Page 86 - DEDU504_EDUCATIONAL_MEASUREMENT_AND_EVALUATION_ENGLISH

Educational Measurement and Evaluation

Notes According to Frederick G. Brown, “Reliability can be defined as that ratio of true variance in a
set of test scores to the total or obtained variance.”
St 2
R =
xx Sx 2
Where, R = Reliability co-efficient of the test
xx
St = Variance of true score
2
2
SX = Variance of observed or obtained score
The more is true variance, the lesser will be error variance, and the more reliable will the test be
considered. In fact, measurement of true variance cannot be done directly, it can only be estimated.
Therefore reliability is a imaginary attribute of test scores.
Some scholars have defined reliability in the following words :
1. “Briefly, reliability is the proportion of the true variance in obtained true scores.”
—Guilford
2. “...defined reliability of measurement as the extent of unsystematic variation in scores of
an individual to some trait when that trait is measured a number of times.” —Ghiseli
3. “The reliability of a test of any measuring instrument depends upon the consistency with
which it gauges the ability to whom it is applied.” —Garrett
4. “The reliability of a test can be defined as the correlation between two or more sets of
scores on equivalent tests from the same group of individuals.” —Stodola and Stordahl
5. “The term reliability refers to the extent to which a test is internally consistent and the
extent to which it gives consistent result on testing and retesting.” —Freeman
It is essential to consider the following points while examining reliability of a test :
1. Reliability is a trait of the test results and not of the test itself.
2. Obtained scores of a test are not equally reliable, that is, the obtained scores will be equal
under all circumstances, this is not meant by reliability, rather it means that they will be
equal under certain and specific circumstances, such as after an interval of time or after the
test is administered on the one-half or one-forth of the group.
3. Reliability of a test affects its validity too, that is, reliability is a necessary but not a
sufficient condition for validity.
4. If the reliability of a test is low, then its validity too will be low; however, it may not be
true that the validity of a test will be in full measure if its reliability is in full measure. It
may be possible test may be 100% reliable but not fully valid.

Test reliability is the aspect of test quality concerned with whether or not a test produces
consistent results.

7.2 Types of Reliability

Decision Consistency
In the descriptions of test-retest and parallel forms reliability given above, the consistency or
dependability of the test scores was emphasized. For many criterion referenced tests (CRTs) a
more useful way to think about reliability may be in terms of examinees’ classifications. For
example, a typical CRT will result in an examinee being classified as either a master or non-
master; the examinee will either pass or fail the test. It is the reliability of this classification

80 LOVELY PROFESSIONAL UNIVERSITY