The researchers came to the conclusion that code written by women, on average, receives higher scores than written by men. But only on the condition that the estimator is unknown which gender was a programmer. The study was conducted on the basis of the analysis of the behavior of users of Github. “We expected that the proposals on introducing amendments to the code made by women, people accept less readily than men,” says the research report. It turned out that it is not.
Suggestions about the edits made by the women, conversely, are accepted actively (in 78.6% of cases) — but only in those cases where the sex has made these proposals, the user cannot immediately identify. The edits proposed by the men, meet approval in 74.6% of cases. According to the study, this is true for the ten most popular programming languages. In those cases when recommendations about changes contribute uniquely identifiable women, users are less willing to accept them.