ELT METHODS AND

PRACTICES

TESTING AND ASSESSMENT

IS TESTING A ‘GOOD’ OR A ‘BAD’ THING?
(1/2)

 Language educators (and not only) are

divided into two camps: the teachers and
the testers who meet in battle.
 Teachers often say things like:
Let's learn to teach before we learn
to test.
We deal with people, you deal with
statistics.
IS TESTING A ‘GOOD’ OR A ‘BAD’ THING?
(2/2)

Testers think that teachers:

tend to be unspecific about their
aims and objectives.
Are disinterested in finding out
whether goals and objectives have
been met.
CAN WE DO WITHOUT TEACHING OR
TESTING?

Probably yes because:

 learning can occur in spite of teaching
and/or testing, despite any kind of
formal evaluation.
 The outcomes of teaching can be assessed
without any form of testing.
 Testing may be used to measure what
people already know.
IS TESTING SYNONYMOUS WITH THE
TERMS BELOW? (1/2)

Evaluation:
 Evaluation may focus on the effectiveness or
impact of a program of instruction,
examination or project.
 Students are usually not asked to evaluate
while teachers carry out or take part in
evaluation only in some contexts.
 ‘Experts’ or the authorities are most
commonly legitimized to carry out formal
evaluation.
IS TESTING SYNONYMOUS WITH THE
TERMS BELOW? (2/2)

 Measurement:
 Measurement is the process of determining
the amount or length of something when
compared with a fixed unit (e.g. using a ruler
to measure length).
 In language teaching measurement
constitutes the quantification of language
proficiency. Aspects of language knowledge,
specific abilities, and skills are measurable
when there are transparent criteria and
precise analysis of data.
IS ASSESSMENT SYNONYMOUS WITH
TESTING? (1/2)

 No, it is not.
 Assessment is a more encompassing term than
testing.
 It is the process of gathering, interpreting, and
sometimes recording and using information
about students' responses to an educational
task in order to provide the next learning step.
IS ASSESSMENT SYNONYMOUS WITH
TESTING? (2/2)

 Assessment is primarily concerned with

providing teachers and/or students with
feedback information.
 In language teaching, it is a local or global
procedure through which one can appraise one
or more aspects of language proficiency.
 Assessment is transparent when clear
assessment criteria have been predetermined.
IS THERE ONE FORM OF ASSESSMENT?

There are different forms of assessment,

including:
 Formative assessment.
 Summative assessment.
 Self-assessment.
 Peer assessment.
THE MOST COMMON FORMS OF
ASSESSMENT? (1/3)

Continuous assessment refers to

the activities required by students
during the conduct of a course.
It takes place within the normal
teaching period and contributes to
the final assessment.
THE MOST COMMON FORMS OF
ASSESSMENT? (2/3)
 Formative assessment refers to observations
which allow one to determine the degree to
which students know or are able to perform a
given task.
 It involves all those activities (assigned by
teachers and performed by students) which
provide information used as feedback so that
teaching may meet students’ needs. It can also
include teacher assessment, feedback and feed-
forward.
THE MOST COMMON FORMS OF
ASSESSMENT? (3/3)

 Summative assessment is usually carried out

at the conclusion of a unit or units of instruction,
activity or plan, in order to assess acquired
knowledge and skills at that particular point in
time.
 It usually serves the purpose of giving a grade
or making a judgment about the students’
achievements in the course.
 OTHER FORMS OF ASSESSMENT? (1/2)
Less frequent but increasingly important forms are:
 Self-assessment occurs when an appraisal
instrument is self-administered for the specific
purpose of providing performance feedback,
diagnosis and prescription recommendations rather
than a pass/fail decision.
 Students engage in a systematic review of their
progress and achievement, usually for the purpose
of improvement. It may involve comparison with an
exemplar, success criteria, or other criteria.
 It may also involve critiquing one's own work or a
description of the achievement obtained.
OTHER FORMS OF ASSESSMENT? (2/2)

 Peer assessment occurs when

students judge one another's work on
the basis of reference criteria.
 This can occur using a range of
strategies. The peer assessment process
needs to be taught and students need
to be supported by opportunities to
practice it regularly in a supportive and
safe (classroom) environment.
DOES ASSESSMENT INCLUDE TESTING?

 Yes, it does.
 Testing is a particular kind of assessment
which focuses on eliciting a specific sample
of performance.
 The implication of this is that in designing a
test we construct specific tasks that will
elicit performance from which we can make
the inferences we want to make about the
characteristics of students, groups or
individuals.
HOW DO WE TEST?

