Code quality Issues in Student Programs

Abstract

Because low quality code can cause serious problems in software systems, students learning to program should pay attention to code quality early. Although many studies have investigated mistakes that students make during programming, we do not know much about the quality of their code. This study examines the presence of quality issues related to program !ow, choice of programming constructs and functions, clarity of expressions, decomposition and modularization in a large set of student Java programs. We investigated which issues occur most frequently, if students are able to solve these issues over time and if the use of code analysis tools has an e"ect on issue occurrence. We found that students hardly #x issues, in particular issues related to modularization, and that the use of tooling does not have much e"ect on the occurrence of issues.