Modeling students’ epistemic cognition in undergraduate chemistry courses: a review†

Thinking about knowledge and knowing (i.e., epistemic cognition) is an important part of student learning and has implications for how they apply their knowledge in future courses, careers, and other aspects of their lives. Three classes of models have emerged from research on epistemic cognition: developmental models, dimensional models, and resources models. These models can be distinguished by how value is assigned to particular epistemic ideas (hierarchy), how consistent epistemic ideas are across time and/or context (stability), and the degree to which people are consciously aware of their own epistemic ideas (explicitness). To determine the extent to which these models inform research on epistemic cognition in chemistry education specifically, we reviewed 54 articles on undergraduate chemistry students’ epistemologies. First, we sought to describe the articles in terms of the courses and unit of study sampled, the methods and study designs implemented, and the means of data collection utilized. We found that most studies focused on the epistemic cognition of individual students enrolled in introductory chemistry courses. The majority were qualitative and employed exploratory or quasi-experimental designs, but a variety of data collection methods were represented. We then coded each article for how it treated epistemic cognition in terms of hierarchy, stability, and explicitness. The overwhelming majority of articles performed a hierarchical analysis of students’ epistemic ideas. An equal number of articles treated epistemic cognition as stable versus unstable across time and/or context. Likewise, about half of the studies asked students directly about their epistemic cognition while approximately half of the studies inferred it from students’ responses, course observations, or written artifacts. These codes were then used to infer the models of epistemic cognition underlying these studies. Eighteen studies were mostly consistent with a developmental or dimensional model, ten were mostly aligned with a resources model, and twenty-six did not provide enough information to reasonably infer a model. We advocate for considering how models of epistemic cognition—and their assumptions about hierarchy, stability, and explicitness—influence the design of studies on students’ epistemic cognition and the conclusions that can be reasonably drawn from them.