Since joining Elsevier, I’ve learned a lot about the important work being done by the Gender Summit in communities around the world and the efforts to forge a deeper understanding of what we all know: that inclusion in research leads to better science more likely to include dimensions such as gender in research, and with more diverse viewpoints on how to approach challenges.
We have made a lot of progress over the last 50 years. There are more women in the workforce overall, and there are more women in the STEM workforce everywhere in the world. Progress depends on steady work, and it’s the effort and dedication of people involved in events like the Gender Summits that makes it happen.
However, we are not where we need to be. A 2019 study examining the funding trajectories of the NIH, the leading funder of health research in the US, highlighted some significant issues. Researchers found that female first-time principal investigators received a median grant of $130k across all grant and institution types during a 10-year period, while men received $170k. That puts women at a disadvantage from the start, systematically providing them with less funding, which impacts equipment, lab resources, and hiring. It’s not just a US story; Prof. Belle Derks’ work in the Netherlands shows similar conclusions.
While disconcerting, data like this is crucial. To make progress in gender inclusion, we need to be able to measure where we are and where we want to be.
It was at a Gender Summit that my colleague Dr. Holly Falk-Krzesinski, our VP of Research Intelligence, had her eureka moment. She realized it was time to tackle the conundrum of sex-disaggregated data — and that Scopus, which contains indexed research from 24,600 journals and 5,000 publishers, would offer an ideal testing ground.
After Holly introduced this methodology at the 2015 Gender Summit in Berlin, we were able to improve it with input from many of you. Two years later, we produced Gender in the Global Research Landscape– an analytical report that explored research output and many other critical indicators between men and women in 12 countries across 27 different research areas. We developed a completely new bibliometric methodology around sex-disaggregation in consultation with many of the experts we met at the Gender Summits, equipping decision-makers with far more granular and important insights into the leaky research pipeline.
In March, we will share a new gender report, expanding our benchmarking efforts to new countries, discipline areas and indicators. As we continue to reveal the strengths and gaps in research being undertaken by women scientists, we will help create a more equal academic playing field.
We’ve come to appreciate that the Gender Summit is not really a conference but a platform for action — action that we at Elsevier also need to take by examining our own core processes in publishing and technology.
Here are the actions we are taking at Elsevier across key areas:
1. Gender balance at conferences
At Elsevier, we organize over 50 international scientific conferences a year. We first looked at the gender split in our conferences in 2015 and saw that, while our delegate list reflected the field at 60 percent men and 40 percent women, our gender balance for invited speakers had a strong male bias of 85 percent to 15 percent.
We made a commitment to address this. Working closely with conference chairs, who are generally scientific leaders in their fields, we examined the demographics in their fields and encouraged them to consider a broader pool of speakers.
In four years, we moved from 15 percent to 32 percent invited women speakers. We avoided all-male panels, we achieved parity in eight conferences, and we saw a significant correlation between a favorable gender split and above-average appreciation scores. Reflecting the necessity of this, my colleague Nigel Clear, Commercial Director, explained:
By addressing the gender imbalance, not only do we challenge an outdated status quo and provide opportunities for talented individuals, we also provide our audiences with a more diverse experience where they will be exposed to viewpoints, styles and fresh ideas that they would not have otherwise encountered.
2. Gender balance on editorial boards
Since 2016, we have worked to balance our editorial boards across all our journals. Today, our female-male ratio is 22 percent to 78 percent. Clearly, we have a long way to go, but the will to improve is there.
Our efforts started when Deborah Logan, Publishing Director for our Energy and Earth Science journals, attended an energy conference where about 25 percent of the delegates were women. She attended the journal reception for 80 editors and found herself to be the only woman in the room. As Deborah said:
When the contrast is so stark, as it was at that reception, you start to ask: ‘What am I really seeing here, right in front of my eyes, and what am I going to do about it?’
