Page 74 - DEDU504_EDUCATIONAL_MEASUREMENT_AND_EVALUATION_ENGLISH
P. 74
Educational Measurement and Evaluation
Notes (d) To verify the result with the help of factor-analysis technique.
“The factoral validity of a test is the correlation between that test and the factors common
to a group of tests or other measures of behaviour.” —A. Anastasi
(e) To compare scores of individuals before placing them in the specific situation and after placing
them in the specific situation; for example, to administer an intelligence test on a group of
students before training and obtain their scores, and then administer an intelligence test on
this group after training, and then go on to compare the two scores. If there is no variation in
scores, then it can be accepted that the test measures intelligence correctly.
Face Validity
When a test appears to measure that element, trait or subject knowledge for which it has been
constructed then it is said that it possesses face validity. In face validity, a test appears to be valid
by the form of its items. This type of validity can be found out by mere looking at the items.
Generally, validity of achievement tests and personality tests is found out in this way. For example,
if we have to measure history-related achievement, then its items can be seen to say that it measures
achievement in history or not. In the same way, a personality inventory which has been constructed
to measure introvert and extrovert aspects of personality, should comprise of such items which can
measure the two aspects equally well. Statistical methods cannot be used too calculate face validity.
For this, the assistance of subject specialists or field specialists is taken. On reading of items, an
effort is made to find out what attribute they measure correctly. Therefore, only qualitative statements
can be given about face validity; for example, high, low, average or poor etc. Mosier has mentioned
four types of face validity : by assumption, by definition, by appearance and by hypothesis.
Factor-analysis is a type of special statistic technique in which the minimum factors existing
in tests are found out with the help of correlation coefficients obtained in different tests.
The credit for developing this technique goes to Thomson (1960) and Burt (1941).
Self Assessment
1. Fill in the blanks :
(i) The extent to which the purpose or objective of a test is realized is called its _________ .
(ii) According to _________, Validity of a test are of _________types.
(iii) The major limitation of content validity is the lack of _________that summarize the degree
of validity.
(iv) In _________, a test appears to be valid by the form of its items.
(v) The major problem faced in finding out _________is the selection of suitable criterion.
6.3 Methods of Validity
The chief aim of validity determination is to know whether a test is measuring that trait to the
required extent and with accuracy, so that meaningful hypotheses can be formulated on its basis,
and important decisions can be taken on its basis. Following are the important methods of calculating
validity :
1. Simple correlation method.
2. Bi-serial correlation method.
3. Point bi-serial correlation method.
68 LOVELY PROFESSIONAL UNIVERSITY