City Research Online

Abstract

First developed by Likert, agreement scales have been widely used in survey questionnaires for many years. However, there are potential shortcomings associated with the format, including the potential for acquiescence bias and the complexity that agreement scales add to the response process, placing an additional, and often unnecessary, burden on respondents. We conducted an experiment to compare responses to agreement-scale questions with those to equivalent questions using item-specific scales which directly capture the dimension of interest. The results of the experiment, and other related research, suggest that using item-specific scales may produce more reliable measures. This should be borne in mind by researchers when designing questionnaires, particularly when they are not constrained by maintaining a time-series, and have the freedom to develop new questions.

Publication Type:	Article
Additional Information:	The author(s) holds copyright of their material.
Subjects:	H Social Sciences > HM Sociology
Departments:	School of Policy & Global Affairs > Sociology & Criminology
SWORD Depositor:	Symplectic Administrator

Export

Downloads