Proceedings of the Annual Meeting of the Cognitive Science Society

This exploratory study looks at analogical transfer in the context of computer science education, particularly during the transition from a visually scaffolded block-based language (i.e., Scratch) to a syntactically complex text-based language (i.e., Python). Using structure mapping theory, we compared surface similarity and structural relationships of fundamental programming constructs (e.g., variables, iteration) in the two languages to predict which students would most likely transfer. 76 students (ages 8-14) proficient in Scratch with little Python experience completed an assessment measuring transfer of these constructs with minimal instruction. Students who saw more similarity among the languages performed better on the Python items, suggesting they perceived structural similarities. However, we did not find evidence that students transfer after controlling for general programming ability. Findings suggest that students need more targeted instruction to conceptualize structural relationships and support robust analogical transfer between block-based and text-based programming languages.