Is there an optimal number of alternatives for Likert-scale items? Effects of testing time and scale properties

Abstract

Investigated 3 criteria used to determine the number of Likert scale rating steps to employ: proportion of scale used, testing time, and usage of an uncertain category. A modified version of the Florida Scale of Civic Beliefs, in which there were 2-19 alternatives for each item, was administered to 360 undergraduates. Results indicate that for cumulative scores from Likert-type items proportion of scale used was independent of the number of scale points, while mean testing time increased and usage of the uncertain category decreased as the number of rating steps increased. Results are integrated with earlier findings and implications for Likert scale construction are described. (15 ref.) (PsycInfo Database Record (c) 2025 APA, all rights reserved)