Athena SWAN personal FAQ

Yesterday I finished writing the draft Athena SWAN submission for the Royal Holloway Mathematics and Information Security Group and sent it to members of the two departments for comments.

In this post I’ll collect the gist of those replies I’ve made to comments that might be of general interest (and can be made public), and a few extra ‘non-FAQS’, that I hope will still be of interest. The post ends with the references from the bid with clickable hyperlinks.

Why is the bid so focused on women and gender? What about disability or LGBT+ people?

The primary purpose of Athena SWAN is to address gender inequalities. In Mathematics and Information Security this means tackling the underrepresentation of women in almost everything we do.

Royal Holloway was one of the first HEIs to get an Race Equality Charter Mark and some of the proposed actions are aimed at BAME (Black, Asian and Minority Ethnic Students). Another action will promote the College workshop ‘How to be an LGBT ally’ and the (excellent) external Safezone training. Yes, we should do more, but this is a Bronze bid so only the start of a long process to address inequalities.

Are women really under-represented?

I think yes. Mathematics is ahead of many sector norms: for example 40.6% of our undergraduate intake is female, compared to the sector mean of 35.7%; 38.8% of A-level Mathematics students are women. Of our staff 25.0% are women, compared to a sector mean of 20.4%; of professors, 23.1% are women, compared to an appalling sector mean of 12.6%. But all that said, women are half the population, and about 40% of new mathematics graduates are women, so we have very far to go.

Isn’t it discrimination to focus so many actions on women?

I argue very firmly no. The question is a ‘non-FAQ’ that I’ve deliberately worded in a pejorative way. (It is impossible to use the word ‘discrimination’ in this context in the positive sense ‘X has a fine discriminating palate for wine’.) By improving our policies and procedures and thinking particularly about women, we very often make life better for everyone. It is not a zero-sum game. It is legitimate to target actions at women to address under-representation. This does not imply that critical decisions, such as recruitment and promotion, will then be biased in favour of women.

Why the focus on unconscious bias?

Unconscious bias training was the most frequently requested form of training when we surveyed all staff and Ph.D. students. There is strong evidence that unconscious bias exists and prevents women from achieving their potential. An important early study is Science faculty’s subtle gender biases favor male students by Moss-Racusin et al, which asked scientists to evaluate two CVs for a job as a lab manager. The CVs were identical except in one the candidate’s first name was ‘John’, and in the other ‘Jennifer’. Both men and women rated Jennifer as less competent than John, and recommended a lower starting salary.

There is evidence that unconscious bias training can be effective for reducing implicit bias (see pages 6 and 16: the overall picture is mixed, but the conclusion is clear). My own experience suggests that high-quality training and reading around the issue has made me more aware of the issues, and at least slightly less likely to rush to (probably poor) conclusions.

What can I read about unconscious bias?

I highly recommend the third part of Cordelia Fine’s book Delusions of gender. The first two parts make a very convincing case that many stereotypical gender traits are not hard-wired, but instead products of culture and upbringing, or even (on closer inspection) non-existent. The final part examines how our remorselessly gendered society creates these biases and misconceptions.

What is unconscious bias?

First of all, I prefer the term ‘implicit bias’, since one can wrongly interpret ‘unconscious bias’ as referring to something that is independent of our thought processes and beyond our control.

Let me introduce my personal answer with an object that should be emotionally neutral and familiar to readers, namely ‘a vector space’. What comes into your head? Is it a finite dimensional vector space over , such as the familiar Euclidean space , or (my answer) an indeterminately large space over a finite field: ? Or perhaps the most important vector spaces you meet are function spaces, in which case you might be imagining Hilbert space, with or without a preferred orthonormal basis. Yet other answers are possible: someone working in cryptography might think of . Quite possibly, I’ve missed your preferred example completely. Or maybe your brain just doesn’t work this way, and all you think about is the abstract definition of vector spaces and the immediate associations are to the main theorems you use when working with them. Anyway, my point is that we don’t think about vector spaces ‘in isolation’: instead they come with a bundle of implicit associations that are deeply shaped by our education and day-to-day experience.

