Diversity data collection in scholarly publishing

Our Joint commitment for action on inclusion and diversity in publishing group has pledged to improve inclusion and diversity in scholarly publishing with and for the research community, and ultimately for the benefit of all of society. Central to that work, which is also supported by our agreed minimum standards for inclusion and diversity for scholarly publishing, is a need to more accurately understand the current demographic diversity of our authors, editorial decision makers and reviewers. This will allow us to understand our research community both as individual publishers and as a collective, so we can put in place actions, set goals and measure progress, holding us accountable to our research community.

Therefore, the joint commitment group endorses the following set of questions for collecting self-reported gender identity as well as race and ethnicity data. We encourage all scholarly publishers and society partners to seriously consider adopting these for use with their stakeholders globally.

As an example, these questions could be used as part of self-reported diversity data for authors, reviewers, and editorial decision makers within an editorial management system.

As a diversity, equity and inclusion initiative, this is an important collaborative effort in service to the research community that took a global-first approach, which is especially important as we continue to see an increasing internationalisation of research. Our coordinated approach also means that publishers can continue to work together using a shared framework toward developing benchmarking capabilities.

Dr Holly Falk-Krzesinski, co-chair, gender equity taskforce and vice president, research intelligence at Elsevier Picture: Elsevier

How were these questions developed?

The questions were developed by the Diversity Data Questions subgroup of the joint commitment group. All members of the subgroup were given space to share their insights and we also engaged with published research in this area. For the race and ethnicity question set, Elsevier engaged a subject matter expert, , who has expertise in racial and ethnic classification on censuses worldwide.

We tested a draft race and ethnicity question set via a survey (sent to 100,000 global active researchers), which we closed when >1% of recipients had responded (1,174 total responses). Respondents were asked about their race and ethnicity (along with other demographic questions). They were also asked about their perceptions of the representativeness of the options provided, and their comfort with sharing race and ethnicity demographic information when in the role of editorial board member vs. reviewer vs. author. We used this incredibly valuable stakeholder feedback, alongside further expert input from Professor Morning, to revise the question set to the current version.

What are the benefits of using standardised questions to collect diversity data?

Inclusion and diversity in research ensures:

entry of new researchers and opportunities for researchers of all backgrounds to advance and excel throughout their careers.
a wider range of topics and research questions will be pursued.
rigorous, reproducible and higher-quality research studies; and
equitable and widespread impact of research outcomes to benefit all of society. Societal challenges of our time make it necessary for us to harness the inclusive contribution of diverse researchers to deliver equitable impact.

Diversity data enables us to define where problems such as bias lie in scholarly publishing, put in place actions, set goals and measure progress. Using a standardised set of questions to collect data puts us in the best position to create comparable (aggregated and anonymised) datasets between journals and publishers, set subject-specific baselines, and identify where action is most needed.

The joint commitment group’s intention is not to devise a single, objective or prescriptive “truth” about researchers’ gender identity nor race and ethnicity, rather develop a set of options that resonate with stakeholders that we serve from around the globe such that they are willing to self-report data. By employing these constructs, we aim for a data-driven approach to inform our goals around diversity, inclusion and equity in scholarly communications and research more broadly. The level of aggregation and number of options these question sets offer to researchers is intended to parallel the scale of diversity publishers can practically accommodate, e.g., diversity on editorial boards, as reviewers, or as invited speakers for conferences. The questions will be periodically reviewed and updated as necessary.

The standardised questions

With which gender do you most identify? Please select one option:

Woman
Man
Non-binary or gender diverse
Self describe* [opens text box]
Prefer not to disclose

What are your ethnic origins or ancestry?

Please select ALL the geographic areas from which your family’s ancestors first originated:

Western Europe (e.g. Greece, Sweden, United Kingdom)
Eastern Europe (e.g. Hungary, Poland, Russia)
North Africa (e.g. Egypt, Morocco, Sudan)
Sub-Saharan Africa (e.g. Kenya, Nigeria, South Africa)
West Asia / Middle East (e.g. Iran, Israel, Saudi Arabia,)
South and Southeast Asia (e.g. India, Indonesia, Singapore)
East and Central Asia (e.g. China, Japan, Uzbekistan)
Pacific / Oceania (e.g. Australia, Papua New Guinea, Fiji)
North America (Canada, United States)
Central America and Caribbean (e.g. Jamaica, Mexico, Panama)
South America (e.g. Brazil, Chile, Colombia)
Self describe* [open text box]
Prefer not to disclose

How would you identify yourself in terms of race?

Please select ALL the groups that apply to you:

Asian or Pacific Islander
Black
Hispanic or Latino/a/x
Indigenous (e.g. North American Indian Navajo, South American Indian Quechua, Aboriginal or Torres Strait Islander)
Middle Eastern or North African
White
Self describe* [open text box]
Prefer not to disclose

*where system functionality does not permit the collection of free-text responses, the use of “other” is an acceptable alternative to “self describe”. Some publishers may also choose not to include a “self describe” or “other” option.

While the options above are recommended for collecting geographic ethnic origins data to best reflect individual identity and maximise responses, it may be necessary at the early stages of data reporting and/or for publishers working with a smaller number of researchers to report on the data in an aggregated form to drive diversity goals.

Aggregated groupings would be labelled: Europe (Western Europe, Eastern Europe), Africa and Middle East (North Africa, Sub-saharan Africa, West Asia/Middle East), Asia and Oceania (South and Southeast Asia, East and Central Asia, Pacific/Oceania), North America (North America), Latin America (Central America and Caribbean, South America).

In collecting sensitive demographic data, it is important to be explicit about why the data are being collected, how they will be used and who will have access. Below is an example statement to use in introducing the questions. Individual organisations can use this as it stands, or as a starting point to devise their own explanatory statement.

[Publisher] is deeply committed to fostering a scientific community that supports and benefits from the talents of researchers from a wide range of backgrounds. Please help us collect the data we need to gauge our success at advancing diversity, inclusion and equity in research by responding to the questions below. Important: individuals’ gender, ethnic, or racial identities will in no way be used when evaluating journal submissions; this data will be aggregated and anonymised before being analysed and reported in order to improve our policies and processes. For details about the storage of and access to these data, please see our [Privacy Policy and/or other public-facing info site].

91AV