STRUCTURAL EQUATION MODELING (SEM) IN BIOLOGY EDUCATION RESEARCH: MOVING BEYOND “WHAT WORKS” TO UNDERSTAND HOW IT WORKS, AND FOR WHOM (Invited Presentation)

Advancing the field of Discipline-Based Education Research (DBER) requires developing theories based on outcomes that integrate across multiple methodologies. The use of structural equation modeling (SEM) is one statistical tool that allows us to address phenomena underlying well-documented trends in higher education. Here I will describe two examples of SEM approaches that address performance gaps in large introductory ‘gateway’ courses in biology. In the first example,we found that unexpected influences underlie exam performance across ten large introductory science courses (N > 1200).Through SEM, we found that for female students only, and regardless of their academic standing,test anxiety negatively impacted exam performance, while interest in the course-specific science topics increased exam performance. These findings challenge traditional approaches that evaluate student knowledge -- particularly those that 'reprimand' women who do poorly on high-stakes assessments like exams that may not be relevant to actual professional skills. In the second example, we observed a transition to active learning pedagogy closed the gap in learning gains between underrepresented minority (URM) students and non-URM students (N = 421). Through SEM, we demonstrated that for URM students, an increase in science self-efficacy mediated the positive effect of active learning pedagogy on two different metrics of student performance. These examples add to a growing body of research that uses diverse quantitative approaches to support varied and inclusive teaching as one pathway to a diversified STEM workforce.