How do mathematics education researchers treat ordinal variables in structural equation models?

Lee, Gabriel Chun-Yeung

Abstract

Mathematics education researchers often work with observations, such as survey responses and tests, related to unobserved abilities or characteristics, such as attitudes and beliefs. Those observations are often measured as ordinal variables. A critical question arises when researchers attempt to implement statistical models to analyse ordinal data, given many statistical analysis techniques assume the data analysed to be continuous: should ordinal data be treated as continuous data? Focusing on structural equation models, including confirmatory factor analysis, this presentation discusses an ongoing debate on the treatment of ordinal data and reports a review on the existing practices of conducting and reporting the analysis of structural equation models, published in five mathematics education journals.

Citation

Lee, G. C.-Y. (2024, July). How do mathematics education researchers treat ordinal variables in structural equation models?. Presented at 15th International Congress on Mathematical Education, Sydney, Australia

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 15th International Congress on Mathematical Education
• Start Date: Jul 7, 2024
• End Date: Jul 14, 2024
• Acceptance Date: Dec 31, 2023
• Publication Date: Jul 12, 2024
• Deposit Date: Feb 20, 2025
• Peer Reviewed: Peer Reviewed
• Public URL: https://nottingham-repository.worktribe.com/output/36578671
• Related Public URLs: https://icme15.org/
• External URL: https://icme15.org/