Page 70 - DEDU504_EDUCATIONAL_MEASUREMENT_AND_EVALUATION_ENGLISH

Educational Measurement and Evaluation

Notes The following formula will prove useful in remembering validity in brief :
Adequacy + Objectivity → Reliability + Aim → Validity

For example, if a test has been constructed to measure intelligence, then the extent to which it will
measure intelligence will be its validity. On the contrary, if an intelligence test comprises of such
questions which arouse his emotions, and he commits error in solving questions, then the influence
of emotion in the test will decrease its validity. In the same way, if in an interview for the selection
of a post of a psychologist, such questions are asked as : Who is Sachin Tendulkar ? Who is Laden
? What are the objectives of the five-year plans ? What is 10 + 2 + 3 system ? Who is Dhanraj Pillai
? etc. Such a test will have a very low reliability, because it contains those elements which do not
measure those factors or qualifications which are necessary for the post of a psychologist.
Psychological tests are constructed keeping many objectives in mind. The extent to which the purpose
or objective of a test is realized is called its validity. A test will be valid only from the standpoint of
its specific reference and realization of its specific objective. This may be valid in reference to one
objective, and it may not be so in reference to another objective; that is, only one test cannot be valid
for all objectives at the same time. Therefore, if we have to know how valid is a test, it is also
necessary to know the purpose for which it is valid. If it is an IQ test, it cannot be valid for measuring
achievement in subjects. It is evident that validity of a test can be ascertained in many ways and in
many contexts. Validity is a very important characteristic of a test. If a test is not valid, then the data
represented by it will not be correct in the context of its purpose for which it was made.

The following factors should be taken into consideration while evaluating validity of a test :
1. Validity is the characteristic of test results, and not of the test.
2. Validity is not a common characteristic of a test, rather it is used in the measurement of specific
abilities.
3. We can see about the validity of a test whether it is highly valid, least valid or ordinarily valid.
A test can never be fully valid or fully invalid.
Some famous scholars have defined validity as follows :
1. “The validity of a test or of any measuring instrument depends upon the fidelity with
which it measures what it purports to measure.” — Garrett (1959)
2. “In general, the term validity refers to effectiveness of a test in achieving specific purposes.”
—Stodola and Stordahi
3. “In a broad sense, validity has to do with the question of what test scores measure and what
they will predict.” —Guilford
4. “A measurement procedure is valid in so far as it correlates with some measurement of
success in the job for which it is being used as a predictor.” —Thorndike
5. “Validity is defined as the proportion of true variance that is relevant to the purposes of the
testing.” —Frederick G. Brown
6. “Validity is the extent to which a test measures what it purports to measure.” —Cronback
7. “It is the correlation of the test with some criterion.” —Gulikson
8. “Perhaps the commonest definition given for the term validity is that it refers to the extent
to which a test or a set of operation measures what it is supposed to measure.” —Ghiselli
If validity is defined on the basis of variance, it becomes clear that the entire variance is formed of
two parts of test scores : 1. true variance and 2. error variance; that is :

2
S 2 X = S + S 2 E
T
Where, S 2 X = Total Variance

64 LOVELY PROFESSIONAL UNIVERSITY