Education Commons^™

Impacts Of Model Misspecification On Individual-Level Differential Item Functioning Detection In Multilevel Data Using Hierarchical Generalized Linear Model, Yue Yin

USF Tampa Graduate Theses and Dissertations

Hierarchical generalized linear modeling (HGLM) has become more and more popular in detecting differential item functioning (DIF) items in multilevel contexts due to its flexibility and efficiency. This dissertation aims to investigate and compare four different HGLM DIF detection models under various DIF conditions with multilevel data. Three studies were conducted in this dissertation. The first two studies were simulation studies and the third one was an empirical study. Simulation Study One compared the performance of two-level HGLM and three-level HGLM with the individual-level covariate (HGLM-IL) on detecting individual-level DIF items with two-level and three-level data structures. Simulation Study Two …

Detection Of Gender-Related Differential Item Functioning (Dif) In The Mathematics Subtests In Turkey, Zeynep Merve Sapmaz

USF Tampa Graduate Theses and Dissertations

The main goal of this study was to investigate differential item functioning by gender in the Fundamental Mathematics (FMS) and Mathematics subtests (MS) of the MSPC-2018 Higher Education Institutions Examination. Each test consists of 40 items and for both subtests random samples of 10.000 students were received from the MSPC separately. To compare non-IRT (Classical Test Theory) and Item Response Theory (IRT) approaches, Cochran-Mantel-Haenszel (C-M-H), Logistic Regression (LR), and 2-PL IRT-LR statistics were used.

For the FMS, C-M-H, LR, and 2-PL IRT-LR procedures identified 18, 16, and 10 out of 40 items that had DIF, respectively. Based on the non-IRT …

The Empirical Selection Of Anchor Items Using A Multistage Approach, Brandon Craig

USF Tampa Graduate Theses and Dissertations

The purpose of this study was to determine if using a multistage approach for the empirical selection of anchor items would lead to more accurate DIF detection rates than the anchor selection methods proposed by Kopf, Zeileis, & Strobl (2015b). A simulation study was conducted in which the sample size, percentage of DIF, and balance of DIF were manipulated. The outcomes of interest were true positive rates, false positive rates, familywise false positive rates, anchor contamination rates, and familywise anchor contamination rates. Results showed the proposed multistage methods produced lower anchor contamination rates than the non-multistage methods under some conditions, …