On the Relationship Between the Developer’s Perceptible Ethnicity and the Evaluation of Contributions in GitHub

Abstract

Context: Open Source Software (OSS) projects are typically the result of collective efforts performed by developers with different backgrounds. Although the quality of developers' contributions should be the only factor influencing the evaluation of the contributions to OSS projects, recent studies have shown that diversity issues affect the acceptance or rejection of their contributions. Objective: This thesis assists this emerging state-of-the-art body on diversity research with the first empirical study that analyzes how perceptible ethnicity relates to the evaluation outcome of the contributions in GitHub. Methodology: We performed a large-scale quantitative analysis of the relationship between developers' perceptible ethnicity and the evaluation of their contributions. We extracted the perceptible ethnicity of developers from their names in GitHub using the tool, Name-Prism, and applied regression modelling of pull request data from GHTorrent and GitHub. Results: We observe that (1) among the developers whose perceptible ethnicity was captured by the tool, only 16.56\% of contributors were perceptible as Non-White; (2) contributions from developers perceived as White have the highest acceptance rate; (3) being perceptible as White have a positive, and being perceptible as Asian, Pacific Islander, Hispanic, or Black might \fix{have} a negative influence on the evaluation of the contributions. Conclusion: While we did not observe any conscious bias against any group, our initial analysis leads us to believe that there may exist an unconscious bias against developers with ethnicity perceptible as Non-White. Thus, our findings reinforce the need for further studies on ethnic diversity in software engineering to foster a healthier OSS community.