Cross-fertilization of content and strategy knowledge: Bayesian insights into fostering scientific literacy in physics education

Background Scientific literacy entails domain-specific content knowledge and domain-general strategies for generating and evaluating evidence. The latter include the control-of-variables strategy (CVS). Aims We examined whether domain-specific physics curricula and CVS trainings complement and cross-fertilize one another. Sample In total, 189 primary school students (100 male, 89 female) participated. Method In a randomized controlled field trial with three intervention stages, we studied the interaction of domain-specific inquiry-learning in physics with a CVS training. In stage 1, 104 students received physics lessons in grade 3 or 4, whereas 85 students received regular science lessons without a focus on physics knowledge. In stage 2, in grade 5 or 6, half of each group received a CVS training, whereas the other half participated in a control training. In stage 3, about five months later, all participants received domain-specific lessons on magnetism. Manipulation checks on pre- and posttests indicated the individual effectiveness of all three trainings. Results Bayesian Hierarchical Models revealed cross-fertilization effects not of a binary indicator of participation in an intervention, but of the amount of actual knowledge acquired in the interventions: Physics knowledge gained in stage 1 benefitted the CVS skills gained in stage 2, and these CVS skills, in turn, predicted students’ increase in content knowledge about magnetism in stage 3. Conclusions Our findings support alternating between domain-specific science lessons and the training of domain-general science principles to support the development of comprehensive scientific literacy.