There are different sorts of testing, including:

 Achievement testing.
 Communicative testing.
 Competence testing.
 Diagnostic testing.
 Integrative testing.
 Performance testing.
 Progress testing.
 Proficiency testing.
 Psychometric testing.
 WHICH KIND OF TESTING IS THE MOST
COMMON? (1/2)

 Achievement testing. It is used to determine

whether or not students have mastered the course
content and how they should proceed. The content
of achievement tests, which are commonly given at
the end of the course, is generally based on the
course syllabus or the course textbook.
Progress Testing

It is used at various stages

throughout a language
course to determine
learners’ progress up to that
point and to see what they
have learnt.
18
 WHICH KIND OF TESTING IS THE MOST
COMMON? (2/2)

 Proficiency testing. It is used to measure learners’

general linguistic knowledge, abilities or skills
without reference to any specific course.
 Some proficiency tests are intended to show whether
students or people outside the formal educational
system have reached a given level of general
language ability.
 Others are designed to show whether candidates
have sufficient ability to be able to use a language in
some specific area such as medicine, tourism etc.
Such tests are often called Specific Purposes tests.
WHICH KIND OF TESTING IS THE LEAST
COMMON? (1/2)

 Diagnostic testing, which seeks to

identify those areas in which a student
needs further help.
 These tests can be fairly general, and
show, for example, whether a student
needs particular help with one of the four
language skills; or they can be more
specific, seeking to identify weaknesses
in a student’s use of grammar.
WHICH KIND OF TESTING IS THE LEAST
COMMON? (2/2)

 Psychometric testing, which is aimed at

measuring psychological traits such as
personality, intelligence, aptitude, ability,
knowledge, skills which makes specific
assumptions about the nature of the ability
tested (e.g. that it is unidimensional and
normally distributed). It includes a lot of
discrete point items.
 WHAT DO TESTS DO? (1/2)

What a test will appraise or measure depends

on what testers wish to know and what the
testers believe a test to be. There is indeed a
difference between:
 Competence testing, which is used to
measure candidates’ acquired capability to
understand and produce a certain level of
foreign language, defined by phonological,
lexical grammatical, sociolinguistic and
discourse constituents.
WHAT DO TESTS DO? (2/2)

Performance testing, which includes direct,

systematic observation of an actual student
performance or examples of student performances
and rating of that performance according to pre-
established performance criteria. Students are
assessed on the result as well as the process engaged
in a complex task or creation of a product.
A performance test measures performance on tasks
requiring the application of learning in an actual or
simulated setting. Either the test stimulus, the desired
response, or both are intended to lend a high degree
of realism to the test situation.
DO ALL LANGUAGE TESTS AIM AT
MEASURING COMMUNICATIVE
COMPETENCE? (1/2)

 No. They may test aspects of language

knowledge and skills which are considered to be
indicators of communicative competence.
 So, are all types of tests ‘communicative’?
 Tests identified as ‘communicative’ are those
which are interaction-based, open-ended (that is,
responses cannot be predicted as in natural
communicative environments), authentic,
behavior-based and so on.
DO ALL LANGUAGE TESTS AIM AT MEASURING
COMMUNICATIVE COMPETENCE? (2/2)

Communicative tests are supposed

to measure communicative
competence which includes:
Linguistic competence,
Sociolinguistic competence,
Discourse competence
Strategic competence.
FOUR ELEMENTS OF
COMMUNICATIVE
COMPETENCE BY CANALE AND
SWAIN:
26
1 – GRAMMATICAL COMPETENCE: WORDS AND
RULES

Knowing how to use the

grammar, syntax, and
vocabulary of a language.
Linguistic competence
asks: What words do I use?
How do I put them into 27

phrases and sentences?)

2 – SOCIOLINGUISTIC COMPETENCE:
APPROPRIATENESS
 Knowing how to use and respond to
language appropriately, given the setting,
the topic, and the relationships among the
people communicating.
 Sociolinguistic competence asks: Which
words and phrases fit this setting and this
topic? How can I express a specific attitude
(courtesy, authority, friendliness, respect)
when I need to? How do I know what attitude
another person is expressing? 28
 THE ELEMENTS OF COMMUNICATIVE
COMPETENCE
3 – Discourse Competence: cohesion
and coherence
Knowing how to interpret the larger
context and how to construct longer
stretches of language so that the parts
make up a coherent whole.
Discourse competence asks: How are
words, phrases and sentences put
together to create conversations, 29

speeches, email messages, newspaper

4- STRATEGIC COMPETENCE
 It is an aspect of communicative competence
that refers to the ability to overcome difficulties
when communication breakdowns occur
(Celce-Murcia, Dörnyei & Thurrell, 1995).
 Strategic competence asks: How do I know
when I’ve misunderstood or when someone has
misunderstood me? What do I say then? How
can I express my ideas if I don’t know the name
of something or the right verb form to use?

