Impact of the Number of Scale Points on Data Characteristics and Respondents’ Evaluations: An Experimental Design Approach Using 5-Point and 7-Point Likert-type Scales

Abstract

A remarkable deal of social research is based on data collected through the use of Likert-type scales. The optimal number of response categories in Likert-type scales has been subject to an academic debate for years. This article studies the differences between 5-and 7-point Likert-type scales using the SERVPERF Scale, which was developed by Cronin and Taylor in 1992, as the measuring instrument. A pretest-posttest control group experimental design was used to test whether the differently pointed response categories lead to any statistical differences in data characteristics, dimensional structure of the scale and data fit. Results do not show any statistically significant differences in terms of normality and reliability whereas different dimensional structures are achieved for the 5-and 7-point scale formats of SERVPERF using Exploratory Factor Analysis. ANCOVA results reveal that the number of response categories is not affective on the participants' evaluations of SERVPERF. The results of confirmatory factor analysis show that the best fit is achieved for the 7-point SERVPERF.