Effect of controlling group heterogeneity on student performance in a graphical programming course

This study assesses whether students perform better in an introductory programming course if they are placed in groups with people of the same or varying programming aptitude. Although the push for teaching code in U.S K-12 classrooms is growing, students arrive to college with a variety of coding experience because only not all schools include coding in their curriculum. Understanding what makes a group productive and efficient is important as the use of active-learning pedagogies across college classrooms grows. The studies presented here aim to 1) predict how well a student will perform in an introductory programming course based on a pre-assessment administered at the beginning of the course and 2) determine whether students perform better in an introductory programming course if they participate in team based projects with people of the same or varying programming aptitude. During the first week of class, students were administered a pre-assessment. Their score determined individual student programming aptitude. Students were assigned to groups based on their performance, where control groups consisted of students who only scored as high programming aptitude or only scored low programming aptitude, and experimental teams consisted of a mix of both. Evidence was found to indicate that the group formation pre-assessment was a good indicator of programming aptitude. It was also found that when completing individual tasks, students who were placed in groups with students of varying programming aptitude outperformed those who were placed in groups with students of the same programming aptitude. The positive findings and their effect on students' abilities to learn a programming language demonstrated that the technique of a short assessment and assigned groups may become an important tool to create more effective groups in large introductory programming courses. Although further testing is required, this technique may shape the future of group formation in large programming courses.