Is the Student Opinion Scale Biased? Investigating Measurement of Minority Students' Motivation

• In Event: 15.059 - Fairmont Royal York Roundtable Session One
  In Roundtable Session: 15.059-4 - Minority Faculty, Students, and Assessment

Abstract

When providing answers to an instrument, students who come from different cultural or ethnic backgrounds may exhibit different response styles (e.g., Supple et al., 2012). Differential responding can be problematic; scores on the instrument may not be valid indicators of the construct of interest. This study examined the invariance of the Student Opinion Scale (Sundre & Moore, 2002), a 10–item measure of test-taking motivation that utilizes a 5-point Likert scale. We conducted likelihood ratio tests using the R package mirt (Chalmers, 2018) to test whether African-American and Hispanic students responded differently to the instrument than their White peers, constraining item slopes and thresholds to be equivalent across items. Results indicated possible differential step functioning for five items.

Authors

• Catherine E. Mintz, University of Iowa

• Derek Sauder, James Madison University