Scaffolding Task Planning Using Abstract Parsons Problems

Interest is growing in the role of metacognition in computing education. Most work to-date has examined metacognitive approaches of novices learning to code. It has been shown that novices navigate through discernible stages of a problem-solving process when working through programming problems, and that scaffolding can be beneficial. In this paper, we describe a novel scaffolding task aimed at guiding novices through a crucial stage of developing and evaluating a problem-solving plan. We presented novices with a problem statement before working through an Abstract Parsons Problem, where the blocks present structural elements rather than complete code, to aid high-level planning before writing code. Comparing groups who experienced this approach with those that did not, revealed that novices who worked on an Abstract Parsons Problem before coding were more successful in solving the task and demonstrated improved metacognitive knowledge related to task planning when asked to identify useful future problem-solving strategies. Our observations from two courses over two years suggest that scaffolding students through a planning step prior to coding can be beneficial for students. We provide directions for future work in exploring strategies for providing this type of guidance, including the use of different types of planning activities, and studying these effects at scale.