Performance of DIMTEST and ATFIND Under Some Commonly Encountered Suboptimal Conditions Including Nonnormal Latent Trait Distributions

• In Event: Advances in Item Response Theory Modeling and Application

Abstract

Assessing test dimensionality is a fundamental part of the evaluation of a test and a prerequisite for using item response theory models. The purpose of this study was to investigate how a popular dimensionality assessment method, DIMTEST, performs under some commonly encountered suboptimal conditions when the items are dichotomously scored. A range of sample sizes, test lengths, latent trait distributions, and inter-trait correlations were manipulated in a simulation study. The performance of DIMTEST was evaluated based on its incorrect rejection rates of unidimensionality (Type I error rate) and its correct rejection rates of unidimensionality (power). There was a tendency for DIMTEST to maintain high power when the latent trait distribution was uniform (skewness = 0; kurtosis = 1.7).

Authors

• Kirsten Hochstedt, The Pennsylvania State University

• Pui-Wa Lei, The Pennsylvania State University

• Jeffrey Borck