The Electronic Journal of Business Research Methods provides perspectives on topics relevant to research in the field of business and management
For general enquiries email administrator@ejbrm.com
Click here to see other Scholarly Electronic Journals published by API
For a range of research text books on this and complimentary topics visit the Academic Bookshop
Information about The European Conference on Research Methodology for Business and Management Studies is available here

linkedin-120 

 

twitter2-125 

 

fb_logo-125 

 

Journal Article

Characteristics of Single‑Item Measures in Likert Scale Format  pp1-12

Aliosha Alexandrov

© Sep 2010 Volume 8 Issue 1, Editor: Ann Brown, pp1 - 62

Look inside Download PDF (free)

Abstract

The use of single‑item measures has been encouraged by several authors asserting that single‑item measures are appropriate and can substitute multiple‑item measures in many cases. This study focuses on the characteristics of single‑item measures in Likert scale format. There are two motives behind it: first, the Likert scale has been called problematic and its usage discouraged by the very proponents of single‑item measures; and second, the reverse wording of Likert items has led to many problems with multiple‑item measures. Because the Likert scale is one of the most used scales in marketing and management, and more researchers may decide to use single‑item measures in Likert scale format, it becomes necessary to answer the question if it is usable or not. This research scrutinizes the characteristics of the Likert scale in a positive‑negative continuum: from positive to negative with different levels of intensities. Based on collected sample data for three popular computer brands, the main conclusion is that only positively worded Likert items with a fairly high level of intensity should be used as single‑item measures. The supporting empirical evidence includes: (1) positively and negatively worded items are not true opposites, (2) items with reversed scores inflate means, (3) items with neutral intensity have unique conceptual meaning, (4) dependent variables are predicted best by independent variables with similar intensity and (5) negatively worded items contain a method factor that limits their ability to capture the measured concept. The results also suggest that the effect of the method factor is expressed more when respondents are not familiar with the object of the measured concept. The findings in this study provide guidelines for the practical use of measures in Likert format. Scales in other formats should undergo similar scrutiny.

 

Keywords: single-item measures, Likert scale, negatively-worded items, reversed items, C-OAR-SE

 

Share |