Plagiarism Detection Approaches for Simple Introductory Programming Assignments

Abstract

Learning to program is often perceived as hard by students and some students try to cheat. Plagiarisms are reported to be a huge problem particularly for summative-like assignments (e.g., crediting courses or bonus points). It is important to fight plagiarisms from early on – even for simple assignments. Especially for larger courses tool support is required. This paper provides an overview of features for commonly used plagiarism detection tools, discusses how these can be integrated into existing assessment systems, and how their results relate to each other for two data sets of quite simple assignments. Additionally, these specialized tools are compared with a simple Levenshtein distance approach. The paper also outlines limits on very simple assignments.