Establishing and nurturing relationships with communities is critical for the success of development research involving human participants. By investing resources and effort into fostering these relationships, researchers can cultivate mutual trust, resulting in a shared understanding with participants and effectively mitigating potential issues such as resistance and unrealistic expectations within the community. Although relationship management is especially prominent at community entry, as this is the point of first contact when researchers and participants set expectations that will guide the research, relationships must also be managed to the study’s end. At the end of the study, the researcher may have accomplished their main goal, namely to collect data, but the community may not. Many communities participate in research studies because community members believe the research will generate findings that benefit them or hope to benefit from the research materially. 

Unfortunately, managing relationships in research is often overlooked, resulting in serious unintended social consequences that can create social harm. The risks of social harm are especially high when researchers and participants come from very different contexts, as these differences in context can inspire distrust and misunderstandings. Relatedly, suppose Global North researchers simply get knowledge in exchange for participant labor and incentives. In that case, fundamental power imbalances are created, potentially perpetuating colonial relationships and dynamics. From our experience, this may cause participants to feel disrespected, leading to a sense that researchers are merely using them to get funds to enrich themselves. This perception often causes participants to believe that the research will not contribute to improving their welfare, leading to careless responses during data collection and, in some cases, outright refusal to participate.

As employees of the Busara, we have more than our fair share of experience with the unforeseen risks that can arise when researchers do not manage relationships with participants effectively. As a nonprofit headquartered in Nairobi and dedicated to advancing and applying behavioral science in pursuit of poverty alleviation, much of our research bridges the worlds of the Global North – where funders and our research partners often live – and the Global South, where much of our research takes place. We often serve as a broker between these worlds. This means we have also had a front-row seat to observe how harm can manifest when the chasms of understanding that separate these worlds are not effectively bridged.

Consider an example from our early days. We were conducting a study investigating whether transferring a large sum of money to lower-income Kenyans ($500 each) would improve people’s economic and psychological well-being (Haushofer et al., 2020). The transfer of money was unconditional, which meant there were no strings attached – from our perspective and the perspective of our Global North research partners, the villagers could do whatever they wished with it. We implemented the study across several villages in Nakuru county, Kenya’s third most populous county that cuts across the Great Rift Valley. Although Nakuru is home to growth and opportunity, it is also home to great poverty: in the villages where the experiment occurred, most villagers earned less than USD 200 per month.

Our study was a randomized experiment with a pure control group, meaning half the villagers did not receive the cash transfer. The problem was these control villagers did not know why. Thus, when the funds were administered after randomization, here is what the villagers outside the experimental group perceived: many of their friends, neighbors, rivals, and relatives suddenly received a windfall worth perhaps two and a half times their monthly earnings for reasons they did not understand. In contrast, they did not receive anything. 

Some villagers were happy with this outcome because they could celebrate the good fortune of their friends and neighbors. However, others were distinctly unhappy – even envious. The result was utter chaos. Villagers who did not receive a payment, both in our control group and outside the experiment, started spreading rumors that the experiment had nefarious aims. For example, contrary to our intent that villagers perceive the money as unconditional, some claimed that the money was ‘blood money’ (cf Schmidt 2022, see also Schmidt, this volume), which compelled the recipient to perform a blood sacrifice. Others claimed that the experiment was a plot concocted by satanic devil worshippers locally known as the Illuminati and that the money was an inducement to recruit villagers into their ranks. In some cases, the intense feelings of anger, envy, fear, and jealousy even escalated to the point of domestic violence, as some men discovered that their wives took part in the experiment – and therefore received more than two months’ earnings – without their knowledge. Tensions across the villages threatened to escalate into inter-household conflict.

We therefore halted our activities during the study endline and spent some time trying to understand what had gone wrong and what we could have done to prevent it.

