Effect of a 2-week scratch intervention in CS1 on learners with varying prior knowledge

Abstract

A large CS1 class often needs to provide scaffolding for novices while keeping advanced learners engaged. Scratch has been shown to be suitable to address a diverse set of requirements. In this study, we determine whether a 2-week Scratch intervention in a CS1 course is useful from two perspectives: i) as a scaffold for novices to learn basic programming concepts and transition to C++, and ii) as a tool for advanced learners to remain engaged and do challenging work. We conducted a field study of 332 first-year undergraduate engineering students, two-thirds of whom were novices. We analyzed student performance on exams and Scratch projects. We administered a survey to determine student perceptions on the usefulness of Scratch. Some key findings of our study are: (i) Novices were able to catch-up to advanced learners in Scratch questions of the type 'Predict the output' and 'Debug the program', (ii) Projects by advanced learners reached 80% of the complexity of 'most loved projects' on the Scratch website, and (iii) 69% of students perceived Scratch to be useful for learning programming concepts and transitioning to C++.