I was part of a research team (class project; never published) that did something like this about a dozen years ago. We were specifically looking at hiring-rel

I was part of a research team (class project; never published) that did something like this about a dozen years ago. We were specifically looking at hiring-related gender bias among different education levels. Our study showed non-statistically significant bias among those without a bachelor's degree, and statistically significant at a moderate correlation level among bachelor's degree holders. The worst were the subjects with graduate degrees; not only was it a strong relationship at a statistically significant level, it was as extreme as our methodology could detect. And it went both ways (e.g., a female identified applicant wouldn't be considered for a mechanic position, but a male identified applicant wouldn't be considered for a nursing position either). Respondents weren't supposed to put specific information on their responses, but one prominent individual did. Our curiosity got the best of us, so after we input and calculated our metrics, we looked at the identifiable subject's responses. This individual was in charge of EEO and non-discrimination policies at a research 1 institution, and she was the most gender discriminatory individual in the study!!!