 Hello, I'm Gwen Evans, the VP of Global Library Relations at Elsevier, and I'm delighted to be introducing my colleague, Holly Falk Krasinski, who will give the CNI Spring 2021 presentation on gender identity data at scale. Holly? Thank you so much, Gwen. Let me get rid of that. Thank you again. I'm Holly Falk Krasinski, and I'm VP of Research Intelligence. I'm the same team as you, Global Strategic Networks. I'm also the co-chair of the Gender Equity Task Force here at Elsevier. And I'm delighted to be able to share with everybody today some of their efforts around gender and race and ethnicity identity data in editorial management systems. Our IND efforts extend far and wide and fit within Elsevier's broader IND initiatives and roadmap in support of the United Nations Sustainability Development Goals, which include Elsevier and Relics commitments to the UN SDG-5 Gender Equality and the Global Research Council Statement and Principles to promote the equality and status of women in research. A lot of the work that I'm going to be sharing started with the strong foundation of our Elsevier Foundation. And as of today, there are five IND priority areas that Elsevier gender is one and race and ethnicity is a second one. I'm also going to share with you some of the information and really drive a driver of a lot of what we're doing is our Inclusion and Diversity Advisory Board that was launched just a year ago today as a matter of fact with executive sponsorship. And the goal of that board is to ensure that Elsevier is working with research leaders, funding bodies, and higher education institutions to drive gender and race and ethnicity equity across the STEM academic career pathway. So much of this work comes by way of the Gender Equity Task Force that I share along with the director of Elsevier Foundation, Elan Sem. And importantly, this group that was formed about six years ago started as an organic hub, a grassroots initiative, if you will, a colleagues committed to and involved in increasing gender diversity and inclusion in research, building on the foundation of the Elsevier Foundation. With strong representation from our journals team especially, we brought together cross business work streams examining our key processes, principles, and systems in the context of gender IND, thinking about ways that Elsevier could support the most robust research possible in the most equitable and inclusive way possible. We also wanted to establish a framework of best practices and policies that other organizations could emulate and or build from. Importantly, we have a number of work streams within the Task Force. Two of those which I'll share today, some you might be familiar with already. We use our considerable data and analytics resources to be able to address issues of gender inclusion and diversity via a strong evidence base. And one year ago today, we released our most recent global gender report, the researcher journey through a gender lens, which you can find available and online. And then the other thing that came from that data or evidence based approach was to think about the ways in which we could capture gender identity demographic data within the editorial management systems. Now it's not that we weren't already doing this to some extent, but admittedly we weren't doing this the best way possible. So what we started with was an approach to update the schema of gender identity and establishing a robust data field that we were in agreement that could be used across all of Elsevier's data systems and not only limited to the editorial system. We developed a five page document that explains how we got there and there is shown in this orange box here. This is the gender identity data field that we are now using, speaks to the ways in which we are aiming to capture this data and for what purpose. And then there is a clear link to our privacy policy. And then we actually made sure we were specific about the actual question we're asking, which gender do you most identify? And then we moved from using only a binary schema that was unfortunately using terminology that really refers more to biological sex, that is female and male, to using something that is not binary and uses the appropriate terminology for gender. That is woman and man. And then we've selected non-binary or gender diverse. And then to fulfill the alignment with GDPR compliance requirements that we always make the option prefer not to disclose available. Now some of the things that we've been having to contemplate as we're working on being able to implement that gender identity data question into our systems really fall into two buckets. The first is the legal and privacy policies that we have to contend with. And the second is the technology issues. So within the first bucket are things like GDPR and in the USCCPA. So these are policies around the privacy and use of personal data. We also had to acknowledge that there are differences between external stakeholders and Elsevier employees. There are labor and HR laws apply that will not allow us to collect any such data from anyone who is actually employed by Elsevier. And we want to make sure that we have really robust and transparent communication and are actually working to elaborate on that now so that we'd have a dedicated webpage that people could go to. On the technology side, well, what we're aiming to do is to introduce this data field into editorial manager, which is part of the ARIES portfolio. This is our existing or legacy system that we have, but we also have other systems where we've been capturing data and looking to have those internal systems interoperable with editorial manager. We want to have the ability to collect information and are looking into the opportunities to do that with open-ended options. And then of course, once the data is within a system, you have to make sure that it is only accessible at appropriate times by appropriate people. We also have to think about the long-term storage and multiple profiles that individuals may have within our own systems. And all of this is being factored into our platform development roadmap. Now one of the opportunities that we've already had to be able to share this data in practice is a part of our new journal homepage displaying editor gender diversity indicators. And we released this and announced it just a month ago on the International Day of Women and Girls in Science. And this is in practice. We're taking some of the data that we've already been able to collect in an internal system that will ultimately interoperate with editorial manager. And being able to put this information by way of an editor gender