She developed a program to guide editors and publishers of energy journals to become agents for positive change. At the time, just 6 percent of editors and editorial board members were women across all energy journals – in a field where 21 percent of researchers are women. Since we took action, the number of decision-making women editors on these journals has risen to around 17 percent, and to around 24 percent on the flagship energy journals, with a commitment to 30 percent in the next 2 years and 50 percent by 2025.
Another example is The Lancet, which has established a 50:50 gender ratio for its board with an aim for all The Lancet journals to reach parity by 2020.
John McConnell, Editor of The Lancet Infectious Diseases, was faced with a tough decision last year: He had 5 women and 22 men on his board, all longstanding and loyal editorial team members. He wrote to 11 of them thanking them for their service and citing the need for greater inclusion and gender balance. Ten immediately responded by agreeing to be retired. He received comments such as, “I agree that it is time to pass the baton to a member of the next generation and help diversify the board.” The board now has 12 women and 11 men.
3. Engaging a broader spectrum of peer reviewers
We are also tackling the peer review process. The Cell team recently looked at their reviewer pool and saw that their that their choices skewed toward American male reviewers. They knew they needed actual numbers, but like most scientific journals, they don’t collect this kind of information on gender.
They examined their 2018 reviewer pool, hand-curating the information through a combination of personal knowledge of reviewers, researcher websites and published interviews. They assigned a male or female gender to the 2,259 scientists who reviewed papers for Cell last year: 82 percent of these reviewers were identified as men and just 18 percent as women. Of the men, 67 percent were from the United States.
As a next step, Cell is incorporating gender self-identification as a mandatory part of personal information, and they have committed to engaging a much broader pool of reviewers across gender, career stage and geography. I was in Boston recently with the Cell team, and the energy was palpable; now that they have the data, they can act on it to drive improvement.
4. The new frontier: technology and diversity
Another thing we have learned at the Gender Summit is that there are always new frontiers in diversity, as the session here on AI and open science showed. How do we create an inclusive scientific environment in a digital future? Supervised machine learning uses existing data sets to train the algorithms, so existing human based biases get reflected by machines. Such issues raise new ethical and legal questions.
At Elsevier, our mission includes contributing to the advancement of science and supporting scientists through services that enhance access to knowledge and data, which is being produced at an accelerating rate. We need to ask ourselves questions on the state of the knowledge itself. Questions like:
- How can a data and analytics approach be applied to evaluating research in support of individual researchers, their institutions and funding bodies?
- How can metrics be used to understand the impact of research and return on investment to society?
- How can we support the fairest, most unbiased and reproducible research?
- And how, as an information analytics provider working in science and health, can we facilitate the equitable creation, dissemination and communication of research results and ensure access to the immense body of knowledge already produced?
During the course of this Gender Summit, these and many more questions were carefully examined and will ultimately be fed into a series of key recommendations for the Dutch National Action Plan. I am genuinely excited about the progress we can make together.
This story was adapted from Kumsal Bayazit’s keynote address at the Gender Summit Europe in Amsterdam October 3.
The Gender Summit
Held throughout the world, the Gender Summit serves as a forum for researchers, scientific institutions, policymakers and other stakeholders to discuss gender impact on scientific research and innovation. Founded by Dr. Elizabeth Pollitzer, Director of Portia, a British nonprofit gender in science organization, the summit aims to “make gender equality in science and research the norm” and promote the inclusion of the gender dimension in STEM research.
Elsevier first became involved in 2011 through the work of the Elsevier Foundation with a grant to Dr Pollitzer, which sought to help early-career women scientists successfully navigate the complex landscape of a research career. Since then, to help support and steer this critical process globally, both Elsevier and the Foundation have partnered with Dr. Pollitzer on the Gender Summits in different capacities. This partnership has provided an excellent environment to absorb best practices on gender in research while showcasing Elsevier’s initiatives such as the gender report and helping us work with the research community worldwide.