Many of the commentators—let’s ignore their sex for the moment—suggested including women in the Feyerabend conference. Then the question was raised, “but are they of the right quality, status, rank?” That is, do they bring down the average quality of the conference in virtue of their being of inferior status, or, in Vincenzo Politi’s words, not “someone whose work is both relevant to the topic of the conference and also as widely recognized as the work of the invited speakers” (HOPOS-L (...) archive, “CFP: Feyerabend Conference,” Tuesday, July 17, 2012, 14:57:20)? It is extremely important that such a discussion of quality, status, and rank recognize the scourge of evaluation bias and its long-term and pervasive consequences. One well-designed study this past year, published by the National Academy of Sciences, established prominent evaluation bias among both male and female science faculty in their evaluations of a student applying for a managerial job, who was randomly assigned either a male or a female name (Moss-Racusin et al. 2012). These professors examined the qualifications of the students and decided whether to hire them, what salary to give them, and whether to mentor themand howmuch to do so.The resultswere that both male and female scientists hired more men, gave them higher salaries, and offered more mentoring to them, even though themale applications were identical to the female applications. When probed about their reasons for not hiring or mentoring the female applicants, the professors explained that they based their decisions on the inferior competence of the applicant: the female applicants were perceived as less competent by all professors (with identical applications between males and females). This is what “evaluation bias” looks like, and it has been established in many, many contexts since the 1970s—this is only the most recent. (shrink)