Page 73 - DEDU504_EDUCATIONAL_MEASUREMENT_AND_EVALUATION_ENGLISH

Unit 6 : Validity – Types , Methods and Usability

Time Notes
Test Criterion
In concurrent validity, this ‘other measure of performance’ is available in the present, and it
indicates to the fact how far the present ability or working efficiency can be measured accurately
by that test.
The major problem faced in finding out criterion-related validity is the selection of the suitable
criterion. Generally, appropriate and valid criterion is not available on the basis of which
predictive validity or concurrent validity of a test under construction can be found out. This
type of validity is established on the basis of certain criteria, and if these criteria are not valid
themselves, the tests constructed on their basis cannot be reliable too. This is an significant
problem. So, Ebel says that validity of psychological tests is a misleading characteristic. He also
says that it is not necessary to find out validity of all tests, nor can it be. So, content validity or
face validity of a test is very important, and it can be relied upon in a greater degree.
3. Construct Validity : According to Cronback and Meehl, when a test has to be analyzed in the
form of a measurer of such attributes which can be defined operationally, in such a situation,
construct validity is needed. This type of validity has been termed by Garrett as factoral validity,
and this has been termed by Guilford as relevant validity. Validation is especially related with
the application of tests. They point out to what extent a test evaluates knowledge of a subject or
future efficiency of individuals of a specific field.
From psychological viewpoint, each test evaluates psychological abilities of an individual,
therefore the scores obtained on a test can be analyzed in the psychological context also. For
example, the scores of a science test indicate the extent of knowledge in science; at the same
time, it can point to the extent of reasoning ability in respect of science. In other words, test
scores express those behaviours or expressions of an individual which become the basis of
success in different fields. These psychological elements as manifested in test scores are the
basis of construct validity.

Garrett has explained this factor by the example of watch. Supposing you have set
your watch faster by ten minutes, it will indicate 10.10 when it is 10 o’clock, but this
time will not be valid because of its difference from the standard time. Of course, the
watch will be reliable certainly.

According to Anastasi, “The construct validity of a test is the extent to which the test may be said
to measure a theoretical construct or trait.”
According to Cronback and Meehl, “With construct validation both the theory and the test are
being validated at the same time.”
The following procedure has to be followed to establish construct validity :
(a) To determine the psychological elements which are behind the test performance.
(b) To construct hypotheses on the basis of those elements.

(c) To verify these hypotheses.
The following methods are used for finding out construct validity :
(a) To compare the scores of different groups of individuals.
(b) To establish correlation with any other test.
(c) After having ascertained the number of test items, to determine what mental faculties or
psychological elements related to the subject become active on responding to these items.

LOVELY PROFESSIONAL UNIVERSITY 67