30
 DO TESTS TEST ONE OR MANY THINGS AT A
TIME?

There are two different types of tests:

 Integrative tests, which include activities that
assess skills and knowledge in an integrated manner
(e.g., reading and writing, listening and speaking).
Less attention is paid to specific lexicogrammatical
points.
 Discrete point tests, which contain items that
ideally reveal the candidate's ability to handle one
level of language and one element of receptive or
productive skills.
FOR WHOM ARE TESTS IMPORTANT? (1/2)

For almost all the people involved in the education

process:
 the learner who wants to know how well s/he is
doing, and also wants the 'piece of paper for
professional and education purposes,
 the teacher wants to know how the learner is
progressing and whether and how well s/he
herself is succeeding in his job,
FOR WHOM ARE TESTS IMPORTANT?

 the parents, who want to make sure

that they’re getting their money’s
worth,
 educational authorities and others
who have some interest in the learner's
progress or his/her proficiency level,
 the potential employer who relies
heavily on what tests tell him/her about
learner proficiency levels.
WHY ELSE IS TESTING IMPORTANT? (1/2)

Because of its backwash effect.

 What does this mean? It is the effect that
testing has on teaching. For better or worse, tests
and exams exert control over what goes on in
classrooms.
 This is because very many language classes are
geared more or less directly to the tests or
examinations the learners will end up taking.
Teachers must often 'teach to' a test.
WHY ELSE IS TESTING IMPORTANT? (2/2)
 Is the quality of tests important for
teaching? Yes.
 If the test is a bad one (or the teacher is
too narrow in his/her interpretation of it),
the result may be negative washback,
where we can say that teaching suffers
because of the test coming at the end of
the course.
 If the test is a good one, and its nature well
understood by the teacher, the effect on
the teaching may be very positive. There
CONSIDERATIONS WHEN CONSTRUCTING A
TEST
There are two basic considerations when
constructing a test. It must be valid and
reliable. The first concept first:
 Validity is commonly defined as 'the
extent to which [a test] measures what it
is supposed to measure and nothing else.
 If a test is valid, the outsider who looks at
an individual's score knows that it is a
true reflection of the individual's skill in
the area the test claims to have covered.
 Content validity. A test is said to have
KINDS OF VALIDITY
content validity if the items or tasks of
which it is made up constitute a
representative sample of items or tasks
for the area of knowledge or ability to be
tested (often related to a syllabus or a
course).
 Construct validity. A test is said to
have construct validity if the scores that
a candidate gets on this relate in the
same way to another test or form of
assessment for the same aspect of
knowledge.
KINDS OF VALIDITY
 Empirical validity. A measure of the
validity of a test arrived at by comparing
the test with one or more criterion
measures.
 Face validity. The extent to which a test
appeals to candidates or to those
choosing it on behalf of the candidates
because it is considered to be an
acceptable measure of the ability they
wish to measure. It is sometimes referred
to as ‘test appeal’.
KINDS OF VALIDITY

 Predictive validity. A type of validity

based on the degree to which a test
accurately predicts future performance. A
language aptitude test for example,
should have predictive validity because
the results of the test should predict the
ability to learn a foreign language.
IMPORTANT CONSIDERATION IN TESTING
Reliability is another very important
consideration when testing.
 Reliability refers to the consistency of a
test. That is, if every time the test is
administered it will have the same outcome.
But reliability does not have to do with the
content of the test alone; it has to do with
marking in two ways:
 ensuring that different raters give
comparable marks to the same script,
 the same raters give the same marks on
KINDS OF RELIABILITY
Reliability is most often(1/2)
estimated with
regard to:
 The internal consistency in a test; that
is, if there is correlation among the
variables comprising the test.
 The results when testing and re-
testing; that is, if there is correlation
between two (or more) administrations of
the same item, scale, or instrument for
different times, locations, or populations,
when the two administrations do not
differ in other relevant variables.
KINDS OF RELIABILITY (2/2)

 Inter-rater reliability, which refers to

the level of agreement between two or
more evaluators/ judges/ raters on a
particular instrument at a particular time.
 They are to apply their marks in a
manner that is predictable and replicable.
Therefore, note that inter-rater reliability
is a property of the testing situation, and
not of the instrument itself.
MORE ABOUT TESTS AND TESTING
 How does one define what will be tested?
 How can tests and feedback provided be a
positive asset in the educational process?
 What types of feedback can teachers
provide to test-takers and how?
 Are tests the best tools for evaluation and
assessment?
 What are the most important things to
remember about raters and marking
(closed and open-ended items) in
integrative and discrete point tests?
STATEMENTS ABOUT FEEDBACK: TRUE OR
FALSE? (1/2)

 The fact that the teacher gives feedback

on student performance implies a power
hierarchy: the teacher above, the
student below.
 Teachers should give their students only
positive feedback, in order to
encourage, raise confidence and
promote feelings of success; negative
feedback demoralizes.
STATEMENTS ABOUT FEEDBACK: TRUE OR
FALSE?
 Giving(2/2)
plenty of praise and
encouragement is important for the
fostering of good teacher-student
relationships.
 Very frequent approval and praise lose
their encouraging effect; and lack of
praise may then be interpreted as
negative feedback.
 Teachers should not let students correct
each other's work, as this is harmful to
their relationships.