The Biology Equality, Diversity and Inclusion committee has compiled a set of guidelines for organizing a diverse and gender-balanced conference or workshop.
Why organize a diverse and gender-balanced conference/workshop?
If early career researchers are to be assured that scientists can succeed regardless of gender, ethnicity, race, disability, sexual orientation and other minority identities, it is important that they encounter successful representative scientists in their everyday careers. Speaking at a conference or workshop is one of the most visible aspects of an academic’s job, and the organizers of these events have a valuable opportunity to promote equality, diversity and inclusion through selection of a balanced program of speakers, presenters and chairs. Careers in science are attractive to all kinds of people, and an important part of retaining as well as encouraging diversity is by visible inspirational role models. For example, an easy way to send out the message that women can be successful scientists is to give women scientists a platform to present their research: the same holds for people from minority ethnic groups and researchers with other protected characteristics. In addition, speaking invitations, particularly at high-profile meetings, contribute importantly to a researcher’s profile. By extending more invitations to people from under-represented sections of the academic community, as well as to early career researchers, conference and workshop organizers boost the visibility and track records of the speakers.
Ten Ways to Promote Equality and Diversity in Conference and Workshop Organization
Here we provide guidance to conference and workshop organizers, in the form of simple and practical steps that they can take that will help them to produce a fairer and more balanced program. Note, our focus here is solely on measures designed to achieve a balanced program of speakers with respect to gender, ethnic background, and other protected characteristics, and we do not provide guidance on other aspects of conference/workshop organization (choosing venues, setting fees, publicity, etc, although keeping EDI in mind when making these decision is key).
Conference and workshop planning often begins with the selection of an organizing committee. This committee generally comprises a small number (1-6) of persons who will make key decisions about the organizational structure of the meeting. Many scientific meetings fail to achieve a diverse and gender-balanced program not because of any overt prejudice amongst the organizers but rather because of “blindness”, where the organizers simply fail to pay attention to equality, diversity and inclusion issues. A good first step to organizing a balanced program, then, is to establish a diverse, informed and balanced program committee. A minimal target is that there should be at least one person, and ideally at least two persons, of different gender amongst the organizers (as it is often more difficult for a lone voice to speak out), and with similar representation from ethnic minorities.
A balanced program is much more likely to be achieved if this is made an explicit target. The organizing committee is encouraged to develop a speaker policy regarding the invited speakers and the acceptance of abstracts. That policy may specify, for example, that the organizing committee sets out to achieve equal numbers of men and women or, alternatively, no more than 60% of speakers of any one gender. The policy should also seek to achieve balance and representation with respect to ethnicity, geographical distribution, level of seniority, scientific approach, and other characteristics. While some fields have a strong gender and ethnicity bias in their audience, the aim should be (if parity is not feasible) to reduce the field-wide bias in the balance of speakers. A further consideration might be a stipulation that each individual can only present one talk, in order to prevent monopolization of slots by senior white men.
In turn, the policy is much more likely to be realized if it is made public. The organizing committee are encouraged to announce the speaker policy prominently, for instance, posting it for everyone to see on the conference or workshop website or Facebook page, in the conference booklet, or on the conference poster. Here, the organizers should be prepared for, but not be intimidated by, resistance. A rather hackneyed response is the claim that selecting “the best speakers” is more important than achieving diversity or gender-balance. However, a burgeoning literature now attests to the fact that achieving diversity is not inconsistent with a high-quality program. Fortunately, overt resistance to diversity and gender-balance in science is becoming rare these days, and in our experience, most participants at scientific meetings very much welcome an explicit commitment to equality and diversity in the conference organization. In addition, individuals with protected characteristics (e.g. LGBTQ+, disability, those with caring responsibilities) will be more likely to attend/present at conferences that demonstrate that they are a welcoming space for them, in turn enhancing the ability of such individuals to succeed in academia. It is particularly important that the policy is made visible at the time of the meeting. All session chairs should be familiar with the policy, and given guidance to promote equality in questioning, thereby preventing senior or dominant figures from monopolizing the question sessions. A ‘code of conduct’ statement can also be added to the conference literature, stating explicitly that inappropriate behaviour will not be tolerated and providing a procedure for the reporting of incidents. Providing a strong chair/moderator who is prepared to shut down aggressive questioners, may be reassuring for some (e.g. early career) speakers.
Each conference or workshop has several distinct prominent roles, including plenary or keynote speakers, long-talk and short-talk speakers, poster presenters, session chairs, the introducers of speakers, selectors and givers of prizes, and the givers of opening/closing addresses. Typically, at least some of these roles will be filled through invitation by the conference/workshop organizers. Here it is often helpful to brainstorm with fellow organizers, to invite suggestions, and to engage with (especially female or minority ethnic) colleagues asking them for nominations, in order to compile a database of potential invitees. It is worth writing down all the roles that require filling, and compiling a list of possible candidates for each, as it is easy to forget one. For instance, it is important to consider the diversity of session chairs, and not just speakers. With a little thought and discussion amongst a committee that is itself diverse, it is usually quite easy to come up with a broad range of candidates for each role. Suggestions include writing a list that alternates gender rather than being top-heavy with men; searching online directories of scientists in underrepresented categories (for example, https://diversifyeeb.com) and erring on the side of early-career speakers rather than filling all slots with established scholars. Demographic and cultural shifts in academia mean that early career individuals are typically more diverse and representative of the general population, and are both motivated to do, and capable of doing, a fantastic presentation. They will typically benefit more from the opportunity and exposure than senior scientists would.
