Author ORCID Identifier

https://orcid.org/0000-0002-9665-4599

Year of Publication

2020

Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation

College

Education

Department

Education Sciences

First Advisor

Dr. Michael D. Toland

Second Advisor

Dr. Xin Ma

Abstract

This is a simulation study that evaluates the performances of two models for the detection of uniform differential item functioning (DIF). Simulated data are generated by a multilevel partial credit model (MLPCM). The purpose of this study was to compare the accuracy of two DIF detection procedures, hierarchical ordinal logistic regression (HOLR) for multilevel data and multilevel generalized Mantel-Haenszel (MGMH: French & Finch, 2013; French, Finch, & Imekus, 2019). Conditions manipulated were the number of participants per cluster (20, 40), number of clusters (50, 100, 200), DIF magnitude (0, .4, .8), and magnitude of intraclass correlation coefficient (.05, .25, .45). Furthermore, only one grouping variable was used within-groups. Data was simulated using R (R Core Team, 2019), whereas analyses will be performed using SAS 9.4 (SAS Institute, 2013) and R. In general, HOLR maintains the Type I error rate better than MGMH and HOLR has more power than MGMH under most simulation conditions.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2020.073

Available for download on Wednesday, October 21, 2020

Share

COinS