Evaluating type I error and power rates using an effect size measure with the logistic regression procedure for DIF detection

The logistic regression (LR) procedure for differential item functioning (D IF) detection is a model-based approach designed to identify both uniform a nd nonuniform DIF. However, this procedure tends to produce inflated Type I errors. This outcome is problematic because it can result in the inefficie nt use of testing resources, and it may interfere with the study of the und erlying causes of DIF. Recently, an effect size measure was developed for t he LR DIF procedure and a classification method was proposed. However, the effect size measure and classification method have not been systematically investigated. In this study, we developed a new classification method based on those established for the Simultaneous Item Bias Test. A simulation stu dy also was conducted to determine if the effect size measure affects the T ype I error and power rates for the LR DIF procedure across sample sizes, a bility distributions, and percentage of DIF items included on a test. The r esults indicate that the inclusion of the effect size measure can substanti ally reduce Type I error rates when large sample sizes are used, although t here is also a reduction in power.