates the information needed. Few mistakes of
grammar, lexis, spelling or punctuation.
5/6 Communicates most of the information needed. Quite a few
mistakes, but it is possible to understand.
3/4 A lot of the information is not communicated and mistakes
make it difficult to understand.
1/2 Has communicated very little.
Assessment task 2 Writing (beginner)
Use the notes to write a description of the animal.
Example:
Her name is Booboo and she is a small animal. She is...
Name
Age
Family
Lives in
Favourite
food
Favourite music
Likes
Dislikes
Marking criteria
Give a mark out of five for the information included and a mark out
of five for accuracy (mistakes in terms of grammar and lexis)
Assessment task 3 Reading (elementary)
You can use dictionaries for this task.
1 Read the information about chameleons. Match the titles with
the paragraphs (1,2 and 3). (3pts)
a Eyes, tongues and tails. b Colour changes. c Insect hunters.
2 Read it again and complete this table (4pts)
Name of animal What they eat
Where they live Their colour
3 Why can chameleons catch insects? List three reasons (3pts)
1 Chameleons are lizards that live in Africa and the island of
Madagascar. They cant move very fast and they are not very
intelligent, but they hunt insects like flies and mosquitoes which are
very fast. How do they do it?
2 First, chameleons can move their eyes in different directions. They
can look in front of them and behind them at the same time! They
have special tails, so that they can hold onto things and they have
very long tongues which can catch the insects.
3 But best of all, the poor insects cannot see them, because
chameleons can completely change colour. If they are sitting in a
tree, they can turn brown and green. If they are sitting on a grey
stone, they can turn grey. So, when a chameleon feels hungry it just
has to sit and wait for its lunch to fly past!
Marking scheme (task 3)
Section 1: one point for every correct answer:
a 2 b 3 c 1
Section 2: one point for every correct answer
Name of animal: chameleon
Where they live: Africa/Madagascar
What they eat: insects (flies/mosquitoes)
Their colour: green/grey/brown different colours
Section 3: three points
Students do not have to write complete sentences. Notes are fine. Do
not penalise students for bad spelling, grammar etc.
Chameleons can move eyes/see in front and behind them at the same
time. They have long tongues. They can change colour/insects cant
see them.
Assessment task 4 Listening (post-elementary)
Listen to the story of the tortoise and the hare.
a Decide if the sentences are true (T), false (F), or there is no
information about them (NI).
Example: The tortoise and the hare lived in the old house. NI
1 The tortoise wanted to have a race.
2 The distance was one kilometre.
3 A lot of animals came to watch the race.
4 The race happened in September.
b Listen again and answer these questions.
Example: What time were they ready to start the race? 3.55
5 When did the race start?
6 When did the hare stop?
7 What time did he go to sleep?
8 What time did the tortoise go past him?
9 When did the hare wake up?
10 What time did the tortoise cross the winning line?
4.00 4.10 4.15 4.50 4.59 5.00
Tapescript
O nce upon a tim e there w as a hare. H e could run very fast and he w anted to race
w ith all the other anim als in the forest. O ne day he decided to have a race w ith the
tortoise. H e thought it w ould be very funny, because the tortoise w as very, very
slow . The hare w as sure that he w ould w in. The tortoise didnt w ant to race but
finally the hare persuaded her. The hare invited all the anim als of the forest to com e
and see the race. The race w as betw een the old house and the river, a distance of
about one kilom etre.
A ll the anim als of the forest cam e and at five to four they w ere ready. The race
started at exactly four oclock. The hare ran dow n the hill very fast. A fter ten
m inutes he w as near the river, near the end of the race. But it w as a very hot day
and he decided to sit dow n and w ait for the tortoise to com e past. H e sat dow n
under a tree and because it w as hot he felt sleepy. H e w ent to sleep at a quarter
past four.
M eanw hile, the tortoise w as going slow ly on. Finally, after fifty m inutes she arrived
at the tree w here the hare w as sleeping. She looked at him and carried on, tow ards
the river.
A few m inutes later, at one m inute to five, the hare w oke up. H e ran very fast and
in one m inute he w as at the w inning line. But, just as he arrived, at exactly five
oclock, the tortoise crossed the line. She had w on the race!!!!
Marking scheme
O ne point for each answ er: Total out of ten.
1F 2 T 3 T 4 N I 5 4.00pm 6 4.10 7 4.15 8 4.50 9 4.59 10 5.00
Assessment task 5 Speaking (elementary)
Part 1
Student A:
Find out this information about your partner:
name/age/family/favourite music/what s/he does at the weekends
Student B:
Find out this information about your partner:
name/age/hobbies/favourite school subject /what s/he does in the
holidays
Part 2
Student A:
Describe this half of the picture to
your partner. Then listen to his/her
description and draw the rest of the
picture. You can ask him/her
questions
Student B:
Listen to your partners description
of the other half of the picture and
draw it. You can ask him/her
questions. Then describe the other
half to him/her.
Part 3
Student A:
Imagine that you are a journalist interviewing a famous film star.
Ask him/her about these things: family/home/dislike/hobbies/plans
for the future
Imagine that you are a famous tennis player. Think about these
things: family / home / hobbies / ambitions /plans
Then answer the journalists questions
Student B:
Imagine that you are a famous film star. Think about these things:
family /home /likes / dislikes/hobbies/plans
Then answer the journalists questions
Imagine that you are a journalist interviewing a famous tennis
player. Ask him/her about these things:
family/home/hobbies/ambitions/plans for the future
Administration
Students to be interviewed in pairs. If possible, interview students
away from the rest of the class (though you could do it at the front
while the others get on with something else). Let them choose who
they want to do the interview with.
Before the interviews, tell the students the components of the interview:
personal information (warmer)
picture description (information gap)
roleplay
If the interview is taking too long and you feel that you have already
seen enough you can miss out one of the above components (eg
picture description).
Then explain to them the criteria that you will be using to assess
them. During the interview, try to say as little as possible. However,
help students if there is a breakdown in communication.
Use this scale to assess students:
Oral assessment rating scale
/5 FLUENCY /5 ACCURACY
5 Tasks done very well
Little hesitation
4 Tasks done quite well
Some hesitation
3 Tasks done adequately
Quite a few pauses
2 Tasks not done adequately
A lot of hesitation
1 Response completely
inadequate
Correct use of structures studied
so far. Clear pronunciation
Use of different structures covered
so far. Not many mistakes
Pronunciation quite clear
Use of some of the structures
covered. Some mistakes, but
reasonable understanding possible
Pronunciation satisfactory
Structure and lexis limited. A lot
of mistakes. Poor pronunciation
Almost incomprehensible
Assessment task 6 Speaking (Intermediate)
Students perform the task on page 45 in pairs. One student is given
Task Sheet A and the other student is given Task Sheet B.
Task Sheet A
You are a member of the two person student committee at your school
responsible for student facilities. Your partner is the other member of
the committee. The school has asked the committee to present a
proposal on how best to spend money available for next year on
different facilities and whether some facilities should be discontinued.
In your opinion, the school should spend money on:
more books for the library (there are too few at present)
more equipment for the science block (present material is very old)
more sports equipment (there is hardly any at present)
more cultural trips (eg theatre, museums, local parks)
more extra-curricular activities with other schools in the area
In your opinion, the following should be discontinued:
student caf (too expensive and not used very much)
school magazine (very little interest and participation)
daily newspaper (nobody reads it)
parent/teacher association (hardly any interest)
school swimming lessons (everyone learns outside school anyway)
Discuss all the above points with your partner and come to an
agreement on what the proposal will be. You start by asking your
partner what he/she thinks about the library.
Task Sheet B
You are a member of the two person student committee at your school
responsible for student facilities. Your partner is the other member of
the committee. The school has asked the committee to present a
proposal on how best to spend money available for next year on
different facilities and whether some facilities should be discontinued.
In your opinion, the school should spend money on:
student caf (people would use it if it were cheaper)
school magazine (it needs somebody to get people interested)
daily newspaper (students should know about current affairs)
parent/teacher association (very useful for solving problems)
school swimming lessons (some students do not learn outside)
In your opinion, the following should be discontinued:
library (very little interest and the books are all old)
equipment for the science block (present material is sufficient)
sports equipment (most students practice sports outside school)
cultural trips (students are here to learn not to travel)
extra-curricular activities with other schools in the area
Discuss all the above points with your partner and come to an
agreement on what the proposal will be. Your partner will start by
asking your opinion about one of the above points.
Oral rating scales
Below are examples of rating scales for use with the example of an
oral test on page 45. After using the rating scales, consider the
following questions and discuss with a colleague if possible:
1 Are the scales easy to use?
2 Did the scales become easier to use after rating 3 or 4
performances?
3 Could the scales be integrated to form one holistic scale?
Communication
5 Fluent communication
4 Good communication
3 Satisfactory communication
2 Communication hesitant
1 Communication minimal
Pronunciation and structure
5 Clear pronunciation and wholly appropriate use of structure
4 Few inaccuracies of structure and pronunciation
3 Inaccuracies of structure and pronunciation do not seriously
impede understanding
2 Inaccuracies of structure and pronunciation impede
understanding
1 Inaccuracies of structure and pronunciation make understanding
almost impossible
Vocabulary
5 Wholly appropriate for task
4 Few limitations
3 Sometimes limited
2 Limitations affected task considerably
1 Inadequate for task
Assessment task 7
Language (elementary)
Total /30
a Complete the description with these prepositions: (10pts)
at (3)
in (5)
on
from
by
Example:
Kristen Eriksen comes..from.. a small town near Oslo, 1.... Norway.
She gets up early, 2.... about 6.30, because her school starts 3.... 8.00
am. She usually goes to school 4.... bus and sometimes her dad takes
her. After school she often goes to the sports centre. She loves
swimming and playing tennis. Then, 5.... the evenings she does
homework and watches television.
6.... the weekends she doesnt have to get up early. 7....
Sunday mornings she goes to the country with her family. 8....
the winter they ski and 9.... the summer they go for long walks.
Sometimes, 10.... July and August, they go to the coast. But Kristen
doesnt swim, because the water is very cold!
b Write questions about Kristen for these answers: (10pts)
Example:
Near Oslo. Where does Kristen live?
1 At 6.30 in the morning. ..................................?
2 Usually by bus. ..........................................?
3 To the sports centre. .....................................?
4 No, she doesnt get up early. ..............................?
5 She goes with her family. ..................................?
c Correct these mistakes. (5pts)
Example:
She gets at school by bus.
She gets to school by bus.
1 Kristens brother studys at university.
2 What time do get you up?
3 How long does you take to go to school?
4 How is the weather in August?
5 It is snow a lot in the winter.
d Find the words for these pictures (5pts)
Example:
sunny
1 ..................................................
2 ..................................................
3 ..................................................
4 ..................................................
5 ..................................................
Marking scheme
a One point for each answer
1 in 2 at 3 at 4 by 5 in 6 at 7 on 8 in 9 in 10 in
b Two points for each correct sentence. If there is only one small
mistake (not related to word order and use of auxiliaries) give one
out of two. Possible alternatives in brackets. (10pts)
1 What time does she (usually) get up in the morning? (during the
week)
2 How does she (usually) go (get) to school?
3 Where does she go after school?
4 Does she get up early at the weekend? (on Saturday and Sunday)
5 Who does she go (with) to the country with?
c One point for each correction
1 studies
2 do you get
3 does it take you
4 what.... like
5 snows
d One point for each word
1 windy
2 cold
3 snowy
4 hot
5 rainy
2.4 Writing, administering and marking tests
A s a teacher, you m ay often find that you are expected to w rite, adm inister and
m ark tests. A ll of these tasks can be tim e-consum ing and w e need to produce
practical tests w hich can assess our students as reliably as possible in the tim e that
w e have available to us.
Writing tests
W hen w e start w riting tests w e need to avoid confusion and am biguity w hich m ay
m ake our test unreliable or invalid. H aving done an initial draft of the test w e then
need to check w hat w e have done. Firstly, w e should do the test ourselves to see if
w e can spot any changes that need to be m ade. Secondly, w e can ask a colleague to
do the test. W hat m ay be obvious to you m ay not be so obvious to another person -
a colleague should help advise you on other changes. O ne w ay of testing w riting
w hich can m ake this process easier is by w riting your tests w ith other colleagues
from the beginning. First you should agree on w hat you w ant to test. Then you can
either divide up the w ork (one person to do a listening test, another a reading test)
or you can sit dow n and w ork on each test together. A collective approach to
w riting tests m akes this difficult task easier and at the sam e tim e it provides you
w ith an opportunity to com pare ideas and attitudes. It is also an excellent tim e to
discuss the criteria that you are going to use to assess perform ance in the tests. This
of course assum es that tim e is available for these activities - you m ay have to
com prom ise.
