Exploring differences in the cognitive and metacognitive scaffolding use of high and low working memory students in block-based programming

This study aimed to clarify how elementary school students' programming performance could be affected by scaffolding types, and how high and low working memory students profited from the intervention. The impact of different scaffolding types was empirically scrutinized with four conditions: cognitive scaffolding, metacognitive scaffolding, a combination of cognitive and metacognitive scaffolding (mixed), or no scaffolding. Students were classified into high and low working memory subjects based on their performance on the working memory test. Two-way MANCOVA results demonstrated that, compared to non-scaffolding instruction, scaffolding had a significantly superior impact on programming knowledge but not on programming skills. The mixed scaffolding group did not show significantly better scores for programming knowledge and skills than the cognitive or metacognitive scaffolding groups. Likewise, no significant difference was found between the metacognitive and cognitive scaffolding groups in programming knowledge and skills. Furthermore, interaction effects were found between scaffolding type and working memory on programming knowledge. High working memory students benefited more from the mixed scaffolding and metacognitive scaffolding. This study offers insightful suggestions for the design and development of different scaffolds to enhance programming instruction for diverse learners, especially learners with different working memory levels.