Effects of Item Grouping and Position of the “Don't Know” Option on Questionnaire Response

Abstract

This study examined the effects of questionnaire item grouping (thematic versus random) and placement of the “don't know” option (before or after the rating scale) on the frequency of nonattitude responses (checking don't know and omission) in rating attitude statements. Response consistency was also evaluated by item grouping. Findings supported the hypothesis that knowledge of an item's theme and of other items related to that theme encouraged attitude responses by reducing the frequency of don't know responses. The authors found no evidence that positioning the don't know option at the end of the rating scale reduced nonattitude responses. Furthermore, thematic grouping enhanced internal consistency reliability for four of six subscales and the total scale.