Page 175 - DEDU504_EDUCATIONAL_MEASUREMENT_AND_EVALUATION_ENGLISH

Kulwinder Pal, Lovely Professional University Unit 13 : Interpretation of Test Scores : Qualitative and Quantitative

Unit 13 : Interpretation of Test Scores : Notes

Qualitative and Quantitative

CONTENTS
Objectives
Introduction
13.1 Relative Rank
13.2 Qualitative Interpretation of Test Scores
13.3 Quantitative Interpretation of Test Scores
13.4 Summary
13.5 Keywords
13.6 Review Questions
13.7 Further Readings

Objectives

The objectives of this unit can be summarised as below :
• To explain about relative rank
• To discuss about Qualitative Interpretation of test scores
• To discuss about Quantitative Interpretation of test scores

Introduction

Scores, whether raw or converted, do not suffice for the complete interpretation of an individual’s
performances on psychological tests. The several, aspects of test standardization thus far presented
are concerned with the performance of groups of persons and with average relationships revealed
by statistical treatment of results. It happens, however that although certain types of test items
meet some or most of the statistical requirements of validity, they are unsatisfactory as indicators
of intelligence when used for clinical purposes. For example, on the Stanford-Binet scale, the
percentage of adults able to repeat eight digits forward (digit-span test) is approximately the same
as the percentage who can solve one of the more difficult reasoning problems. Yet, in clinical
examinations, psychologists find some adult mental defectives who can pass the former test,
although a mental defective can never succeed with the latter. What this means is that statistical
validation of a test item is not always sufficient; it must be supplemented by the pragmatic
criterion of use with a wide variety of individuals in a variety of situations in order to show
whether or not it has discriminative value among individuals at the several levels of ability.

13.1 Relative Rank
The raw score (that is, the actual number of units or points) obtained by an individual on a test
does not in itself have much, if any, significance. One test may yield a maximum score of 150,
another 200, and a third 300. Obviously, then, any point score on one of these tests is not directly
comparable with the same number of points on either of the others; a score of 43 on one test
cannot be directly compared with a score of 43 on another. Furthermore, the average scores of
each of these will in all probability be different, as will the degree of variation of scores (called
the deviation) both above and below the average.

LOVELY PROFESSIONAL UNIVERSITY 169