Thinking fairly throughout: A guide to embedding Equality, Diversity, Inclusion and Intersectionality (EDII) principles within research

Equality, Diversity, Inclusion and Intersectionality (EDII) principles are increasingly recognised as central to robust research practice. Rooted in the civil rights movements, diverse perspectives have proven to boost innovation, creativity, problem-solving, and productivity. By the 1980s and 1990s, diversity management gained prominence, and since the 2000s, EDII has been central to institutional and funding strategies. Challenges like systemic bias, unequal access, and resistance to change persist, highlighting the ongoing need for robust EDII policies to ensure research environments are equitable, inclusive, and impactful. 

By embedding EDII at every level-from team composition to research design and dissemination researchers can, at least aim, to produce findings that are more relevant, equitable, and transformative for a greater number of people[1],[2]. Fostering environments that are not only diverse but also inclusive and equitable, allows research organisations to maximise innovation, productivity, and societal relevance, while addressing systemic barriers that hinder participation and impact[3]^{,[4], [5]}.

For example, it is known that gender-balanced teams are more likely to produce novel and highly cited work[6], while interdisciplinary and cross-institutional collaborations further increase innovation and long-term impact[7].  We also know that diversity alone is insufficient.  For example, underrepresented PhD researchers often produce highly original work, yet their contributions are less likely to be recognised or taken up, highlighting the need for intersectional approaches to equity (known as a diversity-innovation paradox).[8] Inclusion - the equitable integration and recognition of diverse contributions - is critical to translating varied perspectives into creative, impactful research. And yet, the ability to look through an EDII lens when assessing research design, process and outputs has not been a skill explicitly or routinely taught[9] to new researchers until fairly recently.

From the outset BR-UK has placed EDII at the heart of our programme of work.  In BR-UK, we believe that providing equitable opportunities for people to be involved in research ensures that the resulting research is unbiased, of higher calibre and produces outputs with broader insights that are a fair representation of systems and behaviours. In 2024 we developed a range of guidance documents to help our research team consider how to be more equitable and inclusive and in 2025 we published our first reflections on our progress to share with, and learn from, others in an open and transparent way.

In 2025 BR-UK hosted a webinar on the importance of EDII in behavioural research which encouraged participants to consider approaches to foster more equitable and inclusive research. We also presented preliminary findings from a rapid scoping review assessing how successfully EDII principles are represented in the current behavioural research landscape. Webinar attendees told us that while they recognise the importance of EDII, they were either unaware of existing practical guidance about how to embed principles and activities throughout the research process, and/or didn’t feel educated enough to implement approaches that would be successful.

In response to this we have developed a structured framework, based on the literature and our own practices, that integrates EDII considerations at each stage of the research and provides prompts to help others embed EDII into their research. 

Many of us are well versed in highlighting the limitations of research studies and outcomes where the results ‘cannot be generalised’. Often, whilst focusing on the research variables of interest at the design stage, the specific details of who takes part, who will be affected by the study outcomes and who undertakes the research may not be given sufficient consideration. These early design decisions have significant implications for subsequent research stages, i.e., we cannot represent views or experiences of those that are not included or data that we do not collect. So how can we do better?

Inclusive research design involves writing and reviewing research proposals to ensure they include considerations relevant to diverse populations and incorporate intersectional perspectives. Study design should therefore strive for equal representation across demographics and support participation from underrepresented groups.

Ways to do this include conducting a literature review of the existing data and research activities to understand the where the road to addressing the research question has led so far and who it has/has not benefitted in the past or consulting with community stakeholders during the design phase to ensure research questions are appropriate and meaningful.

Good public and patient involvement engagement (PPIE) from the design stage is also fundamental. Organising and liaising with a PPIE group can make sure that research questions are appropriate, the methods (including recruitment) are fair but realistic, and reaches far enough to include greater representation. Importantly, testing ideas with people with lived experience can help obtain feedback on whether the ideas are actually measuring the variables we think they are. For example, do we have good construct validity or would we be introducing confounding factors that as researchers we may not have even considered?

The composition of the research team undertaking the study should also be carefully considered to ensure inclusion of researchers from varied backgrounds who can bring different analytical and interpretive approaches to the work.

It is important to remember that communities (and individuals within them) will have differing experiences and histories that influence whether they would be open to participating in scientific research.  Studies that include diverse participants whilst addressing potential barriers to participation, are more likely to run to time, avoiding extensions and extra costs. Identifying appropriate participants is therefore key.  It is area where having range of advisors can help ensure that the recruitment strategy targets a diverse pool of people and does not present the study in a way that would alienate or turn people away from participating.

The importance of inclusive recruitment strategies is illustrated by the UK Spectrum 10K autism study[10]. Although the project aimed to advance large-scale genetic research, some autistic self-advocates expressed concerns that recruitment and study processes did not fully reflect community priorities or address questions about consent, communication, and data use. This feedback led to uncertainty and reduced trust, and the study was ultimately paused and closed following further consultation. The case highlights the value of early, meaningful engagement with participant communities to ensure recruitment approaches are transparent, respectful, and aligned with participants’ expectations.

Thus, it is important to consider factors such as location, timing, practical constraints (such as childcare or transport costs), as well as appropriateness. Recruitment materials should be made available in plain language, languages other than English and in a range of formats (including Braille) to ensure accessibility across different populations. 

The core ethical principles for all research involving humans[11] includes the expectation that participants should feel comfortable, their dignity is respected and the research maximises benefit whilst minimising harm.

This means that data collection methods should be used that capture the experiences of diverse groups whilst treating all participants equitably. This involves using culturally appropriate communication and tools, providing materials in various languages and formats as mentioned above, and collecting feedback on participant experiences. Data should be collected in ways that enable analysis of intersecting identities, recognising that experiences are shaped by multiple, overlapping demographic characteristics.

