Sachel Villafane, Jennifer Lewis, Xiaoying Xu (University of South Florida, USA)
Structural Equation Modeling (SEM) is a statistical approach used to model complex relationships between one or more dependent or independent variables. Different from classical analysis such as regression analysis, SEM combines a measurement and a structural model to find potential interrelationships among latent and/or observed variables. The aim of this presentation is to introduce SEM as a research tool to evaluate and confirm possible theoretical relationships among constructs of interest. This presentation will discuss what SEM is and when its use is appropriate. Research questions that could be answered with SEM as well as the type of data needed for an analysis will be explored. The information presented will give an overview of SEM as a research tool, specifically showcasing how it can be applied for theory-building in chemistry education research.