As someone who studies academic organizations and their efforts to encourage equality, I believe that dedicated programmes to address inequalities are essential. I also worry they might not be designed to provide support to the very people who need it. Issues such as race, class and overlapping patterns of discrimination must be considered. So, too, must the way that the measurements used for assessment tend to distort what is being measured.
The rationale for programmes to promote women’s representation in science is clear. In the European Union, only 15% of senior academic positions in science are held by women. Numbers are improving (slowly), and gender-equality initiatives deserve some of the credit.
One of the most prominent is the Athena SWAN charter, which was established in the United Kingdom in 2005 and has inspired programmes in other countries. It grants awards to research institutions that perform self-assessments on gender representation and career advancement, and establish well-documented multi-year plans to improve. As of April 2019, more than 160 institutions had collected more than 800 bronze, silver or gold Athena SWAN awards. In 2011, the UK National Institute for Health Research announced that medical schools needed a silver or gold award to be eligible for its funding. Since then, the proportion of applications from relevant departments has increased fourfold.
Initiatives such as Athena SWAN insist on hard data, which are crucial for credibility and accountability. But, too often, these data focus on women as a homogenous group and so overlook intersecting patterns of disadvantage faced by women of colour, early-career researchers and sexual minorities. For example, compared with white women, female academics of colour report limited access to mentoring and higher rates of feeling isolated, excluded, discounted and not belonging.
Another unintended consequence is that women can be penalized by the programmes designed to help them. A 2014 analysis found that women make up more than 70% of Athena SWAN champions, a labour-intensive role that takes time away from their research. My colleague and I found similar patterns in an analysis of 11 institutions with silver SWAN awards: ten teams had more female members than male; eight of the submission teams were led by women. One interviewee described coming in to work on her department’s Athena SWAN application on a Sunday and having e-mail conversations with two women at other institutions who were also working on theirs. Some women had been advised that their promotion would depend on attaining an award, even though the department had not provided essential resources or support. That defies the spirit of the awards. These programmes should be about the ability of the department to support equality.
Perhaps the most unfortunate unintended consequence is that achieving gender equality becomes a box-ticking exercise, divorced from the broader goals for a fairer society. A department looks at gender-equality data not as an opportunity to gain insight and improve the working environment for all, but to present itself in a certain light in order to secure the award; it must assert that inequality is not really that bad within their unit, but that it can make clear improvements. There is a temptation to think more about what can be demonstrated than about what needs to be done.
Gender-equality programmes should be about collaboration and working together to weaken entrenched inequalities in the academic system. How can we make sure that happens? I have some ideas.
First, explicitly consider intersecting patterns of disadvantage. Calculating the percentage of women in various positions is insufficient; it is important to capture experiences of sexual, ethnic and other minorities as well. Fortunately, equality programmes, including Athena SWAN, are broadening reporting requirements around intersectionality and people from gender or sexual minorities, but the focus is still often on simpler, overarching statistics.
Second, the awards should consider more qualitative data about the workplace culture, and make sure that applicants have the resources to support this extra work. Assessments should, ideally, be accompanied by site visits. The award system for Project Juno, a gender-equality and inclusion programme set up by the Institute of Physics in the United Kingdom, offers a visit by evaluators to the research institution and meetings with staff.
Evaluators should look beyond data to find the stories behind career decisions. Women I interviewed told me that they left positions because they felt that their career progression was blocked. None had reported this to her former employer during the exit interview.
Third, create ways for people to report mistakes and disappointments without jeopardizing their award status. Everyone I interviewed had seen unintended consequences of gender actions (everything from unfair burdens, to less willingness to talk about the most difficult problems, to stronger feelings of hypocrisy and cynicism). A dedicated space in the application form to describe these would reduce the incentive to put a misleading positive spin on reports. There are precedents for helping people to air dirty laundry — reports could be anonymized or shared under strict Chatham-House rules (that is, disclosing information, but not identities). Most people who start equality and diversity initiatives truly want to learn from mistakes and to help others achieve the same goal.
Most of all, continue to check that the consequences of programmes match their intent. Gender-equality initiatives are laudable, but their drawbacks and insufficiencies should not be ignored. The only way not to suffer from unintended consequences is to be mindful of them.
Nature 570, 277 (2019)