diversity indicator on journal homepages. And our goal with doing this is to be both transparent in the data that we have and how we're using it and making sure that that data is also contextualized. So we're currently using comparator data for entire fields or portfolios of journals. And we'll be working an addition comparator is benchmarking data that comes by way of the disciplinary gender diversity data from our own 2020 global gender report. Now as we've been working in this effort to do gender identity data capture, we began thinking about the other opportunities for us to extend into different dimensions of the IND. And as I noted earlier, we have five main priority areas and one of the others is race and ethnicity. So we thought how is it that we can extend our efforts around gender identity data collection and extract and extrapolate those over to race and ethnicity identity data as well. This is not as easy as you might think though. In fact, race and ethnicity data is quite a complex challenge in no small part because most of the time race and ethnicity are thought of with schemes that are quite variable and usually nationally focused. For example, the format that the US Census Bureau might use, which by the way is drastically different than the Census Bureau in Canada and the Census Bureau in the UK. There is this lack of universality around race and ethnicity schema that won't work for a global publisher like Elsevier where we really want to think about researchers across the world and especially because we have to take into consideration the fact that there's such an internationalization of research. There are also legal and policy considerations and technology considerations that have to be considered just like with gender identity data. And we've already discovered that there is greater sensitivity to this type of demographic data compared with gender identity. Also we're discovering that there is likely to be an increased hesitancy in terms of providing this kind of personal demographic data when people are being asked to answer multiple personal data questions. So how are we working through the complexity? Well, not alone and that's non-solus. That's part of Elsevier's logo and something that I feel very strongly about. So we've looked to lots of different internal and external groups and stakeholders to try to inform the way in which we'll go about this process, which by the way is also informing the approach that we're taking to the gender identity schema and implementation of that data field as well. So it's everything from looking to the literature and thinking about the Royal Society of Chemistry has a multi-publisher that's RSC group that we can work with, our External Inclusion and Diversity Advisory Board. Also looking to have an external consultant and to make sure that we can actually do some testing of the schema with a global research community. And ultimately we also have our eye on a prize a little bit further out. And that is going back to the data analytics that we focus on. Is there a way then to take the schema that we'll be developing for race and ethnicity and use it in an inferred methodological approach so that we can conduct bibliometrics based analyses at a large scale and extend what has only been gender analyses to include other intersecting social identities as well. So we're working with the Royal Society of Chemistry Publisher Group and all together we developed what we thought were the basics of a draft universal schema. But when we shared this with an external consultant, what we realized is that it's probably not going to be possible to capture both race and ethnicity inside of a single schema. And so this is currently a two-question draft race and ethnicity schema that we're considering. And when I say draft, I mean it really, really is in draft. There's nothing final about this now. But what we've already seen is an evolution from thinking about a single question schema to one where we might have to have a two-question schema. And this is what we would use and they're sharing then with the external consultant and we would test using a global group of researchers. Importantly though, we want to make sure that we're not losing sight of what our overarching purpose is by looking to collect this data as we get stuck in working through the minutiae of the day-to-day issues having to do with these two initiatives. So we are doing data collection and collecting self-reported gender identity and then ultimately race and ethnicity identity data within systems we use to support our editorial workflows. And we're doing that to gain actionable insight that will inform our decision-making processes and help us to develop potential interventions. All of this is in the context of our goal towards greater diversity and inclusion of our editorial boards, reviewer, and author pools across gender and race and ethnicity dimensions. And overall, this is to help us with our overarching mission that research and a research workforce that is more diverse and inclusive and equitable is what we're aiming for. And importantly, in order to do that, we recognize that we have to do that by building trust all throughout this process with the research community that will be working with us. So that ends my presentation and now, Gwen, I'm really happy to answer any questions that you may have about this process. Thank you, Holly. And let's start with question number one. We are recording this on March 8th, International Women's Day. Can you tell me why this issue matters for Elsevier? I think this goes back to just recognizing that Elsevier is being such a large publisher. We are a steward of a considerable amount of research in the STM fields and researchers at the level of our editors, our authors and our reviewers choose to work with Elsevier. And so we have a responsibility to them and to the research community at large. The fact that it is that at the scale at which we operate has the opportunity to have considerable impact. And so in that way, understanding that what we do really matters in the context of the research community. We really feel that it's our obligation to be involved in these activities. Can you talk more about the actionable insights that underlay this effort? What actions are Elsevier and publishers hoping to implement internally? But also, how might others use the insights provided, for example, institutions, governments, authors or librarians? Yeah, that's a good question. And I will be careful in that to speak for other publishers. So I'll just talk about the things that we're thinking about at Elsevier, although these are the information