Improving Middle School Science Learning Using Diagrammatic Reasoning

We explored whether existing curricula can be adapted to increase students' skills at comprehending and learning from diagrams using an intervention delivered by regular middle-school teachers in intact classrooms. Ninety-two teachers in three states implemented our modified materials over six curricular units from two publishers: Holt (a reading-focused curriculum) and Full Option Science System (FOSS) (an inquiry-focused curriculum). Results were compared between two interventions-one based on selected principles of cognitive science (cognitive-science-based) that included instruction in diagram comprehension and one providing professional development in science content only (content-only)-and a business-as-usual control. We analyze posttest items involving different degrees of reliance on diagrams to show how instruction in diagram comprehension can improve comprehension of diagrams during reasoning. At the classroom level, there were significant advantages of the cognitive-science-based intervention over both content-only and business-as-usual with large effect sizes in all FOSS units (d = 0.41-0.52), but only one Holt unit (d = 0.11). Analyses by type of diagram suggested these effects were largest for transfer to diagrams from an uninstructed domain. Further analyses of high-stakes state test scores available for the participants implementing the Holt units showed improved use of diagrams (d = 0.45-0.66). Results suggest that making sense of visualizations (diagrams) is not effortless and self-evident, but that students who receive supports for comprehending can improve their comprehension, and the learning can transfer to new domains.