Gender balance in time-keeping at life science conferences
Preprint posted on February 24, 2018 https://www.biorxiv.org/content/early/2018/02/24/267120
The bigger picture:
Giving talks on conferences is absolutely necessary for networking, obtaining jobs and creating visibility. Presenting research is one (important) way of communicating along with publishing and (to a lesser degree) teaching, grant applications and scientific outreach. Even if there’s a positive trend for inviting female speakers, recent studies still show a gender bias, especially at “post PhD career level”. This recent preprint not only focuses on the importance of presenting research at a conference, and thus the opportunity to show results to a broader community, but also emphases the effort to fit and match a talk into a restricted format and allocated time frame. The authors state that this may influence the career path of young scientists, and how this ability differs between genders which could be one additional factor explaining the demographic shift in male/female ratios at the postdoc and PI level.
Further, conference talk does not necessarily equal conference talk. Allocated talk slots often differ in timing in addition to distinguishing general oral presentations, which are usually selected in a competitive review process from abstracts, from invited plenary talks. The authors assess different steps and levels of this process with respect to gender ratio to identify “key players” that could introduce bias.
This recent preprint by Edlund and her all female co-authors compared time slots of male and female speakers at 12 life science conferences in order to discuss if there is a gender bias in the number of talks given by female and male participants and how often men and women go over their allocated time. The number of female speakers correlated with the number of female participants (33% female participants and 38% female talks). When we discussed these numbers, we were wondering if bias against female participants was already introduced at the registration process? For instance, conference attendance can be difficult for women with small children by facing difficult situations for nursing or child care options, a problem that is recently highlighted in an opinion article by Calisi et al. (PNAS). We think the aim to distribute talks 50/50% between female and male speakers should be independent of the ratio of female participants attending the meeting. Edlund et al. found no gender difference with respect to allocated talk time. However, when comparing how often speakers exceed their allocated time, it became clear that male speakers went over time more often (47% male vs. 41% female). The time-keeping-gap between female and male speakers increased when more factors, such as allocated time, career stage, level of time-keeping enforcement and size of the conference were further taken into account: At large conferences (> 150 participants) 73% of male speakers exceeded their allocated time vs. 49% of female speakers. Clearly a lot of time keeping performance also depends on the time-keeping enforcement of the session chair. Focusing on assessing the effect of this covariate on time keeping performance, the authors remarkably report that male speakers go overtime more often when the chair of the session is male, too. Female speakers exceed their allocated time more often in mixed gender chairs.
Interestingly, gender balance was most equal between male and female PhD students (46.2% of PhD student talks were given by female participants) compared to only 35.7% female postdoc or PI speakers. Female PhD students are also best at staying within their allocated time (74% on time). Notably, only 17.8% talks were given by postdocs, 35% of which were female speakers. The low number of postdoc talks surprised us and it would be desirable if a higher percentage of conference presentations were given by (female and male) postdocs to be able to present their research particularly during this crucial career stage on their way to scientific independence.
Conclusion and future perspective
While the authors admit that it’s difficult to draw statistically significant conclusions from only 12 conferences, this preprint is definitely useful to increase awareness. Not only does the study highlight the remaining gender gaps with respect to the number of male vs. female speakers, but more specifically does it show that there is a huge difference in keeping their allocated time both across genders and career stages. By explicitly analyzing this aspect and highlighting the high mismatch between taken and allocated time, we hope that the results will raise consciousness on the importance of time keeping as one challenge that is part of being a professional scientist. It would be great if this study will result in increased motivation to follow rules about time keeping and preparing and timing lectures in advance.
While there are ~50% female PhD students in Life Science, this number decreases drastically on the postdoc and PI level. Thus, a better representation of female speakers at conferences is important to increase visibility for role models. It would be interesting to see how exceeding the allocated time is perceived among the audience, especially between male and female attendants. Other factors that would be noteworthy to study are the gender balance for key note lectures or differences between younger and more senior PIs. The study also suggests that the composition and style of time-keeping enforcement of session chairs plays an important role on actual performance.
Increasing inclusion on life science conferences does not only include female attendance, also researchers of color or scientists with disabilities (just to mention two examples), should be given equal chances to present their research which has to be taken into account already when organizing conferences and the necessary (for instance financial) support. Interestingly, the authors tried to recruit people to take measurements at conferences but the interest was rather little. This study, however, is a good starting point to increase awareness of a sensitive topic and hopefully organizers and chairs will pay even more attention to gender bias. Consequently, a successful conference presentation should not exclusively be measured and evaluated by its content but also by matching the provided framework/timeslot/program.
The authors give some great examples of initiatives fighting gender bias at conferences, such as BiasWatchNeuro (https://biaswatchneuro.com), Academia Net (http://www.academia-net.org) and Anne’s List (https://anneslist.net). We would like to add Request a Women Scientist (https://500womenscientists.org/request-a-scientist) and eseb (http://eseb.org/prizes-funding/equal-opportunities-initiative) as additional resources. In addition, we would like to encourage the readers of this preprint to add more sources in the comment field or on twitter and we’ll update this collection.
Amon MJ: Looking through the Glass Ceiling: A Qualitative Study of STEM Women’s Career Narratives. Front Psychol. 2017.
Calisi RM et al.: Opinion: How to tackle the childcare–conference conundrum. PNAS 2018.
Posted on: 16th March 2018 , updated on: 17th March 2018Read preprint
Also in the scientific communication and education category:
Tracking the popularity and outcomes of all bioRxiv preprints
|Selected by||Gautam Dey, Zhang-He Goh, Lars Hubatsch, Maiko Kitaoka, Robert Mahen, Connor Rosen, Samantha Seah|
Raincloud plots: a multi-platform tool for robust data visualization
|Selected by||Samantha Seah|
Moving beyond P values: Everyday data analysis with estimation plots
|Selected by||Gautam Dey|