Publications

Group highlights

**At the end of this page, you can find the full list of publications.

Generating Planning Feedback for Open-Ended Programming Exercises with LLMs

Can we provide feedback on high-level planning approaches, even when students have syntax errors in their submissions? In this work, we propose a method for leveraging LLMs to detect patterns in correct and incorrect submissions, enabling a new dimension of feedback in addition to test-case correctness.

Mehmet Arif Demirtaş, Claire Zheng, Max Fowler, Kathryn Cunningham

26th International Conference on Artificial Intelligence in Education (AIED 2025)

PLAID: Supporting Computing Instructors to Identify Domain-Specific Programming Plans at Scale

How do instructors identify common patterns in a computing domain? Can we help them by automating information gathering with LLMs? In this work, we interviewed 10 computing educators to understand their challenges in programming plan identification, and designed PLAID for supporting them in information gathering and refinement tasks through design workshops. An user study with 12 participants showed that PLAID helps instructors to identify programming plans faster, with smaller workload, and with more satisfaction.

Yoshee Jain*, Mehmet Arif Demirtaş*, Kathryn Cunningham

43rd ACM Conference on Human Factors in Computing Systems (CHI 2025)

Validating, Refining, and Identifying Programming Plans Using Learning Curve Analysis on Code Writing Data

Can programming plans represent the key skills in introductory programming? In this work, we applied learning curve analysis using plans as knowledge components and found that some plans are associated with discrete skills acquired by students.

Mehmet Arif Demirtaş, Max Fowler, Nicole Hu, Kathryn Cunningham

20th ACM Conference on International Computing Education Research (ICER 2024)

Profiling Conversational Programmers at University: Insights into their Motivations and Goals from a Broad Sample of Non-Majors

How prevalent are conversational programmers, and how do their attitudes and interests differ from end-user programmers? In this study, we found that conversational programmers are more common among non-majors than end-user programmers, but they face more challenges in their motivation for computing.

Jinyoung Hur, Kathryn Cunningham

20th ACM Conference on International Computing Education Research (ICER 2024)

Reexamining Learning Curve Analysis in Programming Education: The Value of Many Small Problems

Can we automatically detect student development on language structures using abstract syntax trees (AST)? In this replication study, we found that AST nodes can be used to track student learning to some extent.

Mehmet Arif Demirtaş, Max Fowler, Kathryn Cunningham

17th International Conference on Educational Data Mining (EDM 2024)

 

Full List of Publications

Jinyoung Hur, Michael M Kang, Junmee Park,and Kathryn Cunningham. 2025. Applying the Model of Interest Development to Understand Why Non-CS Majors Decide to Persist in or Leave Computing. 25th Koli Calling International Conference on Computing Education Research (Koli Calling ‘25).

Mehmet Arif Demirtaş, Claire Zheng, Max Fowler, Kathryn Cunningham. 2025. Generating Planning Feedback for Open-Ended Programming Exercises with LLMs. 26th International Conference on Artificial Intelligence in Education (AIED 2025).

Jinyoung Hur, Michael M Kang, and Kathryn Cunningham. 2025. Patterns of Major Switching and Persistence in Computing among Students with Disabilities (Poster). In Proceedings of the 56th ACM Technical Symposium on Computer Science Education V. 2 (SIGCSE 2025). To learn more, click here.

Mehmet Arif Demirtaş, Claire Zheng, and Kathryn Cunningham. 2025. Detecting Programming Plans in Open-ended Code Submissions (Poster). In Proceedings of the 56th ACM Technical Symposium on Computer Science Education V. 2 (SIGCSE 2025). To learn more, click here.

Yoshee Jain*, Mehmet Arif Demirtaş*, Kathryn Cunningham. 2025. PLAID: Supporting Computing Instructors to Identify Domain-Specific Programming Plans at Scale. 43rd ACM Conference on Human Factors in Computing Systems (CHI 2025).

Mehmet Arif Demirtaş, Max Fowler, Nicole Hu, Kathryn Cunningham. 2024. Validating, Refining, and Identifying Programming Plans Using Learning Curve Analysis on Code Writing Data. 20th ACM Conference on International Computing Education Research (ICER 2024).

Jinyoung Hur, Kathryn Cunningham. 2024. Profiling Conversational Programmers at University: Insights into their Motivations and Goals from a Broad Sample of Non-Majors. 20th ACM Conference on International Computing Education Research (ICER 2024).

Mehmet Arif Demirtaş, Max Fowler, Kathryn Cunningham. 2024. Reexamining Learning Curve Analysis in Programming Education: The Value of Many Small Problems. 17th International Conference on Educational Data Mining (EDM 2024).

Hongxuan Chen, Ang Li, Geoffrey Challen, and Kathryn Cunningham. 2024. Implementation of Split Deadlines in a Large CS1 Course.. In Proceedings of the 55th ACM Technical Symposium on Computer Science Education V. 1 (SIGCSE 2024).

Kathryn Cunningham, Miranda C. Parker, and Jonathan Zhang. 2023. The Landscape of Computer Science Education Courses: A Syllabi Analysis. 23rd Koli Calling International Conference on Computing Education Research (Koli Calling ‘23).

Kathryn Cunningham, Yike Qiao, Alex Feng, and Eleanor O’Rourke. 2022. Bringing High-level Down to Earth: Gaining Clarity in Conversational Programmer Learning Goals. 53rd ACM Technical Symposium on Computer Science Education (SIGCSE 2022).

Kathryn Cunningham, Barbara Ericson, Rahul Agrawal Bejarano, and Mark Guzdial. 2021. Avoiding the Turing Tarpit: Learning Conversational Programming by Starting from Code’s Purpose. CHI Conference on Human Factors in Computing Systems (CHI ’21).