The Case for Diversity and Inclusion in Research Information

Who benefits from science? Who decides which knowledge is produced? Who participates in research? Which research topics are considered important? Is science really a public good enjoyed by all humankind?

The answers to these questions point in the same direction: today, science is largely conducted in alignment with the interests of rich countries and people. Either directly or indirectly, mainstream scientific knowledge feeds into forms of innovation, like some AI algorithms or engineering for mining, that actively contribute to social and economic inequalities.

This is in contrast with the consensus that science should be a public good and that all humankind should enjoy it benefits. Already in 1948, the Universal Declaration on Human Rights established that “everyone has the right freely to participate in the cultural life of the community, to enjoy the arts and to share in scientific advancement and its benefits.” This view has been reaffirmed and newly articulated in international commitments like the 2021 UNESCO Recommendation on Open Science.

In this piece, I argue that for science to be a public good, there is a need for diversity and inclusion in science—and specifically in research information, i.e., information about what type of knowledge is produced, where, and by whom. In practice, this means: i) fostering pluralism in the research topics investigated; ii) supporting the participation of broader communities in research; and iii) making this diversity visible in research information systems.

Why Does Pluralism in Research Topics Matter?

Given that social and environmental contexts are so disparate across the world, science must embrace a diversity of research topics to improve wellbeing and human development. Even shared issues—like climate change, food production, or vaccination—differ in how they are investigated across cultural and geographical contexts. Scientific agendas too often concentrate on the issues and framings that are relevant to high-income countries and powerful or privileged actors.

The historical role of science in colonialism and environmental degradation offers an example of this concentration of research toward the interests of the powerful rather than the global public good. We continue to see relatively little research on the topics that are more relevant to low-income countries, women, cultural minorities, or indigenous groups.

Research efforts across diseases offer a classic example of the focus on the problems of the rich: there is little research on diseases like malaria and conditions like cardiovascular disease that affect poor populations in comparison to investments in conditions like cancer that are relatively more frequent in richer populations. For example, in India, cancer research amounts to 22 percent of disease-related research although cancer is less than 5 percent of the disease burden; in high-income countries, cancer is the subject of 24 percent of the research and 18 percent of the disease burden. Similarly, in agriculture, there can be a mismatch between topics, like nutrition, that are more relevant to many low-income countries and their share of the research focus.

Another area of imbalance is research on artificial intelligence. In this case, research has been concentrated on topics that are relevant to big corporations, such data-hungry and computationally intensive deep learning methods, rather than the societal and ethical implications of AI.

In short, if science is to address the challenges faced by disparate communities across the world, the contents of research must become more diverse and more equitable.

Why is Inclusion of Diverse Social Groups Important?

To foster a science that addresses a wider range of topics and, in particular, societal challenges, participation in research from diverse social and cultural groups across the globe is key.

Initiatives in this direction are being increasingly developed—for example, research that involves patients, environmental activists, or citizen science organizations. These non-academic participants contribute to science by collecting data, shaping research agendas, or conducting experiments. Writing and communicating research in local languages also facilitates connections with local stakeholders and should be properly valued, as stated in the Helsinki Initiative on Multilingualism.

However, research agendas in global science remain heavily unbalanced and social groups unequally represented, either in terms of gender, cultural identity, or ethnicity. Major efforts need to be pursued to find better ways to deepen and widen the participation of stakeholders so that research is more aligned with the needs and aspiration of societies around the world.

Making Diverse and Inclusive Science Visible Through Open Research Information

But making science more diverse and more inclusive is not enough. Science and scientists that address topics relevant to marginalized communities must also be made more visible.

Scientific visibility is achieved through research information sources like Web of Science or Scopus. However, since these sources are based in North America and Northern Europe, they often make more visible the science produced in, and of interest to, these territories.

This is explained in the Barcelona Declaration on Open Research Information, which warns that science policy decisions are routinely made “based on information that is biased against less privileged languages, geographical regions, and research agendas.”

Research information sources should avoid these biases by becoming more inclusive and comprehensive. I believe that the best way to achieve a comprehensive coverage of topics and authors is to develop a federation of open information sources, as illustrated in Figure 1.

The ideas is to develop a multiple-perspective science observatory (a “Multiversatory”) that can generate research information in a plurality of ways, so that it provides information relevant to particular and diverse contexts and according to transparent data and assumptions. Relevant metrics on agricultural research in the Peruvian Andes, the Pakistani Himalaya, and in California are likely to be different even if they are about the same crops, as researchers in these geographical locations conduct and publish research in very different ways. To achieve this contextualization, multiple information sources, including local ones, need to be federated and made interoperable, while keeping the information in formats that are accessible to many users, as recently shown by Alysson Mazoni and Rodrigo Costas.

Making Knowledge Visible for All

Science has historically been built on unequal relationships within and across countries. As a result, it excludes some populations from participating in its production and from benefiting from its progress.

Scientific institutions should make efforts to become more diverse and inclusive both in terms of topics and people. This diversity needs to be made visible through more open and comprehensive research information sources. Information sources also need to be more accessible so that different potential beneficiaries can share and reach the knowledge that matters for them. Only then can science truly become a global public good.

Acknowledgements: This piece draws from previous contributions to the CWTS Madtrics blog: on diversity and inclusion in science and on open research information sources.

References

Ciarli, T., Aldoh, A., Arora, S., Arza, V., Asinsten, J., et al. (2022). Changing Directions: Steering science, technology and innovation towards the Sustainable Development Goals. STRINGS, SPRU, University of Sussex. https://sussex.figshare.com/articles/report/Changing_Directions_Steering_science_technology_and_innovation_towards_the_Sustainable_Development_Goals/23492681/1

Harding, S. (2008) Sciences from Below: Feminisms, Postcolonialities, and Modernities. Duke University Press.

Leach, M., Stirling, A., & Scoones, I. (2010). Dynamic sustainabilities: technology, environment, social justice. Taylor & Francis. https://library.oapen.org/handle/20.500.12657/52748

Rafols, I., Costas, R., Bezuidenhout, L., & Brasil, A. (2024). The multiversatory: fostering diversity and inclusion in research information by means of a multiple-perspective observatory. Conference on Advancing Social Justice Through Curriculum Realignment. https://osf.io/preprints/socarxiv/dn2ax

Rafols, I., & Stirling, A. (2021). Designing indicators for opening up evaluation: Insights from research assessment. In P. Dahler-Larsen (Ed.), A Research Agenda for Evaluation (pp. 165-193). Edward Elgar Publishing. https://osf.io/preprints/socarxiv/h2fxp