The first area to think about is that of w riting instructions. It is very easy to
concentrate only on the content of the test itself and to forget that students w ill
need to know w hat to do during the test. Students only know w hat they have to do
during the test by reading or hearing the instructions - this is often called the rubric.
You m ay have a test w hich appears to be practical and valid, but w ithout a clear and
concise rubric the test w ill soon lose its validity if students are unclear as to w hat is
expected of them .
C larity is essential in rubrics. The rubric should tell the students exactly w hat they
have to do, how they have to do it and w hat the m arker is looking for. If the
language is unclear, students m ay not know w hat is expected of them and could fail
to perform to their full capability in the test.
Consider which of the following rubrics is clearer, a or b:
a Listen to the tape and put the right answer to the questions.
b Listen to the tape and answer questions 110 below by putting a
cross (X) in the correct box next to each question.
In the above cases, rubric b is clearer, as it tells students w hat they have to w rite
and w here they have to w rite it w hereas rubric a does not tell students exactly w hat
they m ust do to com plete the test.
A nother im portant point is conciseness. The rubric should not be too long. If the
rubric is so long that the student has to take in large am ounts of inform ation, then
it is likely that he/she m ay concentrate on w hat and w here to produce their answ er
rather than on getting the right answ er itself. W ith unnecessarily long rubrics, the
objective of m easuring the students true perform ance, as far as this is possible, w ill
be com prom ised.
Consider which of the following rubrics is more concise, a or b:
a Read the text which appears on the second page of the text booklet
you have been provided with. Look at the answer box which has
columns for information. Underneath the heading of each column,
you will find spaces in which you should write your answer
according to the text. Use the words you hear in the text to fill in
each of the columns. Sometimes, you may find that a column has
no corresponding information in the text. In these cases, do not
write anything in that column. Leave that column blank. All the
other columns will require you to write some information.
b Read text number 7 and fill in the information in each column of
the box next to the text. Write the exact word used in the text in the
appropriate column. If there is no information for any particular
column, leave the column blank.
In the above cases, b is m ore concise than a and has the advantage that students do
not have to spend tim e trying to decipher the m eaning of rubric a w hilst they are
attem pting to do the test. O ne w ay of avoiding any potential problem s can be to
put the test rubric in the studentsm other tongue. W hen you are checking the tasks
them selves you also need to think about background know ledge. It should be
im possible to answ er any questions correctly w ithout reading or listening to a text.
Students should not be able to use their know ledge of the w orld or background
know ledge of certain subject areas to answ er the questions.
Consider the following item, taken from a reading test. Can you
answer this question?
When was the Boeing 747 (Jumbo Jet) first built?
a the 1950s b the 1970s c the 1990s
M any students m ay be able to answ er this question w ithout referring to the reading
text in w hich the answ er appears as a result of their know ledge of the subject area.
Therefore, those students w ho happened to know w hen the Boeing 747 w as built
w ould have an unfavourable advantage over those students w ho did not happen to
know the date.
It is also im portant to look out for any kind of cultural bias w hen w riting or
checking test questions. N o item s in your test should depend on specific know ledge
of certain cultures or custom s. The test should not require the student to
dem onstrate know ledge of a particular culture.
Consider the following item, taken from a reading test:
The Smiths are a typical English working class family and they have
their meals at normal times. They have their evening tea when they
get back from work.
What time do the Smiths have their evening tea?
a 3pm b 6pm c 8pm
This question tests know ledge of custom s in Britain, rather than ability to read. You
cannot answ er the question if you do not know w hat tim e people eat and w hen
they norm ally finish w ork in Britain.
O ne of the m ost im portant considerations is the content of the test. The test should
reflect w hat the students have been doing in class ie should accurately reflect the
syllabus and design of the course in term s of content and form at.
For exam ple, w hen w riting a progress test after the first seven units of a course, it
w ould be unw ise to base the test on the first tw o units alone. A test at the end of
the course should attem pt to reflect the w hole course or year as far as possible ie
the test should sam ple as w idely and unpredictably as possible from the content of
the course, bearing in m ind tim e constraints. Therefore, w hen you are checking your
test it is w orth asking yourself how w ell it reflects w hat you have been doing w ith
your students.
A ll item s in your test should be also relevant in term s of real w orld language use.
The task w hich students are expected to perform should correspond as closely as
possible to som e use of the language in the real w orld. The key here is to m ake the
task appear to the students as som ething w hich they m ight actually have to do w ith
the language.
Consider the following item, taken from a writing test. Is this a task
that students might have to do in the real world?:
Composition: Write a short composition (about 250 words) about
what you did in your summer holidays. Include details about the
journey, the place and the accommodation where you stayed.
W ithout searching too far for potential relevance, it is difficult to think of a context
in the real w orld w hen a language learner m ight need to produce this kind of
w ritten w ork. The sam e type of production m ight be elicited in a m ore authentic
m anner by setting the w ritten piece in the context of, for exam ple, answ ering a
letter from a penfriend asking about the sum m er holidays.
Tim e is also very im portant and your test should not place undue pressure on the
students in term s of tim e needed to com plete the tasks set. In reading tests,
students should not be required to answ er item s in a lim ited tim e w hich w ould be
com pletely unrealistic for a reader in real life. Tim e should be allow ed to read the
text, read the item s and answ er the item s, including a tim e allow ance for re-reading
for clarification and possible am endm ent to answ ers.
In listening tests, students should not be required to answ er too m any item s in a
short period of tim e, especially if the listening text gives answ ers to item s in rapid
succession, perhaps not even giving tim e to note dow n required answ ers.
In w riting tests, students should not be required to produce x am ount of w ords in
an unrealistic tim e lim it, eg a letter to a prospective em ployer in 5-10 m inutes. Tim e
should be given for planning, w riting, reading and possible re-w riting.
In speaking tests w here there is an interlocutor, unrealistic tim e pressures on
answ ers and oral production should be avoided, eg hurrying students into an
answ er w ith no tim e for reflection or checking understanding or asking for
repetition.
Finally, w hen you are w riting your tests you should also take into account practical
adm inistrative factors. For exam ple, you should not w rite tests that require a lot of
photocopying if it is difficult and expensive to do in your situation. You should also
avoid using equipm ent (like video recorders) w hich is difficult for you to obtain.
Checking tests
Look at the test items on the next few pages. Identify the problems in
test items.
Use this checklist to help you:
Is there more than one possible answer?
Is there no correct answer?
Is there enough context provided to choose the correct answer?
Could a test-wise student guess the answer without reading or
listening to the text?
Does it test what it says it is going to test? (or does it test something
else?)
Does it test the ability to do puzzles, or IQ in general, rather than
language?
Does it test students imagination rather than their linguistic ability?
Does it test students skills or content knowledge of other academic
areas?
Does it test general knowledge of the world?
Does it test cultural knowledge rather than language?
Are the rubrics (instructions) clear and concise? Is the language in
the instructions more difficult than that in the text?
Will it be very time-consuming to mark and difficult to work out
scores?
Are there any typing errors that make it difficult to do?
Speaki ng
1 Intervi ewer: Right, now Mara what do you think about cricket?
2 Intervi ewer: Can you tell me the names of the animals in the picture?
3 Intervi ewer: What do you think about the political situation in South Africa?
Li steni ng
4 Li sten to the tal k and choose the correct answer:
a A ll elephants live in A frica.
b Elephants live in South A m erica.
c Elephants live in A frica and India.
5 Li sten to the di al ogue and answer the questi ons. You must wri te compl ete
sentences (remember adverbs of frequency).
Example:
W hat is the w eather like in Sydney in June?
In June it is rainy and som etim es it is cold.
You get three m arks for each correct sentence, one m ark for each adverb of
frequency used and half for each w ord you use related to w eather.
Wri ti ng
6 Wri te a ghost story (200 words). Try to make i t dramati c and exci ti ng!
7 Wri te i nstructi ons for a sci enti fi c experi ment. Draw di agrams to i l l ustrate i t.
8 Wri te a report of an i ntervi ew wi th a famous pop star. Use reported
statements and questi ons. Exampl e: I asked hi m about hi s fami l y and he
tol d me that...
Readi ng
9 Read the text bel ow and answer thi s questi on: What di d the captai n hear?
a A sheep or goat.
b A strange noise from the sea.
c Som ebody in pain.
It was a dark and misty night. The captain was sleeping in his cabin and
everything was quiet. Suddenly, the captain woke up. He heard a strange noise
coming from the deck. It sounded like an animal in pain, maybe a sheep or a
goat. He got dressed and cautiously climbed the ladder.
10 Compl ete the sentences wi th one of these words:
a how ever
b but
c in spite of
d although
They continued playing football .............. it w as raining
Vocabul ary
11 What are these words? nsyun / ureufni tr / angroe / dol c
12 Whi ch i s the odd one out?
a basketball
b table tennis
c ice-hockey
d cricket
Grammar
13 Compl ete the sentence wi th wi l l or goi ng to:
Tom orrow I think it..................rain.
14 Choose the al ternati ve that i s cl osest i n meani ng to the word whi ch i s
underl i ned:
Ive just finished it.
a A m inute ago.
b Yesterday.
c A w eek ago.
15 Marki ng scheme: Exerci se A:
1 = 2.5 2 = 2 3 = 3.25 4 =1.75 5 = 2 6 = .5 TO TA L = 12
Administering tests
A s w e pointed out in the introduction to this book, being tested as a student and
testing learners as a teacher can be traum atic. Therefore w e need to do w hat w e
can to reduce tension. A t the sam e tim e, w e need to m ake sure that form al
assessm ent does take place under test conditions ie that students cannot copy or
help each other. In norm al classroom conditions w e w ant to encourage co-
operation, but w hen carrying out form al assessm ent w e need to do the opposite, to
m ake sure that w e are testing the perform ance of each individual learner. W e thus
need to reduce to a m inim um any cheating that m ight go on in our classroom .
The first thing to consider is the place. M ost of the tim e w e give our tests in our
ow n classroom s. Before giving out the test papers it is a good idea to try to separate
students as m uch as possible, by m oving desks or placing students around the room .
If it is im possible to do this it m ay be w orth trying to m ove to another room . If you
do this, it is w orth choosing a place w here there w ill be a m inim um of interruptions
and outside background noise.
Tim e also has to be considered. Firstly, you should tell students how long they have
at the beginning as w ell as w riting it on the paper. In your ow n classroom tests you
can be relatively flexible and give students a bit longer if you see that they are
having problem s. If it is a school test it is m ore im portant to avoid any unfair
advantage for som e students w ho m ight benefit from m ore tim e.
It is im portant to have m aterials w ell prepared in advance. M ake sure that before
you adm inister the test you have all the necessary printed m aterial and that there are
adequate supplies of all test papers, m aps, charts or any other printed m atter for the
num ber of students w ho w ill be taking the test. M ake sure that photocopying is of a
satisfactory quality and that no errors have been m ade in printing and preparation. If
there are any errors, eg spelling, repeated questions, inform students before the test
begins.
In addition, m ake sure that you have checked any recorded m aterial before the test
is to be adm inistered. Som etim es, you m ay find that there is nothing you can do
about a certain problem , eg poor quality of recording or varying sound recording
levels. In these cases, students should sim ply be w arned. H ow ever, all attem pts
should be m ade to rem edy the problem before the next adm inistration of the test.
Electrical equipm ent also needs to be checked beforehand. M ake sure that any audio
equipm ent is adequate in term s of sound quality and acoustics in the room w here
the test is to be adm inistered.
Your students w ill need to be prepared for the test. If you have short and regular
progress tests they w ill be m uch less w orried than if they have few er but m ore
im portant tests. Tell students in advance that you w ill be giving them a test and at
w hat tim e they are expected to arrive, w hat tim e the test w ill start and w hat tim e
the test w ill finish. Students should also know w hat m aterials they need to bring
w ith them to the test, eg pens, pencils, erasers etc. If dictionaries are to be used in
any test, students should be told to bring their ow n copy of preferably the sam e
edition so that no student has an unfair advantage.
