Gender Bias in Academia: Latest Study Results

I have touched upon the topic of gender equality in my blog on several occasions, be it about gender diversity in the workplace, the gender pay gap, or gender imbalances in expatriation. As such, the topic is far from being new, yet it cannot be dismissed, as the challenge of gender equality continues to persist. For example, according to the World Economic Forum’s gender gap report, as of 2018 globally there was a remaining gender disparity gap of 32%, with the largest disparities between men and women in the areas of political empowerment (77%) and economic participation (42%). Although the annual gender gap report shows steady progress, there is still a long way to go in reducing existing disparities. Thereof, we need to keep the importance of the matter afloat.

Recently I came across the topic reading intriguing research in the domain most relevant to me personally: academia. Echoing other professional domains, research indicates that in spite of having similar starting positions and qualifications to men, women seem to be less likely to succeed in the academic hierarchy and have higher drop-out rates from academia than their male counterparts.

In order to shed more light on this gender gap, Juan Madera with colleagues (Madera, Hebl, Dial, Martin & Valian, 2019) set out to investigate how gender bias manifests in the early stages of the selection process in the academic world. Highlighting some theoretical underpinnings for the hypothesized gender bias, the researchers brought up the implicit human tendencies to divide jobs and tasks according to gender schemas, or gender stereotypes. For example, nursing and teaching jobs and ‘caring’ tasks are traditionally considered to be female gender-typed, while stereotypical male traits that revolve around assertiveness, leadership, agency and competitiveness ‘better suit’ financial and technological jobs. Alas, such gender schemas, and gender-typing of occupations are reaffirmed by actual gender statistics in these professional domains. In relation to the academic domain, the study authors noted that academics’ responsibilities have also been linked to gender norms and stereotypes, and have been historically viewed as more masculine. Referring to data from the U.S. National Center of Education Statistics, the study authors note that 86% of full professors at American institutions are men. Similarly, data from the UK shows that only 26% females are holding professorial positions, and only 36% of females have other senior academic contracts.

Focusing specifically on the early stages of the selection, the study aimed to examine letters of recommendation, and the number of doubt raisers in them. The researchers were interested to know if letters of recommendation for men and women are written differently. Are there any gender-related differences in the amount and type of doubt raisers in the letters? If and how do the doubt raiser influence evaluations of job applicants?

The results showed that recommendation letters for women do indeed include significantly more doubt raisers than for men. This result remained significant even as researchers controlled for a set of background variables, such as number of publications, highest journal impact factor, years in post doc position, number of courses taught etc. In regard to the type of doubt raiser, letters for women contained more negativity (e.g. ‘she/he doesn’t have much teaching experience’), hedging (e.g. ‘might not be the best …’), and faint praises (e.g. ‘she/he no longer needs major supervision’) than the letters for men. Interestingly, the amount of doubt raisers was not influenced by letter-writer gender, meaning that both male and female writers of recommendation letters tended to use significantly more doubt raisers when recommending a woman, compared to a man. As such, it seems that both genders are similarly susceptible to using gender-typing of work, and hence perceive careers in academia as rather masculine endeavours in this study example.

Looking at the influence of doubt raisers on the evaluation of applicants, the researchers found that applicants, whose letters contained negativity and hedging doubt raisers, were evaluated significantly lower than applicants without such doubt raisers, or with another type of doubt raisers (such as irrelevant information or faint praises). Importantly, negativity and hedging had a detrimental effect on evaluations irrespective of applicant’s gender, hence, they hurt both, male and female applicants. In other words, the negative effect of the doubt raisers was not due to a gender bias of applicants’ evaluators, but rather due to the gender bias of the writers of the recommendation letters. As the researchers concluded, ‘Doubt raisers are a minus for everyone, but letter writers assign that minus more often to women than to men’ (pg. 298).

This research holds important implications for academic institutions, and organizations at large, as it is fair to expect that similar biases are present also in other work domains. Given the significant impact letter writers can potentially have on either promoting or hindering women in academia, gender biases should be better identified and reduced. To start with, maybe letters of recommendation should not have such a weight in the selection process after all? Could there be more emphasis on direct (and, importantly, structured) job interviews and CV reviews? Finally, as the study authors propose, part of the solution could lie in more structured letter templates, which minimize recommender’s freedom to include doubt raisers in the letters.

 

 

Further reading:

Madera, J. M., Hebl, M. R., Dial, H., Martin, R., & Valian, V. (2019). Raising Doubt in Letters of Recommendation for Academia: Gender Differences and Their Impact. Journal of Business and Psychology, 34, 3, 287-303.