2023 ASEE Annual Conference & Exposition

We present results from a quasi-experimental study to compare an online section to an in-person section of the same computer science class, where all students were residential and physically present on campus. The use of online sections for computer science classes has been primarily driven by two key factors: (1) the difficulties of finding enough large classrooms to accommodate the growing number of computer science majors, and (2) the desire of some students to take classes online, with the flexibility that this entails.

This study adds to the existing body of knowledge on online versus in-person instruction by specifically addressing the question of how effective online instruction is for students who are physically on campus, where they retain their peer support networks and physical learning environments. In this study we considered online and in-person sections of an upper-level required undergraduate computer science course at a large public university in Spring 2022. Students were free to choose to enroll in either the online or in-person sections, both offered in the same semester, by the same instructor, using the same material. The online section had 222 students and the in-person section had 117 students.

The course followed a flipped format with pre-lecture videos. During class time, students were expected to complete computer-based group activities. The in-person section completed these activities in the classroom while the students in the online section had to arrange to complete the activities with their groups at a time of their choosing, either in person or online. All course assessments were completed using an online platform. Students took quizzes and the final exam online using their own computers, which were proctored by course staff online via Zoom.

We compared assessment performance and sense of belonging for students in the online and in-person sections. The two sections had similar average student GPA at the start of the course (3.78 and 3.73 for online and in-person sections respectively; not significant, Cohen’s d = 0.15 and p = 0.22). Controlling for GPA, the two sections had similar quiz averages and final exam average (the online section was 1.3% higher, which was not significant; Cohen’s d = 0.1 and p = 0.4). Disaggregating by gender and URM status showed similar results.

We also conducted pre and post surveys that included measures of student’s sense of belonging in relation to their instructors and classmates (whole class) and in relation to their collaborative groups (within group). Students in the online and in-person sections report similar levels of sense of belonging on average in both surveys (for example, for the post survey the in-person section was 0.05 points higher on a 5-point scale, which was not significant; p = 0.6). Moreover, the whole class sense of belonging statistically increases from the beginning to the end of the semester in both sections.

Based on these results we conclude that online sections of courses for on-campus students can be an effective way to accommodate large class sizes and increased enrollment pressure, while not disadvantaging students in either learning outcomes or sense of belonging. In our study about 2/3 of students preferred to enroll in the online section, substantially reducing the demands for classroom space.