Tell students about the test conditions. For exam ple, there should be no talking and
if they w ish to ask a question they should raise their hand. Students should also be
seated in a w ay that they cannot copy answ ers from a neighbour or com m unicate
answ ers to a neighbour. You can also tell them w hat w ill happen if they are caught
cheating, for exam ple, they m ight be given a zero score or even disciplinary action
m ight be taken.
If you are adm inistering a test w ith your colleagues you need to agree on conditions.
D ecisions should be taken beforehand about w hat to do in cases of students arriving
late, students copying, students finishing before the allocated tim e etc. M any of
these decisions w ill differ from institution to institution and m ay depend on policy
and internal rules and regulations. If you or your colleagues are to act as
interlocutors in speaking tests, it is essential that everybody agrees on how to act as
interlocutor and w hat is to be expected of students in the test. A ll those w ho are to
act as interlocutors should hold m eetings before the test to agree on criteria for
perform ance and to practise using the test m aterials.
Marking tests
M arking is one of the m ost tim e-consum ing parts of m any teachersjobs. A s w e
suggested earlier (in the sub-section on planning assessm ent program m es), short
and regular assessm ent tasks spread the test m arking load over the w hole term . This
not only avoids stress and exhaustion on your part, but it should also m ean that you
w ill be able to m ark m ore accurately and reliably. W e have also m entioned the need
to consider m arking tim e w hen choosing test form ats. W ith very sm all classes w e
can afford to have m ore labour-intensive test form ats (for exam ple open
com positions and oral interview s). If w e have very large classes w e w ill need to
choose form ats that w ill enable us to m ark large num bers of tests in a short period
of tim e (like m ultiple choice questions).
O n the one hand w e have discrete itemor objective tests. These tests are so called
because of the w ay they are m arked. A n objective test could in theory be m arked by
any person capable of interpreting and applying a m arking key w hich gives the
correct answ ers w hich are unique and not negotiable. A n exam ple w ould be a
tem plate or m ask as often used to m ark m ultiple-choice questions. For exam ple, the
only possible answ er to question 6 is (b)option. The person m arking the test w ill
sim ply apply the tem plate or m arking key and w ill be able to total the correct
answ ers w hich w ill give a raw score. In this sense, objective testing can be
considered as m arking by counting.
Objective tests are easily and quickly m arked by non-specialists. H ow ever, they can
be difficult to w rite so that they are reliable, eg in the case of the m ultiple-choice
question paper it is often difficult to devise suitable and plausible distracters and the
guessing factor is also very high. A nother disadvantage is in the case w here m any
variants of a sim ilar answ er w ould be a suitable and correct answ er to an item . For
exam ple in a listening test, w hat is an acceptable answ er to the question What does
the man want? The answ er key m ight state that the correct answ er is He wants to
buy petrol. But w ould Buy petrol, or Petrol buy or petrol also be acceptable
answ ers? This w ould depend on the test designer, w ho w ould be responsible for
ensuring that answ er keys contained all possible answ ers.
Subjective tests, as opposed to objective tests, are not based on counting, but
depend on som ebodys opinion, a judgem ent, a decision about candidate
perform ance. The person w ho is to m ake the judgem ent is expected to be qualified
to m ake that judgem ent, eg you as a language teacher could m ake judgem ents
about oral perform ances of students in a speaking test. O n the other hand, m any of
the people w ho m ay be perfectly qualified to m ark your objective, m ultiple-choice
listening test m ay not be suitable for use as raters of oral perform ance. Subjective
tests can provide a w ide sam ple of studentslanguage in a relatively short tim e
think of how m uch your students could actually say in 15 m inutes.
They can be objectivised by using rating scales w hich outline a description of w hat
each point on a scale m eans, eg 5 = the ability to ... . Subjective tests m ay take up a
lot of your tim e. For exam ple, if a class of forty students can be tested at the sam e
tim e using a forty m inute reading test, this is m uch m ore practical in term s of tim e
than holding forty interview s of fifteen m inutes each or m arking forty w ritten
com positions.
Reliability of raters is the greatest problem area. The key question is how to ensure
that different raters apply the scales in the sam e w ay. This is called inter-rater
reliability. A lso, another question is how to ensure that the sam e rater w ill apply the
scales in the sam e w ay on different days or at different tim es of the day. This is
called intra-rater reliability. These issues can be addressed by rating w orkshops and
training packages, although this im plies m ore tim e spent by teachers and any other
person expected to carry out rating of tests.
It m ust be pointed out here that objectively m arked tests are not to be considered as
good and subjectively m arked tests as bad. They are sim ply different w ays of
m arking.
W hen you are m arking a test that you have w ritten yourself it is a good idea to
define beforehand the answ er key especially w hen other teachers are going to m ark
the test. The key should be easy to use and leave no doubt in the m arkers m ind as
to w hat is a correct and incorrect answ er. This can be a relatively sim ple process in
the case of m ultiple-choice item s or rather m ore com plicated in the case of open-
ended questions. W ith open-ended questions, ensure that the answ er key covers all
possible answ ers. A fter producing your answ er key, it is a good idea to show it to a
colleague if possible. Your colleague could check the answ ers for possible errors,
additions or deletions to be m ade. A fter that, you should agree on the answ er key
w ith all those w ho w ill be using it to m ark the test.
It is also a good idea to w ork out how you are going to distribute m arks before
adm inistering the test. It can be very useful to w rite the m arking schem e on the test
paper itself, so that the students also know how m uch each section is w orth.
W hen other teachers are going to use the test you w ill need to produce a m arking
schem e. The m arking schem e should be easy to use and should leave no doubt in
the m arkers m ind as to how m any points each item is w orth. M arkers should not be
expected to perform com plicated m athem atical calculations to arrive at a students
final score. Look at the exam ple of a m arking key on page 48.
Subjective tests and rating
For subjective tests w e need to look at rating ie w here results are based on
som ebodys opinion about candidate perform ance, eg in an oral test or a w ritten
com position. W e looked at the subject in the chapter on inform al assessm ent
(particularly in the sub-chapters on speaking and w riting). H ow ever, for form al tests
it is not only im portant to establish criteria. If m ore than one teacher is adm inistering
the test, w e w ill need to agree on interpretation of rating criteria.
W e have already m entioned the tw o kinds of rating scales w hich w e can use. Rating
scales can either contain descriptions of all activities w ithin one level or can break
dow n the activities into separate scales and provide descriptions for each activity at
each level. The first is holistic rating and the second exam ple is called analytic rating
(see page 13).
The advantage of holistic rating for testing is that raters can internalise the
descriptions in a relatively short period of tim e, eg after practice w ith a few sam ple
perform ances. This system is therefore practical and quick to adm inister. The
disadvantage is that student perform ances can often cut across the descriptions, eg
one activity m ay belong to level 3 and another activity to level 4. H ow ever, raters are
trained in all cases to choose the closest description of the perform ance.
The advantage of analytic rating is that raters m ay find it easier to assign a certain
level using sim plified and discrete scales. The disadvantage is that it w ill probably be
less practical than holistic rating in term s of tim e, paper and training.
Look at page 46 for an exam ple of oral rating scales.
W hen you are assessing your students in this w ay, it is im portant to achieve intra-
rater reliability to m ake sure that you rate them consistently. In an attem pt to
increase our ow n intra-rater reliability w e can look at a piece of studentsw ork and
then look at it again tw o w eeks later. Then w e can com pare assessm ents and if there
are differences, think about w hy and w here w e w ent w rong.
A n im portant factor w hen m ore than one teacher is m arking a test is inter-rater
reliability to ensure that all raters assess in the sam e w ay and that all raters agree
on the interpretation and m eaning of the descriptions in the rating scales. The
objective is to m inim ise the possibility of a students m ark being affected by the rater
w ho assesses their perform ance.
In an attem pt to m axim ise inter-rater reliability, you could hold m eetings w ith your
colleagues to discuss rating scales and sam ples of student perform ances could be
provided. Raters could discuss the sam ple perform ances, rate them , and thus see if
they are applying the sam e criteria. W ritten perform ances should be easy to supply.
In the case of oral perform ances, these could be recorded on audio or video tape.
Inter-rater reliability activity
Written performance
Look at the example of the writing assessment task on page 39.
Then look at the criteria for marking it on page 39.
Give copies of three answers on page 58 to one or more of your
colleagues. Also give them a copy of the marking criteria.
Read the letters yourself and rate them.
Ask your colleague(s) to use the marking criteria to rate the
performances.
Compare your results and if there are any discrepancies, discuss the
performances and the rating scales. Try to come to an agreement
about your rating of the compositions. Then compare your marks
with those on page 62.
The next time you are marking written performance from a test, do
the same activity with your colleagues.
By doing activities like these, you will achieve a much greater degree
of reliability for tests at your school. You will also need to work out
criteria that is clear for both you and your colleagues.
Inter/intra-rater reliability activity
Look at the example of a paired oral test on page 45.
Give the test to students in your class and record some of the
discussions onto audio tape. Make sure that the tapes are labelled
with student names. (You might want to use numbers rather than
names for increased reliability of the activity !)
Assess the student performances using the rating scales given on
page 46 and keep a record of the marks assigned.
Inter-rater reliability
Arrange a meeting with one or more of your colleagues and explain
that you want to see if you can agree on the level of student
performances and whether the rating scales are easy to use.
Give your colleague(s) the oral test material and the rating scales
and give them time to familiarise themselves with the task.
Play a selection of student performances (perhaps 3 or 4) including,
if possible, what in your opinion is a clear pass, a clear fail and a
borderline case.
Ask your colleague(s) to use the rating scales to rate the
performances using the rating scales.
Compare your original results and if there are discrepancies, discuss
the performances and the rating scales. Try to come to an
agreement on the criteria to be used for each descriptor in the rating
scales. (Dont worry if there are minor differences in rating these
are to be expected).
Intra-rater reliability
Keep the recordings and original marks in a safe place and, if you
have time, set aside an hour or so one day about two or three weeks
after you administered the oral test.
Play the recordings again and rate the performances again, in a
random order, using the same rating scales. Do not refer to the
original marks.
When you have finished, compare your new ratings with the original
marks.
If the marks are the same, you appear to be quite a reliable rater
within your own performance ! If the marks are a little different,
dont worry this might be expected. If the marks are very different,
think about reasons why this might be so are you stricter or more
lenient? Do you think that perhaps the rating scales are confusing?
How could the problem be solved?
2.5 Results f rom f ormal assessment
O nce students have taken a form al test and it has been m arked, you w ill have
results. These results w ill usually give a m ark or score for each student, for exam ple
68 out of 100. O n the other hand, the results m ight sim ply tell you that a student
has passed or failed the test. But w hat is a pass m ark? W hat is a fail?
O ne w ay of deciding this is by norm-referencing.This consists of putting the students
in a list or scale depending on the m ark they achieved in the test. The student w ith
the low est m ark is at the bottom of the list and the student w ith the highest m ark is
at the top of the list. A ll other students are betw een the tw o extrem es. A pass in the
test m ight be decided as the top 60% of students, w ith 40% failing, or as the top
50% of students w ith 50% failing. This is often used in public exam inations, but it is
not suitable for classroom situations. For exam ple, it m ight create a situation in
w hich a high ability candidate fails because the other students in the sam e class are
of a higher ability. C onversely, it m ight create a situation in w hich a low ability
candidate passes because the other candidates in the class are of a low er ability. The
extrem es m ight be that student X achieves 90% in the test and fails because all
other students achieved betw een 91 and 100% ; or student Y achieves 35% in the
test and passes because all other students achieved betw een 1 and 34%.
Criterion-referencing is a m uch m ore suitable w ay of m aking decisions about
classroom tests. This consists of m aking decisions about w hat is a pass and w hat is a
fail before the results are obtained. It norm ally m eans that w e define certain criteria
that candidates in the test m ust conform to, eg if the candidate can do X, Y and Z,
then he/she passes if he/she cannot do X, Y and Z, then he/she fails the test. A n
exam ple is the driving test in the U nited Kingdom : there is a specified list of things
w hich the candidate m ust be able to do in order to pass the test, eg m ake an
em ergency stop, turn the car round in the road in a num ber of m anoeuvres, reverse
round a corner etc. It can be a fair w ay of interpreting results because a pass or fail
is decided on each candidates perform ance in the test irrespective of the rest of the
candidates w ho take the sam e test. A lso, candidates know that they are not in
com petition w ith the rest of those w ho take the test.
