Exploring Omitted Variable Bias Threat to Differential Item Functioning Assessment

• In Event: Advancements in Detecting Differential Item Functioning and Measurement Invariance

Abstract

The effect of omitted variable bias (OVB) has long been ignored within the context of differential item functioning (DIF) assessment. Consequently, the estimation of DIF effects may not be unbiased, resulting in inflated Type I error rates and/or deflated power rates in DIF assessment. To overcome these problems, the study examined the OVB threat to DIF and incorporated the DIF-free-then-DIF strategy in the exploratory DIF assessment with considering confounding variables. According to the simulation study, the results indicated that the controlled method can well control Type I error rates under all conditions, while the original method lost control when confounding variables exhibited DIF. The controlled method appears promising for helping test practitioners avoiding OVB in DIF assessment.

Authors

• Hsiu-Yi Chao, National Taiwan Ocean University

• Jyun-Hong Chen, National Cheng Kung University

• Chi-Chen Chen, College Entrance Examination Center