Abstract
Advancing the field of discipline-based education research (DBER) requires developing theories based on outcomes that integrate across multiple methodologies. Here, we describe mediation analysis with structural equation modeling as one statistical tool that allows us to further examine mechanisms underlying well-documented trends in higher education. The use of mediation analysis in educational settings is particularly powerful, as learning outcomes result from complex relationships among many variables. We illustrate how mediation analysis can enhance education research, addressing questions that cannot be easily reached otherwise. We walk through critical steps to guide decision-making in mediation analysis and apply them to questions using real data to examine performance gaps in large introductory courses in biology. Through the use of mediation analysis with structural equation modeling, we add to a growing body of research that shows diverse quantitative approaches support evidence-based teaching in higher education.