Style Anomalies Can Suggest Cheating in CS1 Programs
Open Access
- 14 March 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1624-1625
Abstract
Student cheating on at-home programming assignments is a well-known problem. A key contributor is externally obtained solutions from websites, contractors, and recently generative AI. In our experience, such externally obtained solutions often use coding styles that depart from a class's style, which we call "style anomalies". Examples of style anomalies include using untaught or advanced constructs like pointers or ternary operators or having different indenting or brace usage from the class style. We developed a tool to automatically count style anomalies in student code submissions. We used this tool to find suspected cheating in student submissions for lab assignments across five terms of CS1. This poster presents our findings: Some student submissions were suspected of cheating due to high style anomaly counts and were not flagged as suspicious by a code similarity checker. With the rise of externally obtained solutions from websites, contractors, and generative AI, style anomalies may become an important complement to similarity checking for detecting cheating.Keywords
Funding Information
- National Science Foundation (2111323,2313793)
This publication has 4 references indexed in Scilit:
- The Benefits and Challenges of ChatGPT: An OverviewFrontiers in Computing and Intelligent Systems, 2022
- PROGRAMMING STYLE ON SOURCE CODE PLAGIARISM AND COLLUSION DETECTIONInternational Journal of Computing, 2020
- Style Analysis for Source Code Plagiarism Detection — An Analysis of a Dataset of Student Coursework2017
- A coding style-based plagiarism detection2012