7.

Language testing, language assessment and error correction

a. Background, history of testing

- Language testing and assessment build on theories and definitions provided by linguistics,
applied linguistics, language acquisition and language teaching, as well as on the
disciplines of testing, measurement and evaluation.
- Language testing uses these disciplines as foundations for researching, theorizing and
constructing valid language tools assessing and judging the quality of language.
- The field of language testing is therefore viewed as consisting of two major components:
o one focusing on the ‘what’ referring to the material that need to be assessed (trait)
o and the other component pertaining to the ‘how’. which addresses the specific
procedures and strategies used for assessing the ‘what’ (method)
 The trait has been defined by the language testing field, the ‘how’ on the other hand, is
derived mostly from the field of testing and measurement.
- Matching the ‘how’ of testing with the ‘what’ of language uncovers several periods in the
development of the field.
o In the pre-scientific period the focus was on translation and structural accuracy. It is
developed along the lines with grammar-translation teaching.
o Discrete-point testing viewed language as consisting of lexical and structural items
so that the language tests of that era presented isolated items.
o In the integrative era, language tests tapped integrated and discoursal language (I
think: Audio-Lingual Method, etc.)
o In the communicative era, tests aimed to replicate interactions among language
users. Learners are expected to perform tasks taken from real life contexts.
- It is being realized nowadays that language testing is not occurring in homogenous,
uniform and isolated contexts, but rather, in diverse, multilingual and multicultural
societies and thus posing new challenges and questions with regards to what it means to
know languages in education and society.
- The current era can be described as the era of uncertainty, where questions are being raised
about the meaning of language and the possibilities for measuring this complex and
dynamic variable.

b. Basic principles of testing:

- Test what you teach and teach what you test
- Test on a regular basis
- Give feedback quickly to students
- Make testing part of a learning progress
- Make tests reflect the student level
- Avoid using tests for punishment
- Never use testing instead of teaching
- It is not the only measure for grading
- A test should be sample of what you teach and what learners are trying to learn

c. Test types
 - Test: A measurement instrument designed to elicit specific sample of an individual
behaviour (gives evidence of the abilities which are of interest)
- Tests can be categorized according to their purpose
o Placement tests: originally designed to place learners at an appropriate level in a
programme or course. Various types or testing procedures such as dictation,
interview or grammar test can be used for placement purposes.
o Achievement tests: also referred to as attainment or summative tests. Designed to
measure how much of a language someone has learned during a particular course,
end-of-year tests.
o Progress tests: achievement-like tests, designed to measure whether or not learners
develop their skills, knowledge during the course. They help language teachers to
assess the degree of success of their teaching.
o Proficiency tests: measure how much of a language someone has learned. They are
not linked to any particular material, but measure the learners’ general level of
language mastery.
o Diagnostic tests: are aimed to identify students’ weaknesses.

- The score on a language test reflect the language ability of the learner with reference to a
group performance or to a criterion level.
o The norm-references test is designed to show the differences between individuals
compared to the norm or the reference group.
o A criterion-references test is a test which measures students’ performance
according to a particular standard or criterion which has been agreed upon. The
student must reach this level of performance to pass the exam

d. Methods of testing
- Tests can be indirect and direct depending on the way they elicit language samples.
o Indirect way of testing involves measurement of isolated areas of the ability in
question. An example of indirect testing involves making inferences on writing
ability on the basis of a test on punctuation. It assesses knowledge without authentic
application.
o Direct testing measures the language ability in question as it appears in real life
situation. The primary goal is to be as much like real life as possible.
- According to the focus of a test we can distinguish between discrete – point/analytical and
integrated tests.
o The former focuses on separate elements of ability and intend to test them in
isolation whereas the latter one focuses on the performance as consisting of several
components.

e. Qualities of a language test

- Language tests should resemble authentic language use as much as possible, although
sampling the language using the different criteria of authenticity is not easy.
- There are 3 major qualities that a good test possesses. These elements constitute the most
important characteristics of language tests and need careful consideration when designing and
implementing tests in a particular situation with a defined purpose for a certain group of test
takers.
 o Reliability: is the degree to which an assessment tool produces stable and consistent
results
o Validity: measures what it is supposed to and nothing else
o Practicality: practical to administer and solve

f. Assessing the four skills

- When talking about language tests, we usually refer to one or more of the four language
skills. The 4 skills are: reading, listening, writing, speaking.
- These skills are further grouped into two receptive and two productive skills according to the
mode of language production.
- According to the channel of language, reading and writing are written skills, speaking and
listening are oral skills.
- This distribution suggests the possibility of isolating language ability elements and dealing
with them separately.