Throughout the process of sending invitations, and receiving acceptances, as far as possible keep a running total of the diversity (e.g. gender balance, ethnicity) for each role. Be aware that many academics will themselves be put off by all-male panels, and may delay or refuse to participate if a speaker list is all men, or only white people; see point 10 below). Approach key speakers accordingly. Some roles (e.g. conference speakers) may be filled through the submission of abstracts. Make sure that the abstract submission process collects data on gender, and any other relevant identity characteristics (e.g. ethnicity, disability), so that you can use this information in the selection of the program and in providing a safe and welcoming event for all attendees. If participants are asked for this information, and particularly if they are given the opportunity to state any specific needs, they are more likely to have the confidence to attend, knowing their characteristics have been considered.
If the abstracts are to be graded for quality it is good policy first to anonymize the abstracts, removing identifying information, so that they can be graded on the basis of the title and abstract alone. Ideally, each abstract will be rated by multiple independent reviewers, and it is important to establish a diverse pool of reviewers. The mean grades can be used to rank the abstracts for their suitability as potential talks. The organizers can then put together a program that balances quality as reflected in the rankings with ethnicity, gender, career stage, geographical location and other relevant characteristics. It is important to remember that the grading process is imperfect and should be used only as a guideline rather than a rigorous assessment of quality. At least one previous conference organized by a member of the School reported that higher mean grades had been given to abstracts submitted by men compared to women. This may reflect gender differences in writing style, or the tendency to make bold claims. In spite of this, there was no evidence for a gender difference in the quality of the talks. The organizers are likely to face some difficult decisions in how they balance quality and protected characteristics, but should bear in mind that it is often difficult to estimate the quality of a talk on the basis of a short abstract submitted months before the meeting. For this reason, speaker policy targets can generally be achieved, without sacrificing conference quality, if the organizers select the top-ranked submissions from the minority group, irrespective of the absolute grades allotted. Finally, the organizers are encouraged to be explicit about their selection procedures, posting this information on the conference/workshop website.
Usually, some offers of speaking opportunities will be declined, and invited women may be more likely to decline than men. Likewise, individuals from other minorities may experience high demands on their time as perceived “representatives” of various groups. In order to maintain a balanced program, as far as possible, it is helpful to select replacement speakers on a like-for-like basis (i.e. replacing a woman speaker who declined with the next ranked woman, early-career for early-career, etc). It is useful to hold a pool of potential speakers in reserve, so that vacant slots can be filled rapidly.
Women often have primary caring responsibility for children, the elderly, or the sick and disabled, which can limit their ability to travel and to attend conferences. If possible, state explicitly on the conference website, and in the invitation letter, that the conference is ‘family friendly’ and that ‘participants’ children are welcome, provided appropriately supervised’. If resources are available, you can offer support for partner travel, or pay for childcare, for attendees who would otherwise not be able to accept conference/workshop speaking invitations. You could consider whether it might be possible to offer a crèche or to provide a family room that allowed delegates with children to watch conference presentations via video link, and a space for chestfeeding parents. Where a crèche isn’t possible, you could provide a means for attendee carers to connect, gain and share advice, and cooperate to provide care or other resources (e.g. a closed Facebook group). Many conferences offer free drinks, but you could instead choose to spend that money on childcare. Ideally, you should try to avoid holding conferences on weekends and non-working days, and avoid the requirement for weekend travel. Weekend conferences are more likely to prevent the attendance of those with caring responsibilities, who are in turn more likely to be women. Featuring images of women and ethnic minorities on the conference website, poster and booklet, and ensuring that they are represented in any conference photograph, will also help to create a family friendly and diverse atmosphere.
Accountability is an effective policer. For each protected characteristic, briefly summarize the numbers of speakers and participants in each role on the conference website or booklet, ideally on the same page as you outline your policy and procedures.
When you are invited to help organize, attend, or speak at a conference, ask to see the list of invited speakers, and if there isn’t a satisfactory gender balance, or satisfactory representation of ethnic minorities, make a stand. Likewise, if other areas of best practice for EDI are being contravened (such as conferences organized on non-work days). Either decline (although this may be counterproductive some invitees) or make your acceptance conditional on the diversity problem being addressed. Either way, explain your position, and give the organizers specific suggestions of ways to resolve the issue (e.g. listing suitable speakers that they could invite). White-senior-male-biased conference programs often result from institutional and unconscious biases, and/or outdated “benevolent” positions such as “race-blindness”, rather than any overt prejudice. We have found in the past that approaching the organizers respectfully to detail your concerns (rather than criticizing them on social media as a first resort) will usually produce constructive and positive responses.
Download “In a Nutshell”, a 1-page summary of this guidance.