Rarely a week goes by when I don't notice at least one comment in my twitter feed about the lack of female speakers at a particular conference. I've noticed this in person and it is certainly a general topic of discussion in the blog-o-sphere. So, when a paper on the topic came into my email this week, it piqued my interest. Spoiler: the title kinda gives away the punchline.
The authors set up the issue by looking at the general stats in the EU:
In 2006, 36% of EU PhD graduates in Science and Engineering were women, reducing slightly to 33% among post-doctoral researchers (Grade C), then falling dramatically to 11% of the senior academic ranks (Grade A; European Commission, 2011; ﬁgure II.3.13).
Then they use six years (2001-2011) of the biannual ESEB conference to test two hypotheses:
1. Because the scientiﬁc achievements of women may be less visible than the achievements of men (Thelwall et al., 2006; Fernandez et al., 2009), female scientists may be overlooked more often for invitations to talk.
2. Symposia organized only by men will have fewer female invited speakers than symposia that have at least one female organizer.
In order to test these hypotheses they tested for gender difference between all presenters, controlling for population differences (in the field) and career stage factors. In order to get at a slightly more amorphous topic, they also tested gender ratios of "top quality" researchers in the field and those invited to speak.
To calculate the baseline numbers in the field, the authors examined gender ratios of three TT career stages at the top 10 universities for life sciences (as ranked here). To account for major recent discoveries leading to a conference invite, they searched Nature and Science for the genders of first and last authors.
The take home is that women presented fewer talks, proportional to their representation at the meeting (dashed line). The differences were especially pronounced for invited and plenary talks. In addition, they found that women were significantly underrepresented when it came to symposia organizers, but that the presence of women on a symposium organizing committee made no difference in the gender ratio of invited or regular speakers.
As for invittee based on career stage:
Fig. 3 The percentage of invited speakers that were women, in symposia (black bars) and plenaries (white bars), at ESEB congresses in 2001–2011, in comparison with the percentage of women in baseline populations of ﬁrst and last authors in top-tier journals (dark grey bars), and faculty members (light grey bars; Fell. = Fellows, Lect. = Lecturers, Prof. = Professors). Horizontal lines under the x-axis indicate the speciﬁc category groupings that the bars belong to. The horizontal continuous line in the plot indicates the sex ratio among the realized invited speakers at ESEB 2011, and the dashed line indicates the sex ratio among all initially invited speakers at ESEB 2011, including those who declined to participate.
And by "impact":
Fig. 4 The percentage of invited speakers that were women, selected by randomizations from baseline populations of authors in top-tier journals (ﬁrst and last authors) and faculty members (error bars = 95% conﬁdence intervals). Horizontal lines under the x-axis indicate the speciﬁc category groupings that the data points belong to. The horizontal continuous line in the plot indicates the sex ratio among the realized invited speakers at ESEB 2011, the dash-dotted line indicates the sex ratio among symposium organizers at ESEB 2011 and the dashed line indicates the sex ratio among all initially invited speakers at ESEB 2011, including those who declined to participate.
Based on these data, the authors propose three (non-mutually exclusive) reasons for the male bias in speakers at ESEB:
(1) the pool of scientists that could be invited to speak contains fewer women than men, for example due to the ‘leaky pipeline’
(2) women turned down invitations more often than men
(3) there was a bias for selecting men as invited speakers.
Reason (1) is quickly ruled out, based on the data. Interestingly, the process of inviting speakers appears relatively unbiased for the 2011 meeting, as the invitation rate for women was roughly equal to the population levels. However, in discussing (2), the authors report that 50% of women declined the invitation, compared with 26% of men. The authors note that child care is not yet available at ESEB and suggest that further research into the reasons for (and solutions to) women declining speaker invitations is needed. Encouragingly on front (3), at least for the most recent ESEB meeting, there was no bias for invite rate or related to invites from syposium committees
with female members, suggesting progress on the overall gender bias of invitation.
Alright, so all well and good. It's the declines that are resulting in apparently biased meetings! But is that a good excuse? As @scicurious also pointed out when she linked to this, is it even a real excuse?
I think we need to do better than balancing the invite rate if the data are telling us that declines by women are higher. If that's the case, invite more women. If you can't think of any, ask your colleagues. Ask your fellow symposium organizers. Pick up an issue of your favorite journal. Don't take the easy way out and throw up your hands.
Schroeder J, Dugdale HL, Radersma R, Hinsch M, Buehler DM, Saul J, Porter L, Liker A, De Cauwer I, Johnson PJ, Santure AW, Griffin AS, Bolund E, Ross L, Webb TJ, Feulner PG, Winney I, Szulkin M, Komdeur J, Versteegh MA, Hemelrijk CK, Svensson EI, Edwards H, Karlsson M, West SA, Barrett EL, Richardson DS, van den Brink V, Wimpenny JH, Ellwood SA, Rees M, Matson KD, Charmantier A, Dos Remedios N, Schneider NA, Teplitsky C, Laurance WF, Butlin RK, & Horrocks NP (2013). Fewer invited talks by women in evolutionary biology symposia. Journal of evolutionary biology PMID: 23786459