Research Diversity

This short talk introduces ‘research diversity’, why it matters, and how it relates to the question “who is involved in research?” As we’ll see, there is a clear relationship between the diversity of research — in terms of its contributors, participants, methods, output formats, and possible routes to impact — and some of the widely discussed aims of open research. We’ll have a look at some evidence about the relationship between Open Access and research diversity, supporting the conclusion that open access alone isn’t enough to bring about some of the key aims of the open research agenda. We’ll also have a look at some reflections on Open Source Software developer culture for some reflections on what not to do when trying to build a more diverse research culture. Lastly, we’ll offer some practical suggestions about how to incorporate more of the benefits of research diversity in your own research work.

What is research diversity?

The term Research Diversity itself spans several different meanings. What these different meanings have in common is that they are all about including a wide variety of perspectives, experiences, and voices in research. One way to achieve this is through the inclusion and representation of diverse participants and members within research teams. “Research diversity” can also refer to the diversity of research practices, output formats, and impact types.

Why does research diversity matter?

Research diversity can be linked to themes such as academic plurality — a widely held value within academic communities that embraces the harmonious coexistence of different interests, convictions and lifestyles. These values might be exercised (or even put to the test) within multi- or interdisciplinary research. This might also call to mind the varied activities and roles that are involved in creating a given piece of research in the first place. All of these considerations relate to another question: who is involved in research? As we’ll see, the diversity of research topics and output formats is linked to the diversity of researchers themselves.

Diversity of research participants

Other videos in this series point out that certain kinds of research simply cannot be done without meaningfully involving a suitable population of participants. As in the MAMMI study (mentioned in the “Participatory Research” video), first-time mothers were an eminently appropriate group to involve in a research project on maternal health. Similary, in the CLARISSA project (mentioned in the “Open Communities” video), children working in hazardous occupations are an important group to include in research that aims at systematic changes to child labour. Additional factors of diversity and inclusion need to be considered in such projects for the research to be valid and ethical. For example, the MAMMI study needed to reflect relevant factors of diversity among new mothers to be representative and useful. These factors included ethnicity, educational background and socioeconomic status. Researchers used purposive sampling to recruit women from more diverse groups, including those who were single mothers, LGB identifying, or who had experienced homelessness. While involving children working in hazardous occupations — as well as their employers — the CLARISSA project had to take particular care when it came to safeguarding. The overall points to make here are that involving diverse participants can have a range of beneficial impacts, including improving the accuracy of results, reducing inequalities, and avoiding biased results.

Diversity of research contributors

Although open research practices make barriers between academic research and wider communities more permeable, when considering research diversity, it’s useful to consider the demography of the academic research sector itself. The first thing to note is that there are many different factors and markers of diversity, including race, ethnicity, gender, age, and social class, among others. Empirically, there is a strong relationship between the characteristics of researchers and the research topics they study. For example, few people would be surprised to learn that publications about pregnancy are predominantly written by women, or that ethnic minorities tend to write more about racial discrimination. What’s more intriguing is the fact that people from underrepresented groups tend to make more innovative and novel insights, and this is often true for more diverse teams as well. Accordingly, a lack of diversity in research is a potentially significant liability for a country like the UK where women, minorities, and people with working class backgrounds tend to be underrepresented in academia, and particularly so at the higher levels of the academic hierarchy and within elite institutions. One striking set of numbers communicates in stark terms how these kinds of inequalities matter for academics themselves. According to UKRI’s equalities monitoring,

“For the principal investigator role type, applications from the White ethnic group had a significantly higher award rate (31%) than principal investigators from both Asian and Black ethnic groups (23% and 21%, respectively).”

In other words, all else equal, your chances of getting a grant may be up to one third less than that of another applicant based on what’s widely regarded as a protected personal characteristic. The reality, though, is that “all else” typically isn’t equal. Firstly, the make-up of UKRI funding panels doesn’t reflect the population of applicants, which may be associated with a difference of opinion about what kind of research is worth carrying out. In the US, bibliometric research revealed that although black and white investigators published the same number of papers during their PhD and postdoc years, papers from black investigators were cited less than half as much, had fewer co-authors, and often appeared in journals with lower impact factor. Preliminary evidence suggests that robust mentoring interventions and enhanced professional network development could help correct these trends, and reduce discrepancies in grant funding rates. However, if these interventions are given at the PhD level, it may be too late for many people who would have had a lot to offer as researchers — in both the US and UK, black students are significantly underrepresented at PhD level.

