Skip navigation

New CRC guidelines aim to reduce unconscious hiring bias against women

Canada Research Chairs program is responding to research that shows referees write weaker letters of recommendation for female candidates than for male candidates.



Advocates for equitable hiring practices in Canadian higher education are commending the prestigious Canada Research Chairs program for introducing new guidelines that aim to curb bias against women in the program’s nomination process.

Recently, the CRC added guidelines on how to limit unconscious bias in its package on reference-letter writing. Michèle Boutin, executive director of the CRC secretariat, said the new section was inspired by a pair of studies (one by the American Association of University Women in 2010, the other by Frances Trix and Carolyn Psenka of Wayne State University in 2003) that show letters of recommendation for female candidates tend to be shorter or incomplete; include gendered terms such as “lady,” “mother,” or “wife”; use fewer “standout” adjectives like “excellent” or “outstanding”; focus on personal qualities over skills and achievements; and offer more “doubt raisers” such as faint praise or negative language. These biases tend to appear in letters of recommendation regardless of the referee’s gender.

“Because of the importance of reference letters in the CRC nomination process [nominees require three letters of support], it was important to bring this to the attention of letter writers,” Ms. Boutin said.

Among the guidelines by the CRC, referees are encouraged to keep feedback specific to the requirements of the position; to avoid descriptors that play into stereotypes of women as agreeable, maternal, sympathetic; to use formal titles and surnames instead of given names, and to consider “whether your letter unintentionally includes gaps, or doubt-raising, negative or unexplained statements (e.g., ‘might make an excellent leader’ versus ‘is an established leader’).”

According to Janice Dodd, a professor of physiology and pathophysiology and of gender studies at the University of Manitoba, a common barrier to women’s professional advancement in academia is a negative perception of their expertise and competence. That’s why these new guidelines are so important, said Dr. Dodd. “Your letters have to be able to say, ‘This person has enormous potential.’ And if you’re not using words like ‘excellent’ or ‘outstanding,’ how are people going to see that potential?”

Though she has generally noticed the addition of guidelines for how referees and review panels should consider extended interruptions in a candidate’s career for childbirth, illness, caretaking or other reasons, this is the first time Dr. Dodd has seen guidelines on unconscious bias explicitly stated from a granting, research or postsecondary institution.

For her part, Lynne-Marie Postovit said the guidelines do a good job of addressing a form of discrimination that is widespread and yet “hard to get at” since it is generally unintentional. Ideally, these guidelines would be adopted across the tri-council of federal research granting agencies, said Dr. Postovit, an associate professor in the University of Alberta’s department of oncology. (The Social Sciences and Humanities Research Council and the Natural Sciences and Engineering Research Council jointly stated that a tri-council agency working group is looking into measures to promote equity in their various programs and processes.)

Guidelines addressing unconscious bias are just one step in a larger set of initiatives the CRC has introduced in an effort to improve diversity among chairholders, particularly among groups of people identified as historically underrepresented in government programs: women, Aboriginal people, people with disabilities and people belonging to a visible minority. About four  years ago the CRC began working with individual universities to set targets to improve representation of designated marginalized groups among their chairholders. These targets are reviewed and revised every three years.

Additionally, the CRC introduced random monitoring of the nomination process, which looks at how the position was advertised and how the process was handled internally, among other details. An annual recognition program then spotlights one university’s exemplary practices in recruiting a diverse talent pool. The selected institution is given a platform for discussing what it has done to improve access to its chair positions as well as the barriers they’ve encountered along the way. In previous years the program has singled out the University of British Columbia and University of Victoria and this year recognized Lakehead University.

On the flipside, when “deficiencies” are identified in a nomination process, the CRC reports back to the institutions with its concerns “in the hopes that this will improve in the future,” Ms. Boutin said. She said all these developments have encouraged institutions to work on effective equity hiring practices and have created a “dialogue” around the issue of equitable hiring.

The CRC began committing itself to equity practices in earnest after facing allegations of gender-based discrimination. In 2003, a group of eight women filed a complaint of gender
discrimination with the Canadian Human Rights Commission. The case was settled in 2006, and it required the CRC to apply the federal government’s policies on equitable hiring and anti-discrimination to the award process. The CRC came under fire again in 2008 when its multimillion-dollar Canada Excellence Research Chairs program, intended to draw the highest calibre of international researchers to Canada, failed to name a single woman among its first batch of 19 recipients. (The first and only woman CERC was named in September 2013. Luda Diatchenko, who holds the Canada Excellence Research Chair in Human Pain Genetics at McGill University, is one of 22 CERC holders.)

