Classification of Chemical Substances Using Particulate Representations of Matter: An Analysis of Student Thinking

We applied a mixed-method research design to investigate the patterns of reasoning used by novice undergraduate chemistry students to classify chemical substances as elements, compounds, or mixtures based on their particulate representations. We were interested in the identification of the representational features that students use to build a classification system, and in the characterization of the thinking processes that they follow to group substances in different classes. Students in our study used structural and chemical composition features to classify chemical substances into elements, compounds, and mixtures. Many of the students’ classification errors resulted from strong mental associations between concepts (e.g., atom-element, molecule-compound) or from lack of conceptual differentiation (e.g., compound-mixture). Strong concept associations led novice students to reduce the number of relevant features used to differentiate between substances, while the inability to discriminate between two concepts (conceptual undifferentiation) led them to pay too much attention to irrelevant features during the classification tasks. Comparisons of the responses to classification tasks of students with different levels of expertise in chemistry indicate that some of these naïve patterns of reasoning may be strengthened by, rather than weakened by, training in the discipline.