that we'll be sharing with the other publishers in the Royal Society of Chemistry Group. What we're really thinking is that, first of all, we need to have a good understanding of where we're at right now. And before we can talk about the change that we'd like to see and the progress that we're hoping to make, we have to have a good understanding of where we're at. And when we don't have that understanding sort of anchored in data, I think it's too easy to have biases in place. That is that we overestimate the participation of women, for example, and underestimate how underrepresented they are. And so this data will help to bring to light where things are right now, but also tell us where we've already begun to make great strides. And I'm not just talking about representation of women either. We know that in the field such as psychology and nursing, they tend to have a significant underrepresentation of men in research. And we want to ensure that men are not disincluded from those domains of research in the same ways that we want to make sure that women are being included in other areas of STEM research. We'll actively look at that data and help it make sense for us and recommendations to our editors and chief and editorial board members and really talk with them. Here's the data so you can see what the composition of the current editorial board looks like in terms of gender diversity. And let's also take the data from the global gender report and look at what the gender composition is across different disciplines that the journal supports and meets the needs of it in different communities. Can we look to make changes, recognizing that editorial boards are leadership positions? Can we look to do better than just meet the overarching representation of women in a field? Can we look to drive change and catalyze that by changing the representation of the gender representation of editorial boards? So these will be some of the things that we'll be thinking about. We also want to make sure that we're being explicit that these are part of our efforts and share these with our editorial board members and our reviewers and our authors to make them aware that we're thinking about these things. In another set of work streams of the gender equity task force, we see that this has been happening with regards to gender representation at our conferences. So these are conferences that Elsevier supports. Six years ago, an initiative was put in place simply to make the data available to conference organizers and ask them without any mandates and without any minimums to please consider greater gender diversity with regards to the speakers who are being invited to conferences. And in six years, they've been able to go from a little over 15% of women being invited as invited speakers to close to 40% women being invited speakers to the conferences. So we're looking to make explicit this kind of data that helps to overcome implicit biases that we all have. And when I say all, I mean women too. Women also carry biases just as men do. So this is not a problem only that men have. This is a problem that all of us have and implicit biases that we carry. Can you tell me more about the editor gender diversity indicator on the journal homepages? How many journals currently have the indicator and what are the future plans? Oh, yeah, so it's really exciting. We started with just 25 journals as a pilot. Look, Elsevier's journal portfolio is very large. It consists of over 2,000 journals. And just like my time spent as a bench researcher, you never set up a big experiment to start because that usually results in big failures. And so at Elsevier, we tend to try things and to pilot opportunities. And then when they work, we expand those bigger. So we started with 25 journals and their home pages and did a soft launch of that. Last month, when we released and publicly announced, we now have that journal, sorry, that gender diversity indicator on the journal homepages for over 600 journals across our portfolio. And from that, we're using the opportunity now to go out and speak to our society partners, right? So society partners choosing to work with us as their publisher. We don't get to make decisions about their journals. Instead, what we do is think about the innovations and the opportunities that we can bring to them and then discuss how they can start to implement those as well. So I've already had an opportunity over the last couple of weeks myself to talk with the editors in chief of a number of those society titles that are partners with us. And then we'll look to extend this across more and more of our titles. We are limited to some extent in places where we haven't been able to adequately capture good amounts of gender data. And we have a task force, sorry, a work stream of our task force specifically dedicated to thinking about how we can do that better. And then also a task force within our platforms team looking at how we can have good interoperability between different systems to share gender data that's been collected in a GDPR compliant manner so that we don't have to ask investigators for that data or ask them even to complete it into multiple profiles. So that's our goal is to be able to extend it. I would love to say eventually to all of our titles. I can't guarantee that, but that's the goal and that's where we have our site set. Can you talk about the order of magnitude that this initiative represents? How many people are involved and how are you handling issues like multiple legacy data systems? So there's a lot of people, but there are a lot of really dedicated people and just as we think we have one group and can focus on the issues, somebody's like, but did you talk to so-and-so? Like not yet. And then we reach out to so-and-so. So we've got about 12 core individuals working on the project right now. And then at least that many others in peripheral teams and main groups. So like our platform team, we have a couple of people representing that group's interest and the issues that need to be brought to the surface. And then we share that back with a larger group. And so we have lots of regularly scheduled meetings now across the core groups and we've in fact split them off into sub streams so that we can pay attention to the details, but then also bring the big issues back together. And again, not lose the forest through the trees and vice versa even. So we're doing that. But again, what we're thinking about is for all of our journals, which means that if we're using editorial manager, the ARIES platform, this is a legacy system. So we have to be careful, right? It's not like we're starting from scratch