Reconceptualizing language teaching: an in-service teacher education course in uzbekistan

– does a test measure what it is trying to measure? Reliability – does 
the test produce consistent results? Practicality – does the test take 
an adequate amount of time, money, energy and resources to make? 
Impact – is there a positive effect on the teaching and learning with 
the distribution of this test? These four principles work together and are 
sometimes contradictory. For instance, multiple choice assessments, which 
are very practical to distribute and score, are often the most notoriously 
difficult to write. Thus, the closer you become to establishing all four princi-
ples in your assessment practices, the more you learn about your students’ 
abilities, and will thus, be able to make the necessary choices about how to 
design your assessment and the appropriate feedback to give them. Know-
ing and understanding these principles is important not only for those 
who design their own tests, but also for those who selects an assessment 
tool from existing ones. Inappropriate selection from well-constructed 
commercially available tests for a certain purpose of assessment can cause 
as much harm as developing one from a scratch. We discuss the four areas 
in detail below.
Selecting or developing an assessment tool is a very responsible task 
for a teacher, as the information obtained through it will be used for mak-
ing decisions. The administration of the test is as important as the test con-
struction itself. A teacher should consider a fair conduction of assessment 
of any kind, whether it is a mere classroom check or high-stake official ex-
amination. Unfavourable assessment condition can influence the results. 
Validity is critical for assessment. This quality signifies whether the in-
tended purpose of assessment has been met and is reflected in the created 
assessment tool. In other words, by checking the validity of a certain test 
or task, we learn whether it provides the evidence about test-takers that 
we intended to obtain. Validity also influences the way assessment results 
are interpreted by teachers and delivered to stakeholders. It is, therefore, 
vital that we know what the purpose of a certain assessment tool is before 
checking whether it is valid or not. 
There are three types of validity from Cumming &. Berwick (1996), 
which are commonly referred to in the language testing literature:

---

115
CHAPTER THREE: LANGUAGE ASSESSMENT/TESTING
• Construct validity: ‘a mutual verification of the measuring instru-
ment and the theory of the construct it is meant to measure’, p.29; 
‘we examine the psychological trait, or construct, presumed to be 
measured by the test and we cause a continuing, research interplay 
to take place between the scores earned on the test and the theory 
underlying the construct’ (p. 26)
• Content validity: ‘review of the test by subject-matter experts and 
a verification that its content represents a satisfactory sampling of 
the domain’ (p. 22)
• Face validity: ‘the appearance of validity…for example that the lan-
guage and contexts of test items be expressed in ways that would 
look valid and be acceptable to the test taker and to the public 
generally’ (pp. 23-24)