Finally, w hen w e have interpreted the results of our tests and decided w hich
students have passed and failed, w e need to go over tests w ith students. It is
im portant to go over the answ ers to progress tests w ith the w hole class, so that
students realise w here they w ent w rong. Students can then think about w hat they
need to do to do to get better results the next tim e. W hen you give back pieces of
w riting done under test conditions you can use correction codes in the w ay
suggested in the previous section (Inform al assessm ent w riting). W hen you give
students their results on oral tests it is also im portant to identify w hat their strong
and w eak points are.
In this w ay results from form al tests can feed into learning and give students (as w ell
as you the teacher) vital inform ation about both perform ance and progress.
Summary
In conclusion, the follow ing key points should be rem em bered w hen adm inistering
form al tests and im plem enting a form al assessm ent program m e.
N There are m any reasons w hy w e should test our students and form al tests should
not be discarded as a w aste of valuable tim e. O n the contrary, they should be seen
as tim e w ell spent in helping to form an accurate assessm ent of our students
ability.
N A form al assessm ent program m e needs to be carefully planned and should tie in
w ith your educational objectives and the extent and tim ing of other form s of
m easuring studentsability such as self and inform al assessm ent.
N There are m any different types of test and test form ats. Each one is useful in its
ow n context and is less useful in other contexts. M uch w ill depend on w hat you
w ant the test to do for you in your teaching situation.
N W riting, m arking and adm inistering tests m ay appear to be a relatively sim ple
m atter. H ow ever, the extent to w hich these are carried out successfully w ill affect
the success of your tests and ultim ately the reliability of the results you obtain.
This in turn w ill directly affect the decisions you m ake about your students our
aim m ust be to m ake decisions w hich are based on accurate and reliable
inform ation.
Suggested answers to practical activities where necessary
Washback effects
1 Bad w ashback. Students w ill feel that luck w as m ore im portant than hard w ork.
2 Bad w ashback. Students w ill feel that classw ork has been a w aste of tim e.
3 G ood w ashback. The inform ation w ill help you plan your course to suit the needs
of your students.
4 G ood w ashback. Students w ill feel they have been tested fairly.
Your testi ng
a = placem ent
b = progress
c = diagnostic
d = sum m ative
e = proficiency
Identi fyi ng test formats
A ssessm ent task 1: C om bined (reading/w riting)
A ssessm ent task 2: G uided w riting (notes and pictures)
A ssessm ent task 3:
1 Identifying topic (m atching)
2 Table com pletion.
3 Short answ er questions.
a True/false b Short answ er questions
1 Inform ation gap (real inform ation)
2 Picture description
3 Roleplay
Roleplay
a G ap-fill b C ued questions (from answ ers) c Editing
d C ued vocabulary (pictures)
Checki ng tests
1 C ultural know ledge that ss could not be expected to know .
2 C ontent know ledge. A lso m ore of a vocabulary test than a speaking test w hich it
purports to be (therefore low validity).
3 G eneral know ledge of the w orld- w hich ss cannot be expected to know
(particularly young students).
4 You know the answ er w ithout listening. (c)
5 D oesnt test listening but is a test of language.
6 Tests studentsim agination m ore than their English.
7 Tests content know ledge of science.
8 Is a language test rather than a test of w riting (to com m unicate).
9 N o answ er is possible here.
10 This is a test of language not of reading.
11 This anagram activity tests the ability to do puzzles not know ledge of vocab.
12 M ore than one answ er is possible here, (eg basketball = no bat/cricket = played
outside)
13 N ot enough context is provided here, either is possible.
14 N ot enough context is provided. A ll three are possible.
15 The m arking schem e is horribly com plex and w ill w aste a lot of tim e.
Inter-rater rel i abi l i ty for wri tten performance
Suggested m arks for letters
1 4 The letter is badly organised. Because of this and because of the large num ber
of very basic structural and lexical errors, it is very difficult to read and understand.
A nother factor w hich m akes understanding difficult is the handw riting and constant
crossing out of w ords.
2 6 In this letter there is som e real com m unication. If this w ere a real penfriend
letter, the person getting it w ould learn som ething about M arta. H ow ever, there are
quite a few m istakes and crossings out w hich m akes it difficult to follow in parts.
3 9 This letter com m unicates the inform ation needed reasonably clearly and w ith
only a few m istakes of spelling or gram m ar. Finally, the letter is enthusiastic and
interesting to read.
Action points: f ormal assessment
1 A s a result of reading this chapter are you going to m ake any changes in the
w ay you test your students? If so, w hat are they?
2 A re you going to give all your form al tests in one block, or are you going to
spread them out over the term ?
(For example, using shorter assessm ent tasks)
3 W hat kind of w eighting are you going to give to different skills and language?
(For example, 30% speaking / 20% w riting etc)
4 W hat kind of test form ats are you going to use?
(For example, m ultiple-choice, guided w riting etc)
5 The next tim e you have to w rite tests are you going to do it on your ow n or
w ith a group of colleagues?
6 H ow are you going to adm inister your tests? W hat test conditions are you
going to introduce? H ow are you going to try to reduce anxiety am ongst your
students?
7 H ow are you going to m ark your tests? W hat m easures are you going to take to
increase the reliability of your m arking?
8 W hat proportion of m arks are you going to get from form al as opposed to
other form s of assessm ent?
Self -assessment
3.1 Introduction to self -assessment
W e have looked at how w e can assess our students, inform ally through assessing
class and hom ew ork or form ally through tests or assessm ent tasks. If both these
types of assessm ent are done thoroughly, w e w ill probably get a fair picture of our
studentsproblem s and a reasonably accurate m easure of their progress. H ow ever,
w e are ignoring an extrem ely useful source of inform ation, the students them selves.
Self-assessment can provide us w ith useful inform ation about studentsexpectations
and needs, their problem s and w orries, how they feel about their ow n progress,
their reactions to the m aterials and m ethods being used, w hat they think about the
course in general. Self-assessm ent can also be a m uch m ore direct and efficient w ay
of getting inform ation than teacher assessm ent. For exam ple, rather than give a
com prehensive diagnostic test to find out w hat areas of language students are w eak
on, it is m uch quicker to ask students directly w hat problem s they feel they have.
The m ost im portant benefits of self-assessm ent are related to its im pact on learning.
Self-assessm ent is an integral part of learning. To learn anything w e need to assess
w hat w e know already and how w e can im prove. W e norm ally do this
subconsciously. If w e are learning to cook, w e learn by thinking about w hat w e have
done and then im proving on it in the future.
Thinking about your language learning
Look at 112 below. Tick ( ) the points which you have thought
about and put a cross () next to those you have never considered.
1 your performance in another language in a particular situation,
eg speaking with somebody at a party
2 the fact that on some days you can communicate well and on
other days you cant say anything
3 why you can understand people in some situations but not in
others, eg you can understand people when you are talking face to
face, but not on the telephone
4 why some people are easy for you to understand and others arent
5 which common mistakes you keep on making,
eg grammar/vocab/pron
6 which grammatical areas are difficult for you
7 which sounds are difficult for you to say
8 what new words it would be useful for you to learn
9 how much you have learnt in a lesson or over a week
10 how much progress you have made over a course of study
11 which of these areas you need to improve on most: listening
speaking reading writing grammar pronunciation
12 how you can practise and improve on your own
3
C HA P T E R 3 Self-assessm ent
64
Your answers
If you have placed a tick next to m ost of the questions on page 63, you have been
thinking about your learning and assessing yourself. You probably did this on your
ow n, outside the language class. D o you think it w ould have been helpful to have
had guidance and encouragem ent in assessing yourself?
G ive the sam e questionnaire to your students (translate it into their L1 if necessary).
This questionnaire can be useful to start your students thinking about their ow n
learning and realising the im portance of self-assessm ent.
Self-assessm ent is thus a vital part of the learning process. If w e rely totally on
teacher-based assessm ent, w e are only carrying out part of the job of assessm ent.
W e are diagnosing problem s and w e are m easuring the progress that our students
have m ade. This inform ation can feed back into our ow n teaching, but it does not
necessarily feed back into the learning process. Students are often passive and w ait
for us to tell them if they have done w ell or badly. A t no tim e do they need to think
about w hat they have done, and w hy they have done w ell or badly. This can be seen
clearly w ith w riting. W hen students are given back corrected com positions they
often just look at the m ark and do not really learn from the feedback.
There is a saying in English that you learn from your own mistakes. Perhaps w e
could qualify this to you learn from your own mistakes when you think about
them. By conscious self-assessm ent, students are involved in the process of
assessm ent and therefore the inform ation from it can feed back into their ow n
learning. Self-assessm ent helps learners to think about their ow n progress and
problem s and then to find w ays of changing, adapting and im proving. This can be
seen w ith the exam ple m entioned earlier of w riting. If students are encouraged to
assess their ow n w ork, to think about their ow n m istakes and to try to correct them ,
they are m ore likely to im prove than w hen only the teacher assesses and corrects.
Teacher assessment
teacher assessm ent conclusions about progress
of learners progress diagnosis of problem s
teaching:
course content
Teacher and learner assessment
teacher/learner assessm ent conclusions about progress
diagnosis of problem s
learning:
course content
teaching:
course content
In the teaching-based m odel above, the results from assessm ent do feed back into
the content of a course. Rem edial w ork is done w hen necessary and objectives can
be changed as a result of inform ation from assessm ent. H ow ever, learners never
need to think consciously about their ow n problem s. In the learning-based m odel
both the teachers assessm ent and studentsself-assessm ent have a direct effect on
the learning process.
W hile there are m any benefits to be derived from self-assessm ent, m any teachers
m ay have serious reservations about im plem enting it. Firstly, it can take up a lot of
tim e, w hich is at a prem ium in m ost classes. This m eans that it m ust be stream lined
and integrated w ith other classroom activities. Secondly, self-assessm ent im plies a
know ledge about language and learning w hich m ost secondary learners do not
have. Therefore, self-assessm ent can only w ork if it is accom panied by learner
training. Thirdly, m any teachers m ay have their doubts about the m aturity of
students w hen they are asked to give them selves a m ark, w hich w ill be taken into
account in their overall assessm ent. The tendency to give them selves a higher m ark
could m ake self-assessm ent an unreliable influence on assessm ent. In fact m any
students tend to do the opposite and give them selves a low er m ark than they
deserve. In either case the link betw een assessm ent and other areas of assessm ent
needs to be looked at closely.
Perhaps the m ost im portant thing about self-assessm ent is that it cannot w ork in a
context w here m arks have an intrinsic value in them selves and there is com petition
betw een students. In fact m ost self-assessm ent should have nothing at all to do
w ith m arks. It should concentrate on thinking about perform ance and progress in
individual term s. This is particularly so in m ixed ability classes w here consideration of
learnersprogress has to take into account the often very different starting points of
individuals.
In this chapter w e w ill look at laying the foundations for self-assessm ent, starting
from the first w eek of a course. Then w e w ill look at how self-assessm ent of
perform ance can be integrated into classroom activities throughout a course. A fter
that w e w ill look at how students can review their ow n progress system atically and
at the link betw een self-assessm ent and other kinds of assessm ent.
Before w e look in detail at how to im plem ent self-assessm ent in the classroom let us
look at som e of the techniques that w e have at our disposal. M any of these are
sim ilar to those looked at in previous sections on teacher assessm ent. The difference
is that in this case the students them selves carry out the assessm ent.
Techniques for self-assessment
Look at the techniques below. Have you ever used any of them
yourself? Which would you like to try out with your students?
Descriptions
Profiles
students write reports about their English and give them to you.
Learner diaries
students keep learner diaries:
records of what they have done in class
what they have learnt
what problems they have had
Counselling sessions
you talk to students individually about their own progress
Rating
Rating scales
students rate their own performance or progress using a rating scale
with descriptors, eg a scale of 1 to 5. Students can use this structure
with no problems at all. Peer rating of other students can also give
students practice at this.
