One of the most interesting conclusions of the study was that the evaluator's gender, specialization, or age, did not correlate with bias. An old male physics p

One of the most interesting conclusions of the study was that the evaluator's gender, specialization, or age, did not correlate with bias. An old male physics professor was as likely as a young female biology professor to discriminate against Jennifer. This story is sort of old news, but this excerpt raising an interesting point. Namely, the use and abuse of the word 'bias'. Forward looking expectations may rightly take into account expectations of future performance. It does not follow that a resume of past performance is gender-neutral for future performance, even if it is exactly the same. In this respect, the bids for future services may still be (statistically) unbiased estimators. That is, it may actually be less likely that (on average) the gender's will produce over an entire career equal productivity. The more interesting question, is thus: is this true? The study is not structured to actually address the real issue. Furthemore, it appears to assume or imply the answer (viz bias will prevent such equal performance). But this is intellectually questionable. Provided a non-zero probability of, for example, curtailing a career to care for children would be enough to produce the observed result, ordinally. Since this assumption is trivially true, by observation, this supports the hypothesis of merely (heterodox) forward looking expectations. This doesn't leave a warm fuzzy feeling, admittedly, and arguably employment managment is to blame for the underlying root causes. But if these issues are genuine concerns that are to be 'fixed,' first they need to be understood. That being said, it has been reported that academics have returned questionaires ranking a (non-existent) law school as "middle of the pack," and they did so based soley on its brand name. Whether this is bias or sloppy thinking (hueristic vs expertise vs presitige bias), is also a good question. The truth probably lies somewhere in the middle