Links between the personalities, styles and performance in computer programming (1611.10169v1)

Abstract: There are repetitive patterns in strategies of manipulating source code. For example, modifying source code before acquiring knowledge of how a code works is a depth-first style and reading and understanding before modifying source code is a breadth-first style. To the extent we know there is no study on the influence of personality on them. The objective of this study is to understand the influence of personality on programming styles. We did a correlational study with 65 programmers at the University of Stuttgart. Academic achievement, programming experience, attitude towards programming and five personality factors were measured via self-assessed survey. The programming styles were asked in the survey or mined from the software repositories. Performance in programming was composed of bug-proneness of programmers which was mined from software repositories, the grades they got in a software project course and their estimate of their own programming ability. We did statistical analysis and found that Openness to Experience has a positive association with breadth-first style and Conscientiousness has a positive association with depth-first style. We also found that in addition to having more programming experience and better academic achievement, the styles of working depth-first and saving coarse-grained revisions improve performance in programming.