General rating
students give themselves an impression mark for different areas, eg
writing/speaking
Graphs
students rate how much they have understood using a graph
Adjectives
students choose from lists of adjectives to describe their own
attitudes, eg hard-working/relaxed/lazy
Monitoring
Self-editing
students look through their own compositions and try to correct
their mistakes. Peer editing can also give them practice at the same
time as helping their partners
Correction codes
students use your correction code, eg Pr= preposition, to try to
correct their own mistakes
Taping
students listen to a recording of themselves and try to correct
mistakes
Grading mistakes
students grade mistakes in order of seriousness
Test yourself
(self-check) students do short tests to find out how much they know
Questionnaires
Listing
students list specific problems for them, like words that they have
difficulty pronouncing
Ranking preferences
students rank activities in terms of which they enjoy, eg 1= listening
to songs 2= watching video
Ranking problems
students rank areas that are difficult for them, eg grammar: modal
verbs
Multiple choice questions
students answer questions about their habits, eg How much do you
study?
a a lot
b quite a lot
c a little
d not at all
Agreeing/disagreeing
students agree or disagree with statements about learning, eg
Learning is easy. They could use a range of statements from easy to
difficult.
Can/cannot questions
students answer questions about what they can or cannot do, eg
Can you talk about the weather?
Short answer questions
students answer questions about their preferences, performance or
progress, eg What topic areas would you like to look at this year?
Surveys
Group surveys
students find out about each others learning preferences or
problems and then report the group results to the rest of the class.
Class surveys
students go around the class and ask questions about learning to
the others. Then they report the results to the whole class.
3.2 Laying the f oundations
The best tim e to start integrating self-assessm ent into a course and to introduce
students to self-assessm ent is from the first day of class. Before you open a textbook
or begin to do other classroom activities, you have the ideal opportunity to get
students to think about their ow n learning. The best w ay of doing this can be either
through learner questionnaires or through classroom survey activities. In this section
w e w ill look at som e learner training activities for the first tw o w eeks of term in
w hich self-assessm ent features prom inently.
Classroom survey: past learning experiences
Even if students com e from sim ilar learning backgrounds it can be useful to get
them to recall previous learning experiences. A s w ell as rem inding students of their
past learning experiences the feedback from students can give you useful
inform ation to help your ow n planning.
Learner questi onnai re: Ini ti al sel f-assessment
A sim ple questionnaire can help you find out in m ore detail about how your students
feel about their ow n English, doing an initial self-assessm ent of strong and w eak
areas. It is im portant to realise that at this stage any grades that students give are
totally subjective and related to how they feel about previous learning experiences.
These questionnaires can be handed in individually, in w hich case they can provide a
useful initial record of each students feelings tow ards their learning. A nother w ay of
gathering this sort of feedback from older and higher level learners is through a
letter or note directed to you the teacher. Students can w rite a note telling you
about their problem s and how they feel about the different areas of English.
Give this questionnaire to your students at the beginning of the year.
My Learning
a So far have you found learning English:
quite difficult?
quite easy?
very difficult?
b Which of these areas of English are easiest for you? Order them in
these terms (1 = easiest 7 = the most difficult)
speaking
listening
writing
reading
grammar
vocabulary
pronunciation
c Give yourself a mark out of ten for each of the areas above.
Example:
speaking 6/10
Dont worry about other people, think about how you feel about the
different areas. If you feel satisfied, give yourself a good mark; if you
dont give yourself a low mark.
Di agnosti c questi onnai res
If your students com e from a variety of prim ary schools, it can be interesting to find
out w hat they have covered and to ask students w hat they think they are able to
do. By collecting them in and looking at them you can get a better idea of your
students. A t the sam e tim e the students them selves are receiving training in self-
assessm ent.
Look through this diagnostic questionnaire and modify it according to
the needs of your students (putting it into their L1 if necessary). Then
give it to your students at the beginning of the year. You can also
include other areas, eg listening: How well students can understand
different people in different situations.
What can I do in English?
Mark which of the things below you can do like this:
a I can do this well
b I can do it
c I cant do it
Ask the teacher for help, eg about the meaning of a word
Ask the teacher for permission to do things, eg open the window
Say my name, address and telephone number.
Talk about what I like, eg music/hobbies
Talk about my family.
Tell the time.
Buy things in a shop.
Ask information about prices/times of trains or buses.
Describe people and animals.
Talk about what I do every day.
Talk about the weather.
Give and ask for directions.
Talk about what I did last weekend.
Talk about my last holidays.
Talk about my life in the past.
Give my opinions about films/TV/sport.
Give instructions.
Make suggestions to do things with friends.
Invite people to parties.
Agree and disagree with people.
Establishing objectives and goals
H aving thought about past learning experiences that students have had, it is a good
tim e to think about objectives for the com ing year. W hile this can also be done unit
by unit, the start of a course is a good tim e for learners to think about their
priorities and needs. A t the end of the course they w ill then be able to refer back to
them to assess w hether they have reached them or not. In this w ay aw areness of
initial goals is an im portant step in the developm ent of self-assessm ent.
The degree of learner participation in establishing objectives w ill depend on your
teaching situation. In m any secondary situations, the overall objectives w ill already
have been established (by the m inistry or the school departm ent). H ow ever, this
does not m ean that students cannot be consulted about objectives and given the
chance to think about their ow n needs. A gain this is particularly im portant in m ixed
ability situations and w here students can have the opportunity of w orking on their
ow n or in sm all groups.
H aving consulted students at the beginning of the year, their needs can be taken
into account in course planning throughout the year. For exam ple, if you find that
students are particularly interested in a certain topic it m ay be possible to relate a
task-based unit or project to it.
Setting objectives
Give your students a list with some of your objectives for the coming
year. Tell them individually to choose five priorities for themselves (eg
listening to stories/speaking about likes). If they like they can add
other objectives (particularly in terms of topic areas to cover).
They can hand these in to you individually or discuss their answers in
groups and then report them back to the rest of the class.
Objectives (elementary level)
Topic areas
nature and the environment, shopping, food
Speaking
using English in the classroom
talking about your own lives likes, homes, food
Listening
getting used to listening without understanding everything
listening to stories, conversations, cartoons
Reading
getting used to understanding the general idea
reading teenage magazines, comics, simplified readers using
dictionaries
Writing
planning and organising writing
writing postcards, penfriend letters
Grammar
revision and introduction of basic tenses: present simple
present continuous, past simple, future
Vocabulary
related to: classroom, animals, food
Pronunciation
difficult sounds
Learning
organising vocabulary, books, self-assessment
A s w ell as looking at overall course objectives, the objectives of particular m odules
or units can be looked at throughout the course. Before beginning a new m odule
you can inform students of the m ost im portant aim s and encourage them to think
about w hich are m ost useful to them . Then, w hen students have finished the
m odule they w ill be able to look back and decide to w hat degree the objectives
have been achieved.
Learning skills
The beginning of a course is also a good tim e for students to assess how effective
they are as learners and from there to think about how they could becom e better
learners. This involves thinking about both their attitudes tow ards learning and their
study habits.
Self-assessment of learning style
Look through the questionnaire below. Think about how you might
adapt it for your own students (translating it into their L1 if
necessary). Having given it to your students you can discuss the
answers.
Are you a good learner?
1 What is your attitude towards other countries?
a Im very interested in how other people live.
b Im quite interested in how other people live.
c Im only interested in where I live.
2 What is your attitude towards English?
a Im interested in listening to songs in English and watching films
and TV programmes in English.
b I only like songs and films from my own country.
3 For which of these things do you think English may be useful?
a For travelling to different countries.
b For helping to get me a job.
c For studying (at school or university).
4 Do you try to speak English in class?
a Always.
b Sometimes.
c Never.
5 What do you bring to class?
a My coursebook, vocabulary book and a notebook.
b Only my coursebook.
c Nothing.
6 What do you do when you find an important new word?
a I write it down on a piece of paper.
b I dont write it down, I say it to myself.
c I write it in my vocabulary book.
7 When you are listening to the cassette in class and you only
understand a little, how do you feel?
a Fine. I know that some listenings are more difficult than others,
I try to understand as much as I can.
b A bit frustrated but I continue listening.
c Very depressed. I stop listening.
8 When your are reading, which of these things do you do?
a If I see a new word I try to guess the meaning from the context.
If I cannot and the word is important, I use my dictionary.
b I ask the teacher what it means.
c I stop reading.
9 When you get written work back from the teacher, what do
you do?
a Look at the mistakes I have made and think about how I can
improve next time.
b I look through it quickly.
c I only look at the mark, then I throw it away.
10 Which of these things do you try to do outside the class?
a Listen to songs in English.
b Read stories in English.
c Speak in English when I meet foreign people.
d Watch films or TV in English.
e Think about any English words I see on advertisements and
labels.
O ne point to rem em ber w hen students are assessing their ow n attitudes and
learning habits is that there is not alw ays a correctansw er. For different learners
different learning strategies can be effective. For exam ple, if you are considering
how students learn vocabulary best you w ill find that som e strategies (like using
w ord netw orks or draw ing pictures) are suited to som e learners but not to others.
W hat is im portant is for students to assess how effective they are as learners and if
there are any w ays in w hich they can becom e m ore effective.
Learner diaries
A nother im portant step at this stage is to establish one of the m ost im portant
instrum ents in self-assessm ent, the learner diary. Learner diaries are records by
students of w hat has happened in their language learning over a period of tim e (for
instance a lesson or a w eek). They can be very detailed and include a lot of
inform ation, such as w hat w ent on in class, w hat students feel they have learned,
w hat problem s they have had and w hat they are going to do to overcom e them .
C om plex diaries are how ever som ew hat im practical for m ost secondary school
learners, m ainly due to restrictions of tim e. A com plex diary could take up the
m ajority of class tim e in m any situations. Therefore, it is perhaps better for diaries to
be sim ple and short, so that they can be filled in during a spare five or ten m inutes
at the end of a class. In this sim ple form , students can be asked to list the activities
they enjoyed and the problem s that they had. In this w ay, students are given a
rough record of the w eeks classes that they can refer to w hen thinking about their
ow n progress.
Look at the two learner diaries. Which one is most suitable for your
learners?
Simple learner diary
Complete your learner diary every week.
Example:
Week 1
Favourite activities animal guessing game / animal crossword
Problems listening to animal descriptions / new vocabulary
animals
Week 2
Favourite activities-
Problems-
Week 3
Favourite activities-
Problems-
Detailed learner diary
Answer these questions about the lesson:
What have I done? (list the activities)
Which activities were most useful to me?
Which activities were not so useful for me?
Did I participate in the lesson?
Did I work well with my group?
What did I learn?
What problems did I have?
A ll of the activities that w e have looked at in this section, as w ell as providing
general learner training, aim to m ake students reflect on their ow n learning and to
start to assess them selves. The am ount of tim e you spend on these aw areness
activities w ill depend on the tim e you have available. H ow ever, even one or tw o
classes focusing on learning and building up m echanism s for future self-assessm ent
like learner diaries, w ill help students to assess them selves throughout the course
and it is im portant to rem em ber that, if w e w ant our students to take som e
responsibility for their ow n learning, w e have to m ake it clear from the beginning.
3.3 Assessing perf ormance
H aving established the foundations for self-assessm ent in the early part of the course,
self-assessm ent then needs to be integrated into day to day classroom activities. The
assessm ent of studentsow n perform ance should becom e an integral part of
com m unicative activities. In this w ay learners continually keep track of their progress
and think about how they can im prove in the future.
Self-assessm ent can also give the teacher feedback about perform ance w ithout
having to correct every single activity done in the classroom . There are m any
occasions w hen students can get as m uch from correcting or assessing their ow n or
each others w ork, than from feedback from the teacher.and w hen w e do correct and
assess our studentsw ork, it is im portant that w e m ake the m ost of the feedback w e
are giving by getting students to think about their perform ance as w ell.
Reading and listening
Before w e ask students to assess their ow n perform ance in reading and listening, it is
useful to do learner training activities to m ake them aw are of different types of
reading and listening. It is vital for students to realise that often they do not have to
understand everything to do a reading or listening task. They also need to be aw are
that som e texts are m uch m ore difficult than others. If they do not realise this and
only see progress in a purely linear w ay, learners can becom e very frustrated.
