Marginalized Maximum Likelihood Estimation for the 1PL-AG IRT Model

Ryoungsun Park; Keenan A Pituch; Jiseon Kim; Barbara G Dodd; Hyewon Chung

doi:10.1177/0146621615574694

Marginalized Maximum Likelihood Estimation for the 1PL-AG IRT Model

Appl Psychol Meas. 2015 Sep;39(6):448-464. doi: 10.1177/0146621615574694. Epub 2015 Apr 5.

Authors

Ryoungsun Park¹, Keenan A Pituch¹, Jiseon Kim², Barbara G Dodd¹, Hyewon Chung³

Affiliations

¹ University of Texas at Austin, USA.
² University of Washington, Seattle, USA.
³ Chungnam National University, Daejeon, South Korea.

Abstract

Marginal maximum likelihood estimation based on the expectation-maximization algorithm (MML/EM) is developed for the one-parameter logistic model with ability-based guessing (1PL-AG) item response theory (IRT) model. The use of the MML/EM estimator is cross-validated with estimates from NLMIXED procedure (PROC NLMIXED) in Statistical Analysis System. Numerical data are provided for comparisons of results from MML/EM and PROC NLMIXED.

Keywords: 1PL-AG; EM; IRT; MML; estimator.