Skip navigation
The Black Hole

Nice gals finish last: Sexist reference letters or skewed value systems?


Although an identical number of women get a PhD in the life sciences, only 15-20% of tenured positions are secured by women ~Frank Gannon, EMBO 2007

This is a frightening statistic, and I’m not convinced it’s gotten much better since 2007. Gannon goes on to introduce two articles in the same issue of EMBO that present two shocking research article findings:

  1. A blinded study of EMBO fellowship applications showed that women were significantly less likely to win fellowships despite highly similar academic impact
  2. An NIH survey of over 1300 postdoctoral fellows showed that fewer women (1/2) vs. men (2/3) were considering a career as a professor

I think the survey data could probably have been predicted, especially considering the additional data in the paper which cites numerous reasons (family, travel, and spousal employment status) for this trend in the follow-up questions regarding the discouraging factors for such considerations.  As for the EMBO fellowship, the really shocking thing is that it was blinded and normalized for academic impact which means that something else caused men to be selected more often than women.  Much has been speculated in the males being more confident and ambitious in proposal writing or description of oneself (alternatively read as “more willing to embellish/BS or simply less scrupulous”) or women’s desire to value family life over academic life ((Another factor can be extracted from Jennifer Rohn’s recent commentary in Nature on boosting confidence levels of female researchers and how the current low levels of confidence are related to what she labels sexist hiring practices in the media industry)) – but I’m not wholly convinced that these items complete the picture…

When I first read the EMBO articles in 2008 after my colleague (thanks, James Z.) floated them into my Inbox, I had not thought about any additional reasons and simply focused on getting my jaw firmly fixed back into place.  Recently, however, my Inbox was once again infiltrated with a shocking paper that adds a new wrinkle to the situation (thanks, Katharine H) – could reference letters have a gender bias?  When we all sit and devise ways of submitting successful funding applications, one of the things we most panic about arranging, but typically have the least control over in terms of the end product are the reference letters.  As one of the key components to a successful fellowship, academic job, or grant application is extremely good reference letters it stands to reason that the difference between successful male vs. female applicants could come down to what the “respected people in the field” have to say about them.

The summary news article that Katharine forwarded to me referred to a study entitled: Gender and Letters of Recommendation for Academia: Agentic and Communal Differences which describes how males and females who are similar in number of publications, teaching experience, postdoctoral years, and honours have substantial qualitative differences in their letters of reference. It seems that women are described in communal terms that value them as dependable and good team players while men are typically described as independent and driven.  This is all well and good until the second part of the study reveals that communal characteristics have “a negative relationship with hiring decisions in academia”.

Ask yourself – if you were hiring someone for your junior faculty position, who would you hire?  The independent thinker who will push boundaries and influence others or the technically sound team player who can carry out and design experiments and likes to share data and help others?

The really tricky part is when you try to figure out what to do about such a phenomenon.  Perhaps we should try to make referees aware of possible subconscious biases and amend their future letters accordingly… or perhaps we should re-evaluate the qualities we’re looking for in a faculty member?  To me, the list that the study associates with being communal (affectionate, helpful, kind, sympathetic, nurturing, tactful, and agreeable, and behaviors such as helping others, taking direction well and maintaining relationships) seem like the qualities that make someone a good mentor and departmental colleague – sadly, I think the nice guys/gals again end up finishing last.

In fact, I fear that “poster child” mentality is dominating hiring practices at Canadian institutions – who can we hire that has more degrees, more publications, and spends more money on more projects.  How good would that look on the front of our University or Foundation mail out to supporters?  Sometimes I fear that it’s much more about giving outsiders the perception that world class research is being accomplished than actually accomplishing it with people who can work together and train the next generation of scientists.

David Kent
Dr. David Kent is a principal investigator at the York Biomedical Research Institute at the University of York, York, UK. He trained at Western University and the University of British Columbia before spending 10 years at the University of Cambridge, UK where he ran his research group until 2019. His laboratory's research focuses on the fundamental biology of blood stem cells and how changes in their regulation lead to cancers. David has a long history of public engagement and outreach including the creation of The Black Hole in 2009.
Post a comment
University Affairs moderates all comments according to the following guidelines. If approved, comments generally appear within one business day. We may republish particularly insightful remarks in our print edition or elsewhere.

Your email address will not be published. Required fields are marked *

  1. Marianne Stanford / December 9, 2010 at 14:20

    Interesting post, Dave. I guess I’m one of those female postdocs who despite the option to go to academia, instead chose industry. I was in a typical ‘academic track’ but the more and more I interviewed, the more disillusioned I became about this being the true preferred path for me. I also struggled with the idea that an academic position was ‘success’ and everything else was second best. I also experienced the academic interview process was amazed that several times (but certainly not always) I was the only female candidate being interviewed. Maybe as a reasonably independent and ‘driven’ person I bucked my typical ‘gender’ sterotype, but I often did that in interviews with the risk of coming off poorly (i.e full of myself), and likely suffered for it. I have found that in industry interviews, my ambition was seen in a much different light, and much more positively, but being a team player was also a big part of what they were looking for. Maybe in industry, the ability to work as a ‘team player’ is also more valued that being the ‘shining star’. In any case, I feel that at that cusp of making the transition between the postdoc and your chosen ‘career’ a gender bias does exist, but I’m not sure its overt or indentured in the system. I guess time will tell.

  2. SB / December 10, 2010 at 00:40

    Nice post. I too was shell-shocked when I first read that listing the ‘wrong kind’ of positive qualities in a reference letter can actually work to someone’s disadvantage. But it didn’t take long to figure out why.
    I suspect though, and I could be wrong, that the relative (negative) impact of someone’s listed “communal/emotive” traits varies from department to department and from field to field. I have no data to back this up with, but I would hypothesize that those traits could even be seen as an advantage (gasp!) in fields like medicine, where being able to work well with people is presumably a vital part of your job.
    This reminds me of a conversation I once had with a Famous Professor who advised me against going into research because to him, I seemed too “chipper and nice” for a career in the academe. I think he was politely trying to say that I didn’t come across as very intelligent. But I think (hope) that newer generations of scientists are realizing that capacity for independent thought and the ability to work well with others aren’t mutually exclusive.

  3. […] and “confident.” This immediately brought me back to an article I wrote years ago on gender bias in reference letters, where these character traits were associated more often with men than women in letters from […]