For both reading and listening, useful aw areness activities are those in w hich students
grade text types according to difficulty or think about different types of reading and
listening, how m uch of a text they need to understand to do a task. H ere it can be
useful to get students to think about reading and listening in their ow n language as w ell.
Learner training (reading)
Think about these three learner training activities yourself. Then try
them out with your students in students L1 if necessary. Tell students
to think about reading in their language and in English.
1 Rank these reading texts in order of difficulty:
postcards from friends
newspaper articles about politics
comic strips
extracts from childrens fiction
encyclopaedia extracts
2 How much of a text do you need to understand to do the things
below? (eg 1 = 25%)
100% = all of it 75% = most of it 50% = the general idea
25% = some of it
Find out the main stories in a newspaper.
Read a comic story for pleasure.
Follow instructions to assemble a machine.
Get a fact from an encyclopaedia article.
Get an idea of what a magazine is like by flicking through it.
Read a holiday brochure and choose a holiday.
magazine articles about people
notices
instructions for machines
travel brochures
letters from friends
Likew ise for listening it is useful for students to think about som e of the variables
that can affect understanding such as accent, background noise, visual clues and
num ber of speakers. A nd as w ith reading it is useful to think about how m uch w e
need to understand w hen w e are doing different kinds of listening.
Learner training (listening)
Look at the learner training activity for reading and produce a similar
activity for listening.
Ranking different listening situations (eg listening to your teacher
telling a story / listening to a conversation between three native
speakers).
Working out how much you need to understand to do different
things (eg listening to find the main points of the news = 50%).
W hen students are actually doing reading or listening tasks, the m ost obvious w ay
of assessing perform ance is by assessing com pletion of the task. For m ost tasks this
m eans w orking out how m any correct answ ers the student gave and w hy the
answ ers are right and w rong. Students can either do this through self-correction
them selves or other students can correct their answ ers. They can then keep a record
of their ow n results, possibly in their learner diaries.
If these m arks carry w eight in overall assessm ent, this procedure is not totally
reliable as students could be tem pted to cheat. H ow ever, if this is not the case,
students have no need to m islead you about their ow n results. A nsw ers can be gone
over w ith the w hole class and you can elicit from students w hy certain answ ers are
right or w rong. W hen discussing results in class it is very im portant to be positive
and concentrate on w hat students achieved, rather than on w hat they got w rong. It
is also im portant not to m ake students say their ow n results in public, to avoid
highlighting failure and encouraging com petition. A nother alternative is for students
to hand in tasks for you to look at after the class.
In addition to this kind of correction, it can be very useful to establish listening and
reading records, w hich can com plem ent the studentslearner diaries. A sim ple w ay
of doing this is to give students a table or graph, w here they assess w hat proportion
of a text they understood each tim e they read or listened to it. For exam ple, the first
tim e they m ay understand 20% of a text (I could understand w hat it w as about
m ore or less). The second tim e they m ay understand 50% ( I understood the
situation quite w ell, but I m issed a lot of details). The advantages of these records is
that they are easy to fill in and students can keep track of their ow n perform ance in
listening and reading. A t the sam e tim e students can also think about how w ell they
com pleted the task. They can thus be m ade aw are of the fact that to do a task they
do not need to understand everything.
Look at the example of a listening/reading record. Listen to something
in English yourself and complete it. Then copy the record sheets and
try out this activity with your students while doing a listening or
reading activity (the descriptions of levels of understanding can be in
students L1 if necessary).
With difficult texts you may find that students only understand 25%
the first time but up to 65% the second time. However, unless the
tasks are too difficult, they should have been able to complete the
tasks for both first and second listening.
Use these scales to complete the listening/reading graphs below.
At the same time say if you managed to complete the task.
100% = I understood absolutely everything
90% = I understood nearly everything. I only missed a few words.
75% = I understood most of it, but I didnt understand a few
sentences.
50% = I understood the situation quite well, though I missed a lot of
the details.
35% = I got the general idea, but at times I got a bit lost.
20% = I could understand what it was about more or less.
10% = I understood a few words here or there
0% = I didnt understand a single word
Listening/reading record
Date................ Kind of listening/reading:.................................
First time Second time
%
100
90
75
50
35
20
10
0
Task completed: yes/no Task completed: yes/no
Writing
It is difficult to assess ones ow n w ritten production; therefore students do need
correction and assessm ent of their w ork by the teacher. H ow ever, this is not
enough. A s m entioned earlier, w hen teachers give back corrected and assessed
w ork, students often just look at the m ark and ignore the rest. Thus a lot of w ork
on the teachers part is w asted. If learners are to im prove their ow n w riting they
need to think about their ow n perform ance and then com pare their judgem ent w ith
that of the teacher. They also need to try to correct their ow n m istakes and to think
about them , so that in future they w ill be able to avoid them .
It is useful to do learner training activities that m ake students aw are of the various
stages in the w riting process. It is particularly im portant for students to realise the
need to check and assess their ow n w ork w hen they have produced a draft. If there
is tim e they can then go on to w rite a final and corrected draft. O ne w ay of
developing your studentsability to assess them selves is by getting them firstly to
assess texts and then to edit them . In this w ay they are m ade aw are of som e of the
criteria involved in assessm ent.W hen you do w riting activities in class, the first thing
to do is to establish clear criteria for assessm ent before students have started to do
a task. D epending on the level and the kind of w riting task, criteria could include
the follow ing:
N interest and relevance of content N organisation of ideas
N appropriacy of style N linking and punctuation
N presentation N gram m atical accuracy and punctuation
A n alternative, w hich can be particularly useful for project w ork, is to negotiate
criteria w ith students. You can ask students how they think the project should be
m arked and then w ork out a sim ple m arking schem e on the board. This activity
provides the vital function of m aking students aw are of w hat constitutes a
successful piece of w riting. It also involves students in the assessm ent process.
The next step is to m ake sure that w hen students have finished a draft of their
w ork, they check over it, try to correct any sim ple m istakes they have m ade and
assess the strengths and w eaknesses of the w riting. It can be a good idea to m ake
students show you an edited version of a first draft before they go on to the final
draft. A t this stage, peer assessm ent of w ork is also very useful. Students can pass
their com positions to their partners, w ho check through it for m istakes and assess it.
In the case of group w riting, students can check the w riting done by other groups.
H aving produced a final version you can also ask students to m ake a final
assessm ent of their w ork, assigning a m ark according to the criteria agreed at the
beginning. O nce again, other students can assess the w ork and give a m ark. This
can be done by displaying w ork around the classroom or by passing the pieces of
w riting around the class. A t first students find this difficult and m ost tend to give
other students high m arks, as they do not think that it is up to them to criticise their
classm ates and friends. H ow ever, having done self and peer assessm ent for a w hile,
m arks tend to be m ore realistic.
The next stage is assessm ent by the teacher according to the agreed criteria. The
teachers m ark is then added to those of the w riter and of other students. A t the
sam e tim e the teacher should provide feedback about the problem areas in the
com positions. The best w ay of doing this is through the use of correction codes.
W hen the com positions are given back to students see if they agree w ith your
assessm ent. If they can argue coherently that they deserve m ore m arks for one of
the factors considered and you feel that they are justified, you can change the m ark.
In the case of group w ork, the m arks can be discussed w ith the w hole class. You can
explain w hy you gave the m ark you did and alter it if the students give convincing
reasons. W hile this process should not degenerate into a bargaining about m arks, it
is im portant to take studentsopinions into consideration and to show them that
the teacher is not alw ays right. This m ay seem to underm ine the teachers
authority, but this is only the case if it is not part of a general approach.
Look at these procedures for carrying out self assessment of writing
(described in the preceding paragraphs). Try out one or more stages
with a class. While some of the stages are best suited to higher level
learners (eg negotiating criteria), other stages (like stage 2- checking
and editing) can be done with low level students.
Stage 1: Negotiating criteria (especially good for projects)
Before beginning a writing activity, outline or negotiate criteria for
assessing finished work. Write the criteria on the board. One or a
combination of these criteria can be used:
presentation
handwriting/drawings etc
content/message
interesting/relevant to situation
organisation
format/paragraphs/organisation of ideas
linking
use of linking words and expressions/punctuation
vocabulary
adequacy of vocabulary for purpose
accuracy
grammatical inaccuracies/spelling/vocabulary
Decide how many points to be given for each area:
Example:
content /10 accuracy /10
Stage 2: Checking and editing
When students have finished writing their first draft, tell them to show
it to you. Then they check it for mistakes and give it to their partner to
edit.
Stage 3: Final self-assessment
When they have finished the final draft, ask students to assess their
own work according to the established criteria.
Stage 4: Final peer-assessment
Display written work around the classroom. Students circulate and
write their assessments of each others work on forms or pieces of
paper provided. At the same time, go around yourself and assess
work, using the same criteria.
Stage 5: Comparing assessments (better for group writing)
Choose a few pieces of work and compare the different assessments.
Discuss differences. Dont be afraid to modify your assessment if
students give convincing arguments in favour of a different mark.
Speaking
Speaking is another area w here it is very difficult to assess ones ow n perform ance.
H ow ever, a sim ilar approach can be taken as for w riting. A s this cannot be done for
every speaking activity that students do in class, it is best to focus on activities
w here perform ance is an im portant elem ent, for exam ple roleplay or sim ulation
rather than survey activities (w here the inform ation is m ore im portant). It is also
advisable to focus on group or pair perform ance rather than on individual
perform ance, w here peer assessm ent is involved. C riticism of individual perform ance
(how ever positive) can be very dam aging if it com es from fellow classm ates.
A s for w riting, it is im portant to establish criteria for assessm ent before students
begin the task: fluency and lack of hesitation, relevance and interest of the
perform ance, pronunciation: sounds/rhythm /w ord and sentence stress/intonation,
appropriacy of language, gram m atical accuracy and use of suitable vocabulary. A t
low er levels students can be encouraged to think about only one or tw o of these
criteria, in particular interest of perform ance and gram m atical accuracy. A t higher
levels all of these elem ents need to be considered at one stage or another.
Perhaps before you get students to think about criteria for assessing them selves, it is
useful to m ake them aw are of them in general term s. The follow ing learner training
activity helps students to think about w hat is involved in successful com m unication.
It also gives them m odels for oral self-assessm ent.
Oral self-assessment
Do this activity yourself. Then try it out with your students.
a Look at the two self-assessments below, with two people assessing
how well they performed in an oral exam (a shopping roleplay).
Which of them do you think did best in the oral interview?
b Which of these areas do they mention in their self-assessments?
pronunciation
grammatical mistakes
vocabulary
fluency and hesitation
getting the message across
using the appropriate language (in terms of formality) for the
situation
A I made very few grammatical mistakes, but I spoke very slowly.
Sometimes I couldnt think of the word and had to stop completely.
My pronunciation was not too bad, but the other person did not
understand one or two words and I had to repeat them. I only bought
three of the six things I wanted to get.
B I made a lot of grammatical mistakes, but the shop assistant was
able to understand me. I think my pronunciation was not very good
and some words were wrong. I spoke at normal speed and when I
couldnt think of a word in English I described what it was. I bought
what I wanted.
Check your answers on page 88.
A n im portant point about self-assessm ent of speaking is that rather than asking
students to give them selves m arks, the m ain thing is for students to think about
their problem s and their achievem ents.
Peer assessm ent should be done during the activity, so that students perform ing the
activity get feedback im m ediately afterw ards. This can either be done in groups or
w ith the w hole class. For exam ple, in groups of four, tw o students can act out
roleplays w hich they have previously practised. The other pair can then give them
feedback about their perform ance. W hen students act out roleplays or sim ulations
in groups, they can then act them out in front of the class and the other students
can assess the overall perform ance of the group.
Self-assessm ent of perform ance can be done im m ediately after an activity has
finished. Students can assess how they perform ed, either in very general term s or in
term s of fluency, accuracy, pronunciation etc. A udio and video taping is of particular
use here, as students can go back and see or listen to their ow n perform ance. These
recordings can also be used to com pare perform ance at different stages of the
course, giving students concrete evidence of their progress in speaking.
It is im portant to com plem ent self-assessm ent of speaking w ith your ow n
observation and feedback, pointing out problem areas and encouraging students on
w hat they have achieved. A ssessm ent can also be linked w ith learner training
activities. H aving assessed perform ance in an activity, strategies for im proving
com m unication can then be discussed, eg avoiding w ords that they do not know ,
using gesture and m im e, using w ords and expressions to hesitate.
3.4 Reviewing progress
In addition to building self-assessm ent activities into everyday classroom practice, it
is im portant for students to look back and review their ow n progress over a period
of tim e. This is w hen self-assessm ent can be system atically integrated into an overall
assessm ent program m e, and linked to the assessm ent done by the teacher.
H ow ever, review ing ones progress is not an easy task. If students are asked to
assess their progress over the past year, w ithout any training or preparation, it is
unlikely that they w ill be able to do it very w ell. Therefore, review ing progress needs
to be a gradual and cum ulative process, building up to a final, global assessm ent of
progress over a course.
A practical m odel for carrying out self-assessm ent is suggested:
4 Students use thei r answers to the
end of term questi onnai re to make
fi nal concl usi ons
3 Students use thei r progress reports
to assess the terms progress
2 At peri odi c i nterval s, students
revi ew thei r own progress
1 Students keep records of what
happens i n cl ass and how they are
doi ng
CONCLUSIONS ABOUT PROGRESS
END OF TERM QUESTIONNAIRE
PROGRESS PROGRESS PROGRESS
REPORT 1 REPORT 2 REPORT 3
1 /2 /3 /4 /5 1 /2 /3 /4 /5 1 /2 /3 /4 /5
WEEKLY DIARY WEEKLY DIARY WEEKLY DIARY
This involves a pyram id approach, w here learners use all the records that they have
been building up. Firstly, the learner diaries and any other records they m ay have (eg
reading records) can help them com plete the periodic progress questionnaires. These
questionnaires can be done at the end of a unit or m odule of a textbook, at the end
of a project or coinciding w ith institutional assessm ent (w hen the teacher has to
produce som e kind of report on studentsprogress). In turn these progress
questionnaires can be used by learners to help them selves assess their progress over
a school term or the duration of a course.
Revising
Revising is norm ally associated w ith studying for exam s, though this does not have
to be the case. Revision activities involve going over w ork done, identifying
problem s and w eaknesses and learning or reinforcing language that has been
studied. Thus at this stage, there needs to be an im portant elem ent of self-
assessm ent, at least if revision is to be efficient.
Test yourselfor self checkactivities are one w ay in w hich students can check their
learning of language and then focus on the problem areas they have. These are now
becom ing com m on in textbooks, and are basically tests w hich students can do and
m ark them selves. Self check activities norm ally focus on the gram m ar studied in a
particular unit, but they can also check vocabulary and pronunciation. The form ats
of these activities are very sim ilar to those used in form al tests. In this w ay students
get practice at doing tasks w hich they w ill do in exam s. Task types such as editing,
gap-filling, m ultiple choice, w ord sequencing and classification of lexis can all be
used. O ne fairly sim ple w ay of creating these activities is to com pile a list of several
com m on m istakes that your students have been m aking w ith target structures.
Students then have to correct them .
These activities can either be done in class or for hom ew ork. It is how ever useful to
go through them w ith the w hole class. In this w ay you can get an idea of w hat
problem s students still have, and w hat rem edial w ork is necessary. The private
nature of these activities is im portant, as if the teacher collects the results they can
becom e tests. But it is im portant for them to be done w ithout outside help. If
students copy or look at the answ ers, they are cheating them selves about their ow n
progress. If the activity is done properly, students can use the results from self
checkactivities to help them com plete the progress questionnaires.
Self check activity
Look at this activity for beginners. Create a similar activity for your
students and give it to them when you have finished a unit or module.
Self check
Do the exercises. Write your total here: /15
1 Correct these sentences:
Example:
My favourite singer are Phil Collins.
My favourite singer is Phil Collins
a I has got two brothers.
b Where you are from?
c What is yours name?
d Have got they a television?
e Kangaroos is from Australia.
2 Complete the description with these words:
have
is not
is
are
am
has
have
Example:
My name..is..Emily and I ....... from Canada. I .........
got one sister and two brothers. Their names .......... Edith, Sam and
Bill. We ......... got a dog called Hegel and two cats called Kierke and
Garde. Hegel is a labrador and he ......... got black fur. He ..........
very intelligent!
3 Look at the family tree. Complete the sentences.
a John is Emilys ..............
b Susan is her.............
c Bill is her............
d Tom is her.............
O ther revision activities w ith a self-assessm ent com ponent involve students looking
back at their w ork and identifying problem areas. For exam ple, students can look
through their w ritten w ork and identify the five m ost com m on m istakes they m ade.
In groups they can then w ork out w hich m istakes are m ost com m on and report
them to the rest of the class. M istakes can then be w ritten up on the board and all
the students have to correct them . W ith lexis, students can look through their w ork
and identify w ords they find difficult to rem em ber. Sim ilarly w ith pronunciation,
students can look back and select five w ords w hich they have difficulty
pronouncing.
Tests
A s w ell as com paring m arks from form al assessm ent w ith studentsow n assessm ent
of their progress over a period of tim e, it is im portant to include an elem ent of self-
assessm ent w hen giving feedback about tests. This can be done by giving students
questionnaires to com plete, in order to m ake them think about their perform ance in
the test. This can also give you useful feedback about your ow n test.
Fred A nne
John Susan Tom
Em ily Sam Bill
Look at these test follow-up questionnaires. Adapt them and then try
them out with your students after administering a test. Alternatively,
think of a test follow-up activity for a different kind of test,
eg grammar/vocabulary.
Test follow-up Reading/listening
How many questions did you get right?
If you had problems, which of these did you have?
a you had not seen or heard anything like this before
b there was a lot of difficult vocabulary
c (for listening) they spoke too quickly
d (for listening) the sound quality was bad
e the questions were confusing
Why did you get the questions wrong?
a you did not understand the whole text
b you did not understand a word
c you did not understand a structure
d you answered a question without thinking
e you did not have enough time to answer a question
f you did not understand the instructions
Do you think the result reflects your level?
What can you do to improve your listening/reading in the future?
A s w ell as getting students to assess their perform ance in tests, you can get
students to w rite the tests them selves, especially tests of gram m ar and vocabulary.
N ot only is this a good revision activity for students, but it m akes them aw are of
different test form ats. This can be done in groups, w ith one group w riting a test for
another group. Student-generated tests can also be used in form al assessm ent in
the follow ing w ay. First, give students a list of the kind of tasks they can use (eg
gap-filling/w ord sequencing). Then, individually or in pairs students w rite their test
and answ er sheet. H aving collected in all the tests, you can select the best item s and
w rite them up on the board. This is the test, w hich is then adm inistered and m arked
as any other test. The advantage of this activity is that students feel m uch m ore
identified w ith the assessm ent process. A nd w hile one m ight expect students to
w rite very easy item s, this does not happen. In fact the opposite tends to be the case.
Progress questionnaires
Progress questionnaires give learners an opportunity to review their learning over
the past few w eeks or m onths. To be able to do this they should use their learner
diaries and their listening and reading records. They can also look through their
w ritten w ork: their com positions, notebook and vocabulary book. They also need to
consider their results in self checkactivities and tests. The progress questionnaire
pulls together all these different threads to try to give students a global picture of
their ow n progress.
Like learner diaries, the com plexity of such questionnaires can vary enorm ously,
depending on tim e available as w ell as studentsage.
Look at the progress questionnaires. Which one is most suitable for
your students?
Questionnaire 1
Think about your progress this term. Give yourself a mark out of ten
for these areas:
speaking /10 pronunciation /10
listening /10 grammar /10
reading /10 vocabulary /10
writing /10
Questionnaire 2
Use your learner diary to complete the tables with areas that you have
covered this term (see examples). Also look at the textbook and your
notebook to help you.
Then grade yourself using the scales provided.
Structures studied Grade
present continuous 4
5 I have no problems with this structure and I never make mistakes.
4 I only make mistakes occasionally
3 I sometimes have problems using it
2 I dont understand it very well and I make a lot of mistakes
1 I dont understand it at all
Vocabulary area Grade
food and drink 3
5 I have learnt a lot of words
4 I have learnt quite a few words
3 I have learnt some new words
2 I have learnt a few new words
1 I have not learnt any new words
Speaking activities in class Grade
asking for tourist information 2
5 I can do this very well, without hesitating and making mistakes
4 I can do this quite well, but I sometimes hesitate and make mistakes
3 I can do this, but I speak slowly and make quite a few mistakes
2 I cannot do this very well: I get stuck and cant think of what to say
1 I cannot do this at all
Writing activities done Grade
letter to penfriend 6
Grade yourself out of ten. Think about these things:
presentation
content
organisation
punctuation
grammar
vocabulary
Reading/listening activities Grade
reading stories 3
5 I did the task with no problems at all
4 I had a few problems
3 I answered half of the questions correctly
2 I only answered a few of the questions
1 I could not do the task
Grade these things (1-5) related to effort and attitude:
participation in class
use of English in class
homework and projects
working in groups
planning and working on my own
Questionnaire 3
Tick which of these things you can do now ().
Put two ticks if you can do it very well ()
Put a cross if you cant do it ()
a I can talk about what I did yesterday.
b I can ask other people about what they did yesterday.
c I can write about what I did yesterday.
d I can give and ask for directions.
e I can understand simple directions.
f I can write simple directions.
g I can talk about how I get to and from school.
h I can understand a description of a place.
i I can write a description about a place.
j I can describe places.
W hen your students have finished progress questionnaires you can look at them
yourself. If they cover the period of a school term you can link them in w ith term ly
assessm ent. If not, they should be kept along w ith learnersother records until they
can be used for global assessm ent of progress.
3.5 Results f rom self -assessment
H aving looked at w ays of carrying out self-assessm ent, w e need to think about how
w e can use the results from it and how our studentsassessm ent of them selves can
be linked in w ith our assessm ent of them . This can be done initially w hen students
com plete the periodic progress questionnaires. By collecting these in and looking at
them , w e can find out how students feel about their ow n progress and com pare
them w ith our ow n results. If there is a big difference betw een the tw o there is
obviously a problem , w hich needs to be sorted out by talking to the student in
question.
H ow ever, the logical tim e to link self-assessm ent w ith teacher assessm ent is at the
end of a term or course. Students can use the progress questionnaires and other
records to m ake a global assessm ent of their ow n progress. This should run parallel
to the assessm ent you do as a teacher. If possible, learners should use the sam e
report or m ark sheet that you use. In this w ay the report done by the student can be
directly com pared w ith a draft of the teachers report.
If there is tim e, a counselling session can enable you and students to discuss and
com pare the tw o global assessm ents. Firstly, the com m ents on the students final
report can be negotiated. Secondly, it is possible to discuss the m arks. The students
m arks from teacher assessm ent can be com pared w ith his/her ow n self-assessm ent.
In m ost cases you w ill find that student m arks w ill be the sam e as yours or even
low er. W hen their m arks are higher than yours you can ask students to say w hy they
think this is the case. It is possible for students to give convincing reasons for this:,
eg I did very badly on one of the tests because I was very nervous, but I dont think
the mark reflects my real level.O n such occasions it can be w orthw hile m odifying
his/her m ark upw ards.
It is very im portant that these counselling sessions do not degenerate into
bargaining sessions about m arks. W hen a student has no reasons to support his/her
argum ents for a higher m ark, the m ark should stay the sam e. But a certain am ount
of flexibility enables students to feel that their self-assessm ent is taken seriously and
that they are being involved in the assessm ent process. In the great m ajority of cases
students respond to this by being honest about their progress.
Look at the following procedures for linking end of term/course self-
assessment with your own assessment. Would you do it in the same
way or would you change anything?
Work out a form to put students marks and your own comments on
(comments can be in students L1). Then complete it for each
student.
Example:
Name Ana Gomez Garca
Listening 7
Speaking 5
Reading 8
Writing 7
Grammar 5
Comments
Ana has worked well this year, especially since Christmas. She still
lacks confidence in speaking but she has a good level of
understanding. She is still weak in some grammar areas and she
needs to revise them thoroughly. However, her compositions have
improved and her stories have been excellent and very imaginative. A
good years work!