g. Item types
- There are several item types to choose from and the choice should be made carefully,
because item types affect performance on the test. If the candidate is not familiar with the
item type or it is not clear what is required from him, the elicited performance will not reflect
his ability.
- There are two broad item types: objective and subjective.
o Objective item can also be called recognition, selected or fixed response type, as
the candidate has to choose from several options.
 true-false, multiple-choice, matching, ordering, gap-filling, one-word
answers
o Subjective item types, which are also known as constructed or extended response
items require language production from the candidate.
 compositions, essays, oral interviews, information-gap activities

h. Scoring methods and rating scale

- Depending on the test method, language is elicited using different item types and then,
candidate responses are assessed. Measurement usually results in a score, according to which
inferences can be made about the candidates’ language ability. The examiner compares the
response to the key provided by the test designer. The quality of the response is established by
an examiner or more examiners.
- For both objective and subjective types of assessment the criteria for correctness should be
specified and the procedures of arriving at a score established.
-A rating scale in general is a scale that consists of ranked categories according to which
subjective judgements are made. In language testing the different scales to assess language
performance contain descriptors to make decision clear.
- Scales have three main functions in measurement: they describe the level of performance,
they guide assessors how to rate performance and provide test designers with information on
test specifications.
-According their purpose, scales can be used for finding out about progress or achievement
and can be used for comparisons which are either norm or criterion-references.
- Generally there are 2 types of scales used in language performance assessment:
 o Holistic scale: to assess the performance as a whole
o Analytic scale: to assess the performance according to its performance

- A wide variety of rating scales can be produces depending on the purpose of the test and
the area of language ability to be tested.
- Test-takers’ behaviour, mood at the time of the test and their attitude towards the testing
situation determines the performance on the test to a large extent.

i. Error correction in the classroom

Error correction is one of the main dilemmas of teachers. It is always tricky to know
when and if to correct students and how to go about it.
The danger of over-correcting is that students will lose motivation and you may even
destroy the flow of the class or the activity by butting in and correcting every single mistake.
The other extreme is to let the conversation flow and not to correct any mistakes. There are
times when this is appropriate but most students do want to have some of their mistakes
corrected as it gives them a basis for improvement.

Are you working on accuracy or fluency?

Before you begin an activity, bear in mind whether you are concentrating on accuracy or
fluency. For a class discussion for example, fluency would be appropriate. The important
thing is that students are expressing themselves. However if students have had time to prepare
a role-play and are then going to perform it you may want to encourage accuracy. Be clear of
the aims of the task and make sure students are aware of what you expect from them. Don’t
present an activity as a fluency task and then pick them up on every single mistake.

Ways of correcting mistakes

 Self correction / Peer correction
Students can often correct themselves when they realise they’ve made a mistake.
Sometimes the mistake is simply a ‘slip’ and they are aware of the correct version. Give
students a chance, and time, to correct themselves. Often by just raising your eyebrows or
repeating the mistake students will know what you mean and back track to correct the error
themselves. Some teachers create all sorts of hand signals to indicate the type of error.
Students can also correct one another. Peer correction often helps to create a positive class
atmosphere as students realise you are not the only source of error correction and they can
learn a lot from one another.

 Correction slots
One way to focus on students’ mistakes is to take ‘time out’ of an activity and look at
mistakes as a group. When students are doing a speaking task in pairs or groups, monitor the
students and listen in on what they’re saying. Make a note of the mistakes that you hear;
whether they are pronunciation, grammatical or lexical. Collect a selection of their errors and
then stop the activity. They can have a notebook and pen and make notes of mistakes they
hear. If they do their job well they could even run the correction slot with their mistakes
instead of you. Usually most of the mistakes can corrected by the students themselves.

 On the spot correction

Students often appreciate instant correction. Think about what type of activity it is before
deciding whether or not it’s appropriate to correct on the spot. You don’t want to destroy the
flow of the task by cutting in. Students can also be responsible for on the spot correction if
they are encouraged to pick up on each other’s mistakes.