Examining Power and Type I Error for Step- and Item-Level Tests of Invariance

• In Event: 14.046 - Poster Session 1
  In Poster Session: 14.046-6 - Contemporary Models and Their Applications: A Poster Session

Abstract

Recent studies have explored within item invariance for polytomous items; the lack of invariance at this level is referred to as differential step functioning. This study investigates the effects of polytomous item features on the power and Type I error of nonparametric invariance tests of differential step and item functioning. Additionally, this study examines which statistical procedures are most effective for adjusting per comparison Type I errors for DIF detection in polytomous items. Independent variables are impact, whether or not items exhibit DSF, number of score levels, DSF pattern, and the model used to generate data. Dependent variables are Type I error and power.

Authors

• Alicia Ayodele, University of Minnesota

• Ernest C. Davenport, Jr., University of Minnesota