Data collection that involves sensitive data, both qualitative and quantitative, should be tested with representatives from the target group, to make sure the process is comfortable, non-threatening and does not transgress power dynamics.  Before launching a smoking cessation trial, the research team sought input from people with lived experience of homelessness to ensure our research approach was equitable, inclusive, and respectful. Their feedback highlighted that some baseline questions felt unnecessarily intrusive and risked reinforcing power imbalances, leaving participants feeling vulnerable or uncertain about our intentions. In particular, participants questioned the relevance of questions about other drug use or parental status within a smoking cessation study. This highlights how research practices can unintentionally exclude or alienate participants, and underscores the importance of co-designing research tools that prioritise relevance, transparency, and trust.

Including data analysts with diverse backgrounds helps ensure multiple perspectives are considered, reducing bias and enriching the interpretation of research findings. Data should be analysed in ways that treat all groups equitably and avoid reinforcing stereotypes. For example, interpreting alcohol use differently by gender without considering context or intersectional factors can reinforce stereotypes. Equitable analysis ensures findings reflect real patterns, not assumptions, producing fairer and more accurate insights.

Intersectional data analysis is increasingly recognised as methodologically and ethically important, especially in research aiming to inform policy or practice[12]. It can examine how combinations of identity factors influence outcomes and provide an understanding of the impact of intersecting identities on research findings. NB: this type of analyses is not always possible if, for example, the sample size and statistical power is too small or the relevant data has not been collected. 

Considering EDII when disseminating research is vital to ensure the findings reach, resonate with, and benefit all relevant communities. Without careful attention, dissemination can unintentionally exclude underrepresented groups, miscommunicate results, or reinforce stereotypes. By using inclusive language, accessible formats, and channels trusted by diverse audiences, researchers can ensure their findings are understood, relevant, and actionable, enhancing both the ethical and practical impact of their work.

This might include the use of social media platforms, publishing policy briefs, and hosting community events to reach diverse audiences. When designing dissemination materials, they must be accessible and sensitive to intersecting identities, with content available in formats that accommodate different needs.  Offering opportunities for participants or those most affected by the research outcomes to review the materials prior to publishing can avoid any issues.

Involving the people most affected by your research in dissemination, by providing financial assistance to attend events (for example using funder-backed travel costs or participation grants), helps them feel included and builds trust while strengthening ongoing relationships. It is a great way to make people feel included and nurture relationships. 

EDII is critical to ensure research informs policy in ways that are equitable and effective. Engaging diverse stakeholders in research dissemination and policy dialogue helps ensure that findings reflect the experiences and needs of all populations. This approach supports intersectional policymaking, addressing compounded inequalities faced by people with multiple marginalised identities, and increases the likelihood that policies are relevant, inclusive, and impactful across society.

During the COVID‑19 vaccine rollout in the UK, engaging diverse populations - including community leaders and advocacy groups - ensured policies addressed barriers faced by marginalised populations, such as ethnic minorities and people experiencing homelessness. This inclusive approach increased vaccine access, reduced disparities, and made policy more equitable and impactful[13].

EDII measures must be tailored to the population, setting, and research goals; generic measures may not capture meaningful differences. Avoid simply counting numbers (for example simple percentages of women or ethnic minority participants), do not focus on single variables which can obscure how overlapping social identities shape experiences or outcomes, and remember that diverse representation does not guarantee equitable participation or influence; inclusion must be actively fostered.

Measuring EDII for impact requires a combination of quantitative, qualitative, and process-focused indicators that reflect both who is involved and how they are included. Avoiding tokenism, respecting privacy, and incorporating intersectionality are critical to producing valid, actionable insights. It also ensures that research and policy initiatives actually achieve meaningful change, rather than just ticking boxes.

Systematic evaluation and reflection on EDII implementation by tracking team and participant demographics, assessing representation of diverse groups, collecting feedback on accessibility of methods and materials, and using intersectional analysis to examine how identity combinations affect findings is essential. Measuring whether participants or stakeholders feel valued, heard, and engaged is a means to inform future approaches and highlight areas where things didn’t go as well as was hoped.

Measurement approaches include surveys and questionnaires, focus groups and interviews, demographic analysis, intersectional analysis, and ongoing feedback mechanisms.

Specific metrics to collect may include the demographic composition of research teams, representation of participants across demographic groups, accessibility of recruitment materials, inclusivity of data collection methods, bias in data analysis, reach of dissemination efforts, and stakeholder engagement from varied backgrounds.

For data analysis, we recommend conducting subgroup analyses to explore differences in outcomes and using statistical techniques to identify and mitigate biases.

For dissemination, monitoring audience demographics and reviewing materials using accessibility checkers in advance is best. Assessment of policy impact is facilitated by active follow-up with policymakers and stakeholders to understand how findings have influenced policy and practice.

Inclusive research environments matter because they make all team members feel valued, foster collaboration, open communication, and idea sharing. They attract and retain diverse talent, reduce bias, and improve research quality and relevance. By incorporating multiple perspectives, inclusive teams produce findings that are more innovative, equitable, and applicable to a wider range of populations.

Embedding EDII principles throughout the research design process ensures that studies are equitable, inclusive, and responsive to the needs of all stakeholders. Integrating EDII considerations across all research stages results in more ethical, inclusive, and impactful research that better serves diverse populations. From planning and recruitment to analysis and dissemination, prioritising diversity, inclusion, and intersectionality enhances the validity, relevance, and impact of research while fostering collaborative, respectful, and innovative research environments. This approach requires ongoing reflection, learning, and adaptation. However, the benefits demonstrate the value of systematic EDII integration in research practice.