What I meant is that these characteristics were observed in only the small number of workplaces that have employed me. And I could have gone on to correlate th

What I meant is that these characteristics were observed in only the small number of workplaces that have employed me. And I could have gone on to correlate these behaviors to other characteristics but purposely didn't do that. More importantly, I could elicit corroboration from others who worked in these locations including the woman who first pointed this trend out to me. Of course, the better technique would be to collect aanonymized data from a wide range of HR departments and look for statistical trends. When complete you'd still see biases ... just like only studying trends by analyzing Twitter.