Who is involved in open research?

The inequalities that characterise research trajectories are reflected in the classic sociological concept of cumulative advantage, otherwise known as the Matthew effect (based on the principle “For to every one who has will more be given”, or in short, “success breeds success”). Open research doesn’t make this issue go away. As Ross-Hellauer et al. write in Royal Society Open Science,

“

aking processes open will not per se drive wide reuse or participation unless also accompanied by the capacity (in terms of knowledge, skills, financial resources, technological readiness and motivation) to do so. These capacities vary considerably across regions, institutions and demographics. Those advantaged by such factors will remain potentially privileged, putting Open Science's agenda of inclusivity at risk of propagating conditions of ‘cumulative advantage’.

A more explicitly inclusive approach to building an open research culture would be to foster discussions around what works and what doesn’t work, who is included, and who is excluded by default. Let’s turn to some of these points now.

Open Access and research diversity implications

In terms of getting ideas out there, open-access papers draw more citations from a broader readership: this hints at key benefits including a diverse readership and diverse possibilities for reuse. However, it’s currently the case that OA articles with Article Processing Charges (or APCs) are more likely to be written by people possessed of the male gender, employment at a prestigious institution, association with a STEM discipline, more research funding, and a more advanced career stage. It’s the Matthew effect all over again. It should come as no surprise that the Higher Education Policy Institute concludes that Open Access is not enough to foster a vibrant open research ecosystem — they touch on the necessity of developing capabilities, connections, coordination, collaboration, and co-production strategies: themes that have also appeared in this video series.

Building more diverse teams can start with something as simple as working on projects that you wouldn’t normally work on, with people you wouldn’t normally interact with. When it comes time to publish, mechanisms like the CRediT taxonomy can help make sure that the diverse inputs to a research publication are suitability accounted for. That said, the rise of open research is associated with a broader ongoing culture shift away from an exclusive focus on publications and citation counts, and towards other ways of evaluating the impact of research. In the next REF, the weighting given to academic publications will be reduced from 60% in REF 2021 (down from 65% in REF 2014) to just 50% in REF 2029. There will be a corresponding increase in weighting given to ‘people, culture and environment’ — i.e., the factors that support the production of high-quality research and impact in the first place. As we’ve noted, involving diverse participants — both as research subjects and as research collaborators — can increase the robustness and novelty of your research. However, research diversity it’s not just about research quality in terms of outputs. Diversity in research is crucial both for reasons of social equity, and because building our capacity to work with difference enhances our own learning.

Matthew Gandy. (2024) Losing control: REF 2029 and the downgrading of academic outputs https://rgs-ibg.onlinelibrary.wiley.com/doi/10.1111/tran.12713

Hui, W. (2020). The Humanities in China: History and Challenges. In History of Humanities (Vol. 5, Issue 2, pp. 309–331). University of Chicago Press. https://doi.org/10.1086/710276

Ecological Thinking: The Politics of Epistemic Location (Oxford University Press, 2006)

David Rock and Heidi Grant (2016). Why Diverse Teams Are Smarter. Harvard Business Review. https://hbr.org/2016/11/why-diverse-teams-are-smarter

Robin J. Ely and David A. Thomas (2020). Getting Serious About Diversity: Enough Already with the Business Case. Harvard Business Review. https://hbr.org/2020/11/getting-serious-about-diversity-enough-already-with-the-business-case

Bas Hofstra, Vivek V. Kulkarni, Sebastian Munoz-Najar Galvez, Bryan He, Dan Jurafsky, and Daniel A. McFarland, “The Diversity-Innovation Paradox in Science,” Proceedings of the National Academy of Sciences, vol. 117, no. 17, 2020, pp. 9284-9291.

Ginther, D. K., Basner, J., Jensen, U., Schnell, J., Kington, R., & Schaffer, W. T. (2018). Publications as predictors of racial and ethnic differences in NIH research awards. In L. G. Koniaris (Ed.), PLOS ONE (Vol. 13, Issue 11, p. e0205929). Public Library of Science (PLoS). https://doi.org/10.1371/journal.pone.0205929

Ben Green (ed.). “Technology Ethics in Action: Critical and Interdisciplinary Perspectives,” Special Issue of the Journal of Social Computing. 2021.

Marylin Vantard, Claire Galland, Martina Knoop; Interdisciplinary research: Motivations and challenges for researcher careers. Quantitative Science Studies 2023; 4 (3): 711–727. doi: https://doi.org/10.1162/qss_a_00265