Now instead think about ‘Mathematics professor’. Without claiming the thought processes are completely analogous, I hope you will agree that something similar goes on, with a bunch of implicit associations coming into our heads. For instance, I immediately start thinking about some of my professorial colleagues in Mathematics and ISG. In this respect I’m lucky: because I have personal examples to draw on, my immediate mental image is not the stereotypical old white man.

Taking this as a roughly accurate impression of human cognition, we now see a mechanism in which bias can enter our decisions. For instance, in the Stanford lab manager study, the implicit associations around the word ‘manager’ bring to mind men, and so the male candidate is favoured. I suspect either you will readily accept this point, or feel it is completely unwarranted, so I won’t argue it any further, but instead refer you to the literature.

What about the Implicit Association Test?

My reading suggests that the Implicit Association Test is valuable as a way of raising awarenss about the potential for implicit bias. But is has been much criticised, and it is not clear the biases it identifies translate into unfair discrimination.

Isn’t this a huge piece of bureaucracy?

A ‘non-FAQ’, although the question has occurred to others. The answer is ‘Yes’. A recent report has many criticism of the Athena SWAN process. For instance, from the summary on page 3:

The application process must be streamlined and the administrative burden on staff, particularly female staff, reduced.

For what it’s worth, I think I could have written a major grant application or completed a substantial research project in the time it took just to draft the submission. Even this rough measure takes no account of the hours of time (not just mine) spent consulting over draft actions and the many weeks of work that the College’s Equality and Diversity Coordinator put into the bid.

What’s the point of doing all this when it clearly wouldn’t address Y (where Y is the manifest injustice of your choice)?

Just because it (probably) wouldn’t have prevented Y, doesn’t mean it isn’t worth doing for other reasons.

Do all Athena SWAN applications have references to the research literature on gender equality and feminism?

(A blatant ‘non-FAQ’.) No. In fact ours is the first I’ve seen. Probably it’s also the first Athena SWAN bid in which the Action Plan is generated by a customised database written from scratch in a functional programming language and outputting to LaTeX.

References

1. Pragya Agarwal, Unravelling unconscious bias, Bloomsbury, 2020.
2. Doyin Atewologun, Tinu Cornish and Fatima Tresh, Unconscious bias training: An assessment of the evidence for effectiveness, Equality and Human Rights Commission research report 113, March 2018.
3. Anne Boring, Kellie Ottoboni and Philip B. Start, Student evaluations of teaching (mostly) do not measure teaching effectiveness, ScienceOpen Research (2016)
4. Caroline Criado-Perez, Invisible Women: Exposing Data Bias in a World Designed for Men, Chatto \& Windus, 2019.
5. Cordelia Fine, Delusions of Gender, Icon Books Ltd, 2005.
6. Cordelia Fine, Testosterone Rex Icon Books Ltd, 2017.
7. Uta Frith, Understanding unconscious bias. Royal Society.
8. Cassandra M. Guarino and Victor M. H. Borden, Faculty service loads and gender: are women taking care of the academic family?, Research in Higher Education (2017) 58 672–694.
9. Nancy Hopkins, Diversification of a university faculty: Observations on hiring women faculty in the Schools of Science and Engineering at MIT, MIT Faculty Newsletter 2006 XVIII.

10. Corinne A. Moss-Racusin, John F. Dovidio, Victoria L. Brescoll, Mark J. Grahama,, and Jo Handelsmana, Science faculty’s subtle gender biases favor male students, PNAS (2012) 109 16474–16479.
11. Ruth Pearce, Certifying equality? Critical reflections on Athena SWAN and equality accreditation, report for Centre for Women and Gender, University of Warwick, July 2017.
12. Athena SWAN Charter Review Independent Steering Group for Advance HE, The Future of Athena SWAN, March 2020.
13. Research Excellence Framework, Guidance on submissions 2021, January 2019,
14. Safezone training.