and introducing a data field when a system's being started. In fact, what we have to do is make sure that we're not going to break the legacy system by inclusion of a new data field. And we have to recognize that other parts and other types of data that are being collected in editorial manager can be shared in ways that the gender identity and race and ethnicity data cannot. For example, the gender and race and ethnicity data cannot be made available at any point during the peer review process period. And we have to ensure that that's the case. So thinking about that on scale, but also the different groups and individuals who have access to different information throughout the editorial workflow process, we have to actually think through each of the steps. And so we'll be working out the entire workflow and considering that and essentially establishing a set of rules that govern it. You mentioned the recognition that no schema is perfect and that Elsevier didn't want the perfect to be the enemy of the good. Librarians have an ever-increasing awareness of how library schema have to be revised and updated as problematic aspects emerge within categories, terminology, and even the hierarchy and location of certain categories within the entirety. Can you talk about how consultative this effort is and how critiques will be handled? Yeah, so look, we do know that we will unlikely get to a final perfect schema, not for gender identity and not for race and ethnicity identity data either. But we don't want the perfect to get in the way of the good and work that should be done. And just already as we've done for gender, look, we were collecting data. We were collecting it in a way that at this time, we don't think we were doing it as robustly and as well as could be done. So we revised our approach. That doesn't mean that anybody was doing something wrong before. It's just that times have changed. Information has made its way to us and we had the opportunity to reconsider the approach that we take to that schema and how we're including it into our system. So we're no longer only capturing data in a binary fashion. We make sure that it's not only that. So we have gender diverse and non-binary option available for individuals. And we have, and making sure that the prefer not to disclose optional be there because that is essential to meeting compliance. The same time we also thought about ways that we care about inclusivity but also being careful not to collect data that we don't need. So we had concerns about being able to capture information for cis and transgender individuals but recognize that if we ask somebody to disclose if there's cis or transgender, then we'd also have information about sex, biological sex for those individuals. So while we're not specific and our schema lacks specificity, we are inclusive. And women is intended for both cis and transgender individuals and gender fluid individuals who selected that as the identity that they most identify with. Similarly for race and ethnicity, we're gonna need to think about that. And that is really quite complex. So we're having to look to the literature. This is great, right? We are an STM publisher and we're working with a group of about 34 other STM publishers, which is fantastic. So we have a lot of literature to draw upon. We also decided that it might be good to have a third party expert, somebody who is not within publishing. And so we have identified a scholar and researcher in this area. She is a sociologist. And so we've just recently secured her as an external expert to be working not only with Elsevier but to provide guidance and feedback into the RSC-led multi-publisher group as well. And whatever schema we develop, we also recognize that we want that input from the global research community. So we'll use the resources that we have available at Elsevier to be able to test that schema. And we will look very carefully at the feedback that we take and make changes to the draft schema. We'll share that with our external consultant with all 35 publisher partners in the RSC-led group and consider what changes we need to make. And then we will settle on a schema and that is what we will introduce. And just as we did with gender, we will revisit this schema. So we will revisit the gender identity schema. We will revisit the race and ethnicity. What we're thinking is not less frequently than on an annual basis, which doesn't mean we'll have to change things on an annual basis, but not less frequently than once a year we look at it and say, does this still meet our needs? Does the way that we are presented the schema and asking the questions and the options we make available suit the needs of the research community that we're serving? And we'll ask those questions and make changes. We have a gender equity task force with lots of work streams and these are in place and we have support from the highest executive levels of the organization. So I don't think that anybody is going away. In fact, more people seem to be asking to join our efforts across the organization and having 35 publishers all wanting to work together towards this common goal suggests to me that there is going to be a sustainability that we're gonna have in that regard. We have about two minutes left. Is there any final thoughts you have that you wanna convey about this initiative? Certainly if people have any resources, especially from the literature that they'd love to share with me, that's great. We've actually established a Mendeley library where we've begun to capture all of this literature and we'll be happy to invite to join this library and have access to it. Anybody within the library community or elsewhere that would be interested in that and certainly then the opportunity to be able to share those resources with us. So that's great. And since our slides are going to be shared when yours and my email addresses are on that last thank you slide. So everybody will have ready access to our emails. But that's really what we're looking to do. And then we encourage those of you who are researchers and editors and reviewers who work with us in the various journals and on the society partner journals that we publish that you'll take the opportunity once these data fields are available to think about sharing your data as part of this process so that we can work with you to ensure that there is better diversity and inclusivity at least the gender and the race and ethnicity dimension and that will allow us to start to think about other dimensions of IND as well. Thank you Holly. Thank you Gwen. This is Gwen Evans and Holly Falkrasinski saying thank you for your time. Thanks.