Gender bias in hiring at Carleton is real, plus a possible solution (26 September 2014)

Several faculty members in the biology department criticized my assertion in earlier blog postings regarding gender bias in hiring. Their criticisms took two forms: (1) that I provided no data showing gender bias and (2) even if data existed to support my assertion, that I failed to provide possible solutions. Several individuals even asked me to retract my assertion of gender bias, which I respectfully decline to do. You should never counteract putatively offensive speech with censorship, but instead counteract it with more speech. And, to present a balanced portrait, several biology faculty members were supportive of my views. Regardless, I maintain the belief that we have a substantial problem with gender bias in hiring faculty members at Carleton, especially in science and engineering, and thus here provide some data and a possible solution.

The biology department provides a small sample size, but still seems to show a gender bias. Since I started at Carleton in July 2006, there have been three tenure-track hires and two hires of instructors. All five individuals self-identify as male. Given that half of the population is female and roughly half the recently earned PhDs in biology are female, ceteris paribus, the probability of all five hires being male is one in 32. That could be a random effect, but seems unlikely. Of the three tenure-track hires, one was a Banting post-doc, so there were no competitive interviews of multiple candidates for his line. But for the other two tenure-track positions, the biology department interviewed a total of nine candidates, only one of which was female. Assuming, possibly erroneously, that the applicant pool had a 50:50 sex/gender ratio, the probability of choosing one or fewer women out of nine candidates is approximately one in 50 (actually, 10/512). That could be a random effect, but also seems unlikely. The biology department website lists 30 faculty members. Of these, only seven are female, i.e. less than one-quarter (but please see the first footnote). Assuming equal probability of hiring women and men, there is approximately a one in 643 million chance of seven or fewer women in a department of 30 people (see the second footnote for the formula). These sorts of statistics seem to provide a prima facie case for gender bias.

I am not complaining about the individual people that the biology department hired since 2006 or even before that. They all seem amazing. I am also not stating what the cause of this apparent gender bias is, which is possibly unconscious and not an intentional bias. Nor is a gender bias in hiring faculty members endemic to the biology department. The university’s equity services office keeps track of gender of faculty members, in a table titled “Summary Workforce Data”, which shows that Carleton does worse than average, over 10% worse than the average for other universities, for hiring women into non-management faculty positions. The problem is that some of my colleagues believe there is no gender bias at all, even after looking at the above data.

What can we do to increase representation of female faculty members in science and engineering, as well as in the rest of the university? Mary Ann Mason gives a few good ideas in her piece in Chronicles of Higher Education. One of her major recommendations was to provide better child-care. Unfortunately, Carleton’s master plan shows a new building displacing the existing child-care centre. This seems like a step backwards in terms of recruiting and retaining female faculty members.

I want to highlight one other idea for improving gender diversity of faculty, an idea that was mentioned by our distinguished guest earlier this week, the 19th prime minister of Canada, Kim Campbell, who admitted to stumbling upon this idea after first being elected as an MLA in a double-member riding in Vancouver Centre. Well, she was not talking about gender equity in academic appointments, but rather the House of Commons. But the same approach is applicable, although nobody seemed to mention the transferability of her idea after her lovely repartée with the audience on Monday. Kim Campbell’s proposal was to hire a pair of tenure-track people with each search. Interview double the usual number of people and then hire the best female and the best male candidate from that search. This means that searches would happen half as often, but that really is not a problem insofar as deans can juggle hires amongst their many units. For Kim Campbell, this translated into two-member ridings, sometimes called double member ridings, with the proviso that the two elected members must be of different genders (see footnote 3).

While I really like Kim Campbell’s idea, I should also highlight pitfalls. First, as our colleague Tom Smallwood has rightly pointed out, as undoubtedly also would Lara Karaian and Dan Irving, there are more than two genders. Second, we should not require job applicants to self-identify gender. And worse, we should not try to surmise their gender, as I very reluctantly did in the second paragraph of this posting. In fact, in “Fundamental differences between females and males?” I showed that you cannot consistently distinguish females from males, even if you forced candidates to interview naked. Also see Anne Fausto-Sterling’s discussions of intersex and many different authors discussions of trans sex/gender (e.g. Transgender 101). Third, even if there would magically (dystopian magic) be a gender dichotomy, this would not solve problems with other forms of hiring inequity, such as racial inequities, where so-called ‘at-large’ ridings are often created to water-down minority votes (see this week’s L.A. Times). Let me thus end with some questions, for which I am not entirely sure of the answers. Even if not perfect, would Kim Campbell’s proposal of hiring pairs of people help correct our seemingly obvious gender bias in faculty members? Would the benefits outweigh the costs? What can or should be done for other disadvantaged groups? Please let Carleton and me know what you think. Thanks.


  1. My estimate of number of female faculty members and female job candidates is largely based on gross gender stereotypes. I did not ask human resources for gender data on these individuals. More poignantly, I did not ask these individuals for their preferred gender pronoun (PGP). I am therefore guilty of surmising gender and sincerely apologize if I have consequently insulted anybody. This is the principal reason that I did not present gender data in my previous blog postings and only do so here with great reluctance. Mea culpa.
  2. [Sum from n=0,…,7 of C(30,n)]/2^30, where C(30,n) are the number of combinations of 30 take n.
  3. I have not conducted a thorough search to see who else might have suggested this alternative of two-member ridings with one female and one male member.
  4. Thanks to many colleagues for discussions on these matters, especially faculty in biology and Melissa Haussman. While I take full blame for what is written here, they deserve most of the credit.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: