Search

Promptly addressing students’ help requests on their programming assignments has become more and more challenging in computer science education. Since the pandemic, most instructors use online office hours to answer questions. Prior studies have shown increased student participation with online office hours. This popularity has led to significantly longer wait times in the office hours queue, and various strategies for selecting the next student to help may impact wait time. For example, prioritizing students who have not been seen on the day of the deadline will extend the wait time for students who are frequently rejoining the queue. To better understand this problem, we explored students’ behavior when they are waiting in the queue. We investigate the amount of time students are willing to wait in the queue by modeling the distribution of cancellation time. We find that after waiting for 49 minutes, most students will cancel their help request. Then, we looked at students’ coding actions during the waiting period and found that only 21% of students have commits while waiting. Surprisingly, students who waited for hours did not commit their work for automated feedback. Our findings suggest that time in the queue should be considered in addition to other factors like last interaction when selecting the next student to help during office hours to minimize canceled interactions.

Few studies have analyzed students’ teamwork (pairwork) habits in programming projects due to the challenges and high cost of analyzing complex, long-term collaborative processes. In this work, we analyze student teamwork data collected from the GitHub platform with the goal of identifying specific pair teamwork styles. This analysis builds on an initial corpus of commit message data that was manually labeled by subject matter experts. We then extend this annotation through the use of self-supervised, semi-supervised learning to develop a large-scale annotated dataset that covers multiple course offerings from a second-semester CS2 course. Further, we develop a series of predictive models to automatically identify student teamwork styles. Finally, we compare trends in students’ performance and team selection for each teamwork style to see if any of them reflected better student outcomes or different trends of help-seekingamong students. Our analysis showed that applying self-supervised semi-supervised methods helps us to label larger subsets of data automatically and maintains and even sometimes improves the performance of the fully supervised models on a held-out validation set. Our analysis also showed that members of teams in which all members have significant contributions tend to have better performance in class, but their help-seeking behaviors are not significantly different.

Demand for education in Computer Science has increased markedly in recent years. With increased demand has come to an increased need for student support, especially for courses with large programming projects. Instructors commonly provide online post forums or office hours to address this massive demand for help requests. Identifying what types of questions students are asking in those interactions and what triggers their help requests can in turn assist instructors in better managing limited help-providing resources. In this study, we aim to explore students’ help-seeking actions from the two separate approaches we mentioned before and investigate their coding actions before help requests to understand better what motivates students to seek help in programming projects. We collected students’ help request data and commit logs from two Fall offerings of a CS2 course. In our analysis, we first believe that different types of questions should be related to different behavioral patterns. Therefore, we first categorized students’ help requests based on their content (e.g., Implementation, General Debugging, or Addressing Teaching Staff (TS) Test Failures). We found that General Debugging is the most frequently asked question. Then we analyzed how the popularity of each type of request changed over time. Our results suggest that implementation is more popular in the early stage of the project cycle, and it changes to General Debugging and Addressing TS Failures in the later stage. We also calculated the accuracy of students’ commit frequency one hour before their help requests; the results show that before Implementation requests, the commit frequency is significantly lower, and before TS failure requests, the frequency is significantly higher. Moreover, we checked before any help request whether students changed their source code or test code. The results show implementation requests related to higher chances of source code changes and coverage questions related to more test code changes. Moreover, we use a Markov Chain model to show students’ action sequences before, during, and after the requests. And finally, we explored students’ progress after the office hours interaction and found that over half of the students improved the correctness of their code after 20 minutes of their office hours interaction addressing TS failures ends.