CAREER: Improving Undergraduate Computing Education by Scaffolding Write Code Problems with Automatically Generated, Personalized, and Adaptive Parsons Problems

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2). This project aims to serve the national interest by creating and testing tools to improve student success in introductory programming courses. Programming courses usually require students to practice by writing complete programs. However, this can be very difficult for students who are new to programming. Struggling while writing code can increase the risk that students will fail a course or leave the field. This project will help students succeed in writing complete programs by providing them with a hint in the form of a mixed-up code problem. In these problems, students place mixed-up blocks of code in the correct order. Prior research has shown that students solve these problems faster than writing complete programs and yet still learn a similar amount. This project will automatically generate two types of mixed-up code problems from student-written code: (1) a problem based on the most common student-written solution, and (2) a personalized problem based on a student's incorrect solution. The goals of this project are to investigate the effects of two types of mixed-up code problems, also called Parsons problems, with respect to student perceptions, learning gains, efficiency, problem completion rates, self-efficacy, and retention. Methods include think-aloud within-subjects studies, between subject studies, and surveys. Effects will be tested in multiple educational contexts: an undergraduate Python course, an online Python course, programs to help secondary students succeed, an undergraduate Java course, and a high-school level Java course. The tools and generated problems will be integrated into Runestone, an open-source ebook platform that is already used by hundreds of institutions and tens of thousands of students. Research results will be disseminated via blogs, conference papers, and workshops for computing instructors. This project will add new knowledge in the effort to scaffold learning for students who are struggling while writing complete programs. Techniques used in this project, if successful, could be used in other STEM subjects to improve success and retention rates. This CAREER award is supported by NSF's IUSE:EDU Program which supports research and development projects to improve the effectiveness of STEM education for all students. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.