Date of Award

Spring 5-12-2017

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Educational Policy Studies

First Advisor

T. Chris Oshima

Second Advisor

Hongli Li

Third Advisor

Tim Sass

Fourth Advisor

Nicole Patton Terry

Fifth Advisor

Keith Wright

Abstract

Advances in computational ability have produced new software, such as Mplus 7 (Muthén & Muthén, 2011) and IRTPRO 3 (Cai, Thissen, & du Toit, 2015), which can estimate multidimensional item parameters and examinee abilities. Due to its nascence, little research has been done on the ability of IRTPRO 3 to estimate multidimensional item parameters and examinee abilities in comparison to other available software. This study investigates the capability of Mplus 7 and IRTPRO 3 to recover multidimensional item parameter and examinee ability levels under different conditions, including estimation techniques, test lengths, sample sizes, correlations and test structures. The results show that certain estimation techniques available in each software package work well under the given conditions. There are, however, some instances where each estimation techniques encountered challenges. Practitioners should take into consideration the test model and the information needed when selecting the most appropriate software and estimation technique.

Share

COinS