Following these incidents, the Council of Canadian Academies struck an expert panel (which included Dr. Dodd and Dr. Postovit) to issue a report on the status of women researchers in Canada. Published in 2012, Strengthening Canada’s Research Capacity: The Gender Dimension outlines a number of problems women face in furthering their academic careers and offers recommendations for fixing the “leaky pipeline,” including “the cumulative effects of sexism, stereotyping and recruitment and evaluation biases [that] can lead to the undervaluation of women researchers’ excellence.”

Both Dr. Dodd and Dr. Postovit say they are cautiously optimistic about the various changes to the CRC program, but hope to see more targeted initiatives. “In the last round there were supposed to be Canada Excellence Research Chairs in areas that were more likely to capture women candidates, and despite that initiative, we still only seem to have one woman out of 22 who hold what we call ‘super chairs.’ We have to be very, very aware of continuing to exclude women from these prestigious and important positions,” said Dr. Dodd.

As of October 2014, women occupy 26.8 percent of all filled Canada Research Chairs (449 women out of 1,667 chairs). More women hold Tier 2 chairs for emerging scholars (315 out of 856 positions) than Tier 1 chairs for established scholars (134 of 811 positions).

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. CAM / December 4, 2014 at 14:56

    Excellent initiative. Now if only we can get rid of the 27 red dresses representing women in the infographic! What is with this dumbing down of data? Do we really need to visualize 100 stick figures to get this percentage breakdown?

    • Reuben Kaufman / December 4, 2014 at 21:44

      Interesting comment!

      I think there are two issues here:

      1) Why indeed, in a University publication, is a simplistic graphic needed to convey a percentage? Surely most of us would agree that it isn’t needed!

      2) But given that a graphic was used, does CAM have a particular objection to using dresses and trousers, considering that these are common public symbols to distinguish between men and women? Does CAM consider them sexist in some way? If so, what would be an acceptable alternative?

      • Editor, University Affairs / December 5, 2014 at 09:52

        I can try to explain this from an editorial-design point of view. We created the graphic illustration for the print magazine, where we need to fill a block of space on the page with an image. We didn’t have any appropriate photos to go with this story, so we created this to illustrate one of the points of the article. When we posted the story on the website, should have used this image, or none at all? Since we had it, we used it. Perhaps next time we won’t. Thanks for your comments.

        • Reuben Kaufman / December 5, 2014 at 13:22

          I don’t have a problem with the image …. the fact that it was used but wasn’t “needed”, doesn’t bother me at all (and the editor has explained clearly why it was used).

          But I sensed a sub-text that the red dresses offended CAM as sexist. As I read the comment again now, I may have totally mis-understood that ….. His/her comment clearly emphasizes the dumbing down, not the red dresses!

  2. Murielle Nagy / December 4, 2014 at 18:53

    For Tier 2 Canada Research Chairs, the real issue is the discrimination against researchers who obtained their PhD more than 10 years ago. This restriction stops many women from applying.

  3. AEMcDonald / December 5, 2014 at 13:54

    Great article that calls attention to an issue that many of us have not acknowledged or thought much about. You can find my thoughts at:

  4. Reuben Kaufman / December 6, 2014 at 18:59

    One other point, which I think is far more important to consider:

    The original study on all this (which I haven’t read) may well have shown a correlation (but how strong was the correlation?) between the types of adjectives used in letters of reference and the gender of the applicant. Although, according to the UA article, the percentage of female CRC Chairs is only 27% of the total, is there any way of knowing the extent to which the “unconscious bias” in reference letters influences hiring committees in their decisions?

    A casual looking over past letters of reference that I’ve written for students (although I haven’t been asked to support someone for a CRC Chair) suggests that I’ve used some of the frowned upon adjectives for both male and female applicants. Furthermore, when I’ve been on committees evaluating students for graduate admission, scholarships and the like, I can’t imagine that my decision one way or the other would have been negatively influenced by the referee using first names rather than Mr/Ms/Dr, or negatively influenced by reading cheerful things concerning the personality of the applicant. And frankly, when I’ve encountered superlative praise in a reference letter, I’m just as likely to regard it with some suspicion until I can see some objective evidence to support it in the application.

    In brief: what evidence is there that our choices in language contribute to whatever discrimination women may face in securing prestigious posts?

  5. Jo VanEvery / December 9, 2014 at 18:26

    This is good news but it doesn’t address the underlying problem, which is that reference letters (really letters of recommendation) are relied upon very early in the process. As I understand it good HR practice for equal opportunities hiring suggests that reference letters should ONLY be requested at the end of the process for the candidates you choose to appoint and at that point they should be used only to confirm what the hiring committee has already learned from application, interview and other direct evidence.

    This is a problem not only for CRCs but for academic hiring as a whole, where letters of recommendation are often submitted early in the process and where the name of the person referring is often taking into account as some kind of evidence of the quality of the candidate.

Click to fill out a quick survey