Polytechnic University of Valencia Congress, Third International Conference on Higher Education Advances

Font Size: 
Automated Program Analysis for Novice Programmers
Ansgar Fehnker, Tim Blok

Last modified: 07-06-2017

Abstract


This paper describes how to adapt a static code analyzer to provide feedback novice programmers and their teachers. Current analyzers have been built to give feedback to experienced programmers who work on software projects or systems. The type of feedback and the type of analysis of these tools focusses on mistakes that are relevant within that context, and help with debugging  software system. When teaching novice programmers this type of advice is often not particularly useful. It would be instead more useful to use these techniques to identify problem in the understanding of students of important programming concepts.

This paper first explores in what respect static analyzers support the learning and teaching of programming, and what can be implemented based on existing static analysis technology. It presents an extension of static analyzer PMD to create feedback that is more valuable to novice programmers. To answer the question if these techniques are able to find conceptual mistakes that are characteristic for novice programmers make, we ran it over a number of student projects, and compared these results with publicly available mature software projects.


Full Text: PDF