This seems unlikely to be true in this case because the study was looking at github projects, and it seems unlikely the sample had enough code from "uptight" work places, to have an affect one way or another