The Effect of Response Format on Ratings of Teaching

Abstract

An experiment is reported in which students rated the effectiveness of their lecturers by using three different types of response format. No differences were found between a scale in which all the points were verbally labelled and one in which only the end points were labelled. However a third scale, in which each point was given a precise percentage label, differed from the other two, and proved to give a more nearly accurate measure of the performance being rated. It is concluded that such percentage scales should be used wherever possible, although they may be inappropriate for some groups of subjects.