Give blank copies of the form to your students. Tell them to use their
progress reviews to fill it in. Ask them to write at least a sentence
about their progress and what they should do to improve next course.
Give the class some puzzles, word games or other activities where they
can work on their own. Then call out students one by one. Briefly
compare the two reports (yours and the students). Decide on the final
marks and comments. Dont be afraid to change your mark if you feel
the student is right.
Tim e is alw ays a problem , but counselling can be done in the last few days of term
after all form al tests have been given and w hen there is no pressure to cover course
m aterial. A ctivities such as puzzles or reading can enable you to call students to the
front of the class w hile the others are getting on w ith w ork on their ow n.
W hen a student has a very different m ark from yours, it is probably a clear sign that
he/she is unable to assess his/her ow n progress. This m ay be due to a lack of
m aturity in outlook or to a real inability to m easure progress. In both cases the
inability to carry out self-assessm ent provides clear evidence of problem s of attitude.
Thus, self-assessm ent itself provides you as a teacher w ith im portant data for
assessing non-linguistic factors.
C ounselling as m entioned here before can obviously be very tim e consum ing.
H ow ever, it can be done w hile students are doing som e other activity like reading or
doing a project in groups. If at all possible all students should have som e kind of
direct contact w ith the teacher, even if it is very brief. This can be either done in the
studentsL1 or in English w here a students level perm its.
In conclusion, this system atic approach to self-assessm ent builds up to a direct
participation of students in the assessm ent process. A s w ell as providing valuable
data for the student, it develops the skill of self-assessm ent w hich is not only
valuable in language learning. The ability to carry out self-assessm ent should be a
broad educational objective at secondary level.
Suggested answers oral self -assessment
a The second person did m uch better as he/she w as m uch m ore fluent and achieved
com m unication, even though he/she m ade m ore m istakes.
b Both of them m ention all of the things except for appropriacy of language.
Action points: self -assessment
N A s a result of reading this chapter are you going to im plem ent any of the ideas
about self-assessm ent suggested here? W hy/w hy not?
N If you are going to introduce self-assessm ent w ith your students, w hen are you
going to do it?
a at the start of a course
b during norm al classroom activities
c w hen you are revising or review ing a unit or m odule of your course
N W hich self-assessm ent activities are you going to try out?
Example:
learner diaries
N A re you going to link self-assessm ent w ith any other learner training activities?
Example:
discussing reading strategies
N A re you going to take studentsself-assessm ent into consideration w hen
w orking out your ow n final assessm ent? If so, how are you going to do this?
N A re you going to take into account the view s that your studentsexpress during
self-assessm ent activities? (For exam ple, in term s of learning preferences,
activities, m aterials etc)
Glossary of terms used in the handbook
accountability
the responsibility w e have as teachers to be able to explain the rationale behind our
assessm ent techniques and results to students, parents and institutions
administer tests
the com plete process of giving a test including giving out and collection of papers
and invigilating or conducting an oral interview
analytic rating scale
scale for rating perform ance or progress in w hich different activities are divided into
constituent parts and a different band is produced for each activity
assessment
m easuring our studentsperform ance in any one of m any different w ays, diagnosing
the problem s and m easuring the progress students m ake
band scale
level of perform ance in a rating scale w hich describes w hat a student has achieved
in a test
competence
level of perform ance in the language w hich our students are actually capable of
correction code
code often used by teachers to signal student errors in w ritten w ork
counselling
individual m eetings betw een students and teachers to discuss assessm ent results
usually in relation to self-assessm ent
criterion/criteria
descriptions of w hat our students should be able to do w ith the language
criterion-referencing
using descriptions of w hat our students should be able to do w ith the language in
order to determ ine the pass score in a test or inform al assessm ent
descriptor
definition of a level of perform ance in a band scale
diagnostic questionnaires
learner questionnaire used to find out w hat are our studentsproblem areas w ith
the language; usually given at the beginning of a course
diagnostic test
type of test used to find out w hat are our studentsproblem areas w ith the
language; m any progress tests have a diagnostic elem ent
discrete item test format
test form at in w hich there are usually m any item s requiring short answ ers
editing
checking our studentsw ritten w ork and correcting m istakes
entry/placement test
test w hich w ill indicate at w hich level a learner w ill learn m ost effectively in the case
of different levels or stream s
G lossary
evaluation
consideration of all the factors that influence the learning process such as syllabus
objectives, course design, m aterials, m ethodology, teachers and assessm ent
examination
form al sum m ative or proficiency test usually adm inistered by an institution; often
associated w ith tests supplied by exam ination boards
formal assessment
tests given under conditions w hich ensure the assessm ent of individual perform ance
in any given area
format
test form ats are the tasks and activities w hich students are required to do (eg
m ultiple choice)
formative assessment
type of assessm ent w hich feeds back into learning and gives the learner inform ation
on his/her progress throughout a course thus helping him /her to be a m ore efficient
learner
grade
w ay of expressing overall results using a num ber or letter
holistic rating scale
scale in w hich different activities are included over several bands to produce a m ulti-
activity scale
impression mark
num ber or letter given by a teacher to studentsw ork as a result of inform al
observation w ithout using a rating scale
informal assessment
system for observation and collection of data about studentsperform ance under
norm al classroom conditions
integrative test format
type of test form at w hich involves the use of m ore than one skill by students and
w hich is open-ended involving com m unication and interaction
inter-rater reliability
w ay of describing to w hat extent different raters or teachers assess perform ance in a
test in the sam e w ay
interlocutor
teacher or other trained person w ho during a test acts as the person w ith w hom the
student or candidate interacts in order to com plete a speaking task
intra-rater reliability
w ay of describing to w hat extent the sam e rater or teacher assesses perform ance in
a test in the sam e w ay
item
individual question in a test w hich requires the student or candidate to produce an
answ er
learner training
w ays of helping learners to find strategies to learn m ore effectively; these strategies
should suit their individual learning style
G lossary
learner diary
record of studentslearning experiences containing w hat they have done in class,
the progress they have m ade and any problem s they have
learning strategies
w ays of organising learning w hich help students to learn m ore effectively
learning styles/preferences
different w ays of learning w hich learners em ploy to achieve their objectives
linguistic factors
aspects affecting assessm ent w hich are strictly to do w ith the language such as the
four skills
lockstep
situation in w hich all students in a class are engaged in the sam e activity at the
sam e tim e, all progressing through a task at the sam e rate
monitoring
observing and m aking of assessm ents of w hat is happening in the classroom during
learning activities
negotiated syllabus
studentsneeds and learning preferences are taken into account during a course;
these needs w ill have been discussed by teachers and students together
non-linguistic factors
aspects affecting assessm ent w hich are not to do w ith language per se but are m ore
connected to other factors such as attitude, w orking w ithin groups and co-
operation
norm-referencing
listing students in order of test results and passing them or failing them according to
their position on the list
objective marking
w here only one answ er is possible and this is given in an answ er key; possible to be
interpreted by all m arkers
open-ended test format
test form at that requires no specific response, but w hich is open to interpretation
paralinguistic
aspects of com m unication w hich are outside the scope of the spoken w ord such as
gestures or expressions
peer assessment
w here students assess one another during class activities
peer editing
w here checking of studentsw ritten w ork and correction of m istakes is carried out
by other students
peer monitoring
w here observation and assessm ents of w hat is happening in the classroom during
learning activities is carried out for students by their class colleagues
performance
how our students did in a form al or inform al assessm ent procedure regardless of
their actual com petence; perform ance m ay be low er than com petence
G lossary
practicality
all aspects concerning tests w hich affect tim e and resources
proficiency tests
type of test w hich aim s to describe w hat a student is capable of doing in a foreign
language; usually supplied by external exam ination boards
profile
w ritten description of studentsperform ance or progress; often used in reporting
assessm ent results
progress questionnaire
learner questionnaire in w hich students reflect on their ow n progress over a given
period of study
progress tests
type of test w hich aim s to find out how w ell students have grasped the learning
objectives over a particular period of tim e such as a m onth, a term or a year or over
a num ber of course m odules
rating criteria
the aspects of perform ance in language or non-linguistic factors on w hich teachers
w ish to assess their students
rating
assessing student perform ance using pre-established scales; usually assessing spoken
perform ance
raw score
num ber of correct answ ers obtained by a student in a test; from this score the final
grade is often calculated
reliability
the consistency of any form of assessm ent w hich m eans that under the sam e
conditions and w ith the sam e student perform ance the assessm ent procedure w ould
produce the sam e results
report
docum ent w hich describes studentsprogress and perform ance
reporting
the process of com m unicating assessm ent results to students, their parents and the
institution; usually through w ritten reports
rubric
instructions in a test or any classroom activity w hich indicate to the candidate or
student w hat he/she has to do to com plete any given task
sample
the am ount of language and content from syllabus plans or teaching records w hich
a test or any classroom activity elicits
scanning
reading a test quickly in order to obtain specific inform ation
self-assessment
assessm ent carried out by students them selves designed to m easure their ow n
perform ance and progress
G lossary
self-check activity
w here students check their ow n perform ance by com pleting an exercise and then
looking at their ow n results
self-editing
w here on com pletion of a w ritten piece of w ork, students go through the piece and
check them selves for any m istakes
self-monitoring
w here students correct their ow n speech production either at the tim e of speaking
or w hen listening to a recorded sam ple of their perform ance
skimming
reading a text quickly in order to obtain a general idea of the content area
standardisation
agreem ent betw een raters of student perform ance on the m eaning and
interpretation of criteria used for assessm ent
subjective marking
w here the m ark or grade given to a perform ance depends on som ebodys opinion or
judgem ent such as in all speaking tests
summative assessment
type of assessm ent w hich aim s to m easure studentsperform ance at the end of a
period of study
summative test
type of test usually adm inistered at the end of courses; often used as a w ay of
deciding w hether students m ove to a higher level or not or obtain a particular
certificate or not
test format
description of the task type w hich is used to elicit any given language sam ple from
candidates and students
test type
overall description of a test in term s of the purpose it serves and its objectivity or
subjectivity
test
any form of form al assessm ent in any language area w hich is adm inistered under
conditions w hich ensure m easurem ent of individual perform ance in any given area
validity
w here a form of assessm ent effectively m easures w hat it intends to m easure and not
som ething different
washback effect
the influence of tests or exam inations on the teaching and learning leading up to
the assessm ent
weighting
the relative im portance of different skills and language w hich is assigned in the
assessm ent process
BIBLIOGRAPHY
Baker, D . Language Testing: A Critical Survey and Practical Guide (1989). U K: Edw ard A rnold.
Black, H .B. & W .B. D ockrell, Diagnostic Assessment in Secondary Schools. A Teachers Handbook (1980).
Edinburgh: Scottish C ouncil for Research in Education.
Brindley, G . Assessing Achievement in the Learner-centred Curriculum (1989). Sydney: N ational C entre for
English Language Teaching and Research.
C arroll, B.J. Testing Communicative Performance (1980). U K: Pergam on.
C arroll, B.J & P.J. H all, Make your own language tests: a practical guide to writing language performance
tests. (1985). O xford: Pergam on.
D avies, A . Principles of Language Testing (1990). U K: Blackw ell.
H eaton, J.B. Writing English Language Tests (1988). U K: Longm an.
H eckhausen, N . The Anatomy of Achievement Motivation (1967). N ew York: A cadem ic Press.
H ughes, A . Testing for Language Teachers (1989). C am bridge: C U P.
N unan, D . The Learner-centred Curriculum( 1988). C am bridge: C U P.
O ller, J.W . Jr. Language Tests at School (1979). U K: Longm an.
O skarsson, M . Approaches to Self-assessment in Foreign Language Learning (1978). O xford:
Pergam on Press.
O skarsson, M . Subjective and objective assessm ent of foreign language perform ance, (1981) in Read,
J.A .S. (ed.) 1981: 225-39.
Read, J.A .S. (ed.). D irections in language testing (1981). SEA M EO Regional Language C entre,
A nthology Series 9.
Sum ner, R. The Role of Testing in Schools (1987). U K: N FER-N elson.
W eir, C .J. Communicative Language Testing (1990). U SA : Prentice-H all.

Assessment

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Assessment

Uploaded by

Copyright:

Available Formats

Handbooks for the Engl i sh Cl assr oom

You might also like