Measurement invariance in factor analytic and item response theory framework

0580475 - ÚI 2024 RIV BE eng A - Abstract
Martinková, Patrícia - Pavlech, Ján
Measurement invariance in factor analytic and item response theory framework.
The 10th European Congress of Methodology (EAM2023) Book of Abstracts. Ghent: Ghent University, 2023. s. 193-193.
[EAM2023: European Congress of Methodology /10./. 11.07.2023-13.07.2023, Ghent]
R&D Projects: GA ČR(CZ) GA21-03658S
Institutional support: RVO:67985807
Keywords : measurement invariance * differential item functioning * item response theory
OECD category: Statistics and probability
https://eam2023.ugent.be/images/eam2023_abstracts_book.pdf

ZÁKLADNÍ ÚDAJE: The 10th European Congress of Methodology (EAM2023) Book of Abstracts. Ghent: Ghent University, 2023. s. 193-193. [EAM2023: European Congress of Methodology /10./. 11.07.2023-13.07.2023, Ghent]. ABSTRAKT: The relations between binary factor analysis (FA) model and 2-parameter item response theory (IRT) model were established some time ago (Takane & de Leeuw 1987, Kamata & Bauer 2008). However, when accounting for multiple groups and testing for between-group differences on an item level, the methods based on FA and IRT frameworks are usually considered supplementary rather than equivalent. This work focuses on relations between FA models used for testing measurement invariance and IRT models used for testing differential item functioning (DIF). We focus on equivalence of the two types of group-specific models under a set of conditions. Different types of invariance are discussed, namely configural, weak (metric), strong (scalar), and partial, together with models testing for non-uniform DIF, uniform DIF, and no DIF in all items, respectively. Computational aspects and extensions to multiple groups and ordinal items are considered. A real data example is provided to demonstrate the presented relations.
Permanent Link: https://hdl.handle.net/11104/0349244