Latent Class Analysis in Educational Research: A Conceptual Overview and R-Based Application

Latent Class Analysis (LCA) is a statistical method used to uncover unobserved subgroups within a population based on individuals’ responses to categorical indicators. In educational technology research, LCA provides a powerful approach to identify hidden patterns in teachers’ or students’ behaviors, attitudes, and engagement with technology. For example, you want to understand whether teachers' practice of using open educational resources (OER) can be grouped based on the barriers that they encountered - some may be restrained from school support, others lack professional development, and some just do not have the capacity of effectively using OERs. With LCA, researchers can classify teachers into subgroups based on those barriers in order to identify group-specific strategies to support those teachers. This method offers a data-driven way to move beyond single-variable analyses and capture the heterogeneity underlying observed responses. The chapter introduces the conceptual foundations of LCA, including the assumption of local independence, and demonstrates its application in R using the poLCA package. Step-by-step examples illustrate how to specify models, interpret class profiles, and include covariates and distal outcomes. By applying LCA, educational researchers can better understand diverse user populations, improve targeted interventions, and advance theory-driven analyses of technology use and learning outcomes.