NotesFAQContact Us
Collection
Advanced
Search Tips
Peer reviewed Peer reviewed
Direct linkDirect link
ERIC Number: EJ1088364
Record Type: Journal
Publication Date: 2016-Feb
Pages: 9
Abstractor: ERIC
ISBN: N/A
ISSN: ISSN-1076-9986
EISSN: N/A
Bad Questions: An Essay Involving Item Response Theory
Thissen, David
Journal of Educational and Behavioral Statistics, v41 n1 p81-89 Feb 2016
David Thissen, a professor in the Department of Psychology and Neuroscience, Quantitative Program at the University of North Carolina, has consulted and served on technical advisory committees for assessment programs that use item response theory (IRT) over the past couple decades. He has come to the conclusion that there are usually two purposes for using IRT in the contexts of assessments of educational achievement: (1) Item selection, which in practice means removing items that perform poorly in item tryout; and(2) Computing comparable scores from alternate forms that may differ in difficulty (this includes computerized adaptive testing). Given these purposes, there are questions asked (or answers given) that are "bad questions" (or "bad answers"). The bad questions are about the analyses or the results, and are asked in a way that invite (or insist upon) a yes or no answer, when the right answer should be either qualified or quantified. These bad questions are asked by intelligent and well-trained persons, and that suggests a need to improve training in measurement. This article provides examples of "bad questions" involving IRT, and discusses possibilities for improvements in this area.
SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: http://sagepub.com
Publication Type: Journal Articles; Reports - Descriptive
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Grant or Contract Numbers: N/A