Optimum number of response categories /

Abstract

Through linear transformations of raw item scores, the paper converts 3-ponit, 4-point, 5-point and 7-point items to continuous, monotonic, normally distributed scores ranging between 1 to 5. This provides a platform for meaningful comparisons of scales with different number of response categories with respect to parameters like reliability, validity, discriminating power, and undertakes analysis in parametric set up. The method makes no assumption of continuous nature or linearity or normality for the observed variables or the underlying variable being measured. Thus, the assumption-free simple method can have wide applicability. Use of such methods of converting scores of Likert items is recommended for clear theoretical advantages and easiness in calculations. Inverse relationship derived between new measures of test discriminating value in terms of co-efficient of variation (CV) and theoretically defined test reliability. Empirically, such inverse relationship was observed for the scales. Number of response categories did not show much influence on discriminating value, reliability and factorial validity, even for the transformed normalized scores in the range 1 to 5. Thus, the study could not find optimum number of response categories which maximize validity, reliability or discriminating value. Future studies with multi-data set for generalization of findings are suggested.