Comparison of Single-Stimulus Format and Forced-Choice Format Instruments Using Thurstonian Item Response Theory

• In Event: 58.028 - Applications of Item Response Theory

Abstract

Self-beliefs such as teacher self-efficacy are present in all learning experiences, and play a lead role in guiding teacher behaviors (Fives & Buehl, 2008). Teacher self-efficacy has been researched extensively in the past decade; however, Henson (2002) observed that self-efficacy scales are often adopted despite glaring psychometric problems. One significant source of measurement bias is the commonly used traditionally single-response Likert response format. This study aims to provide solid evidence for an alternative comparative response format by comparing two versions of one teacher self-efficacy scale, each with a different response format. Thurstonian item response analysis will be used to demonstrate the benefits of the comparative format as it applies to the development of self-efficacy scales and self-report scales more generally.

Authors

• David Dueber, University of Kentucky

• Abigail M. A. Love, University of Kentucky

• Michael Toland, University of Kentucky

• Trisha Anne Turner, University of Kentucky