What went wrong in Nakuru County

We halted our study for a total of three months. During that time, we performed an extensive debrief. We interviewed villagers, engaged community leaders, and held informal community meetings, or barazas as they are known in Kenya, where we explained the study’s objectives, introduced the research team, and allowed the villagers to air their concerns. Crucial to these proceedings was the fact that the people who we hired to help carry out the project, our enumerators, were from Nakuru county and, we assume, therefore, carried a degree of trust in the eyes of our participants – although many enumerators were not from the villages included in our study.

We learned from this reflection that we had incorrectly managed our relationship with our participants. Instead of holding our barazas and engaging the community before we started our study, allowing us to establish the baseline level of mutual knowledge and trust necessary to disburse such large sums of money to a random subset of villagers, we held them afterward. This allowed rumors to fill the gap in understanding and context between the source of the money and the recipients. Instead of engaging the community after the funds were disbursed to remind them that they would not receive any money, we left them to wonder when and if their cash would be received. Instead of treating our potential participants – both those who made it into our final sample and those who did not – as partners who could help us identify issues in the research process and help co-create solutions, we treated them as objects from whom insights are extracted. In short, we had failed our responsibility of bridging the gaps in culture, levels of income and wealth, job status, and other sources of context that exist across the research world, which the Global North heavily influences, and the world of our participants.

Lessons in relationship management

Our story of the cash transfer study in Nakuru County has a positive ending. After we learned about the relationship management mistakes that we had made and implemented corrective actions, we completed the remainder of the study without incident. We conducted community exit barazas to notify the community about the conclusion of the study and to address their expectations regarding cash transfers proactively. As a result, we have successfully carried out other studies in the region without the same resistance we faced during the endline phase of this project. This suggests that our barazas successfully managed our relationships with these communities by showing them we respect them enough to return the knowledge we gained from the study to them.

Yet, our story also illustrates the harm and challenges that can occur when researchers do not prioritize developing a foundation of trust with their participants. Ineffective relationship management increases the risk of social harm, and research that does harm is unethical. Researchers, therefore, have to plan their studies to minimize the risk of this harm. If that means that studies become more expensive, that is a cost that researchers and funders have an ethical duty to shoulder.

Better relationship management strategies

Cultivating a productive relationship with community members is a resource-intensive endeavor as it requires gathering insights from and communicating with diverse individuals. Researchers often engage a specific subset of individuals, typically community leaders, to represent the broader community. While this approach may initially seem efficient in terms of time and resource management, it often falls short in providing researchers with a comprehensive understanding of potential challenges that may arise when conducting research within a specific community. Furthermore, these leaders may not consistently relay information about the study to the rest of the community.

For instance, in our study in Nakuru, we maintained regular contact with community leaders throughout the study. However, this communication did not effectively translate into a broader understanding of our research activities within the community. When we completed the baseline and intervention phases of the study, we signaled the conclusion to the community leaders. Still, we failed to adequately inform the study participants, particularly those who received no cash. This oversight resulted in significant hostility and disengagement when we returned to collect end-line data, even sometimes putting our enumerators at risk. As mentioned earlier, we were forced to halt the study and organize debrief sessions with the participants and community leaders. In these sessions, we assessed whether the expectations set at the beginning of the study were met and to clarify why not everyone received financial support. This experience underscores the importance of managing relationships effectively, from community entry and maintenance to exit.

Finally, we emphasize that the ‘communities’ that play host to research are not homogeneous but are rife with all the complex political and social tensions that one might expect from moderate to large groups of humans. This point underscores the importance of not treating communities as homogeneous wholes and thereby unintentionally perpetuating colonialist stereotypes. Instead, you should take the time to understand the political and social dynamics at play in your workplace.

Below, we highlight the steps we consider crucial for managing community relationships.

Step 1: Hire Locally. Make sure your enumerators are locals. They know the context better than you and will help you flag potential harm in advance.

Step 2: Identify focal points. To gain support and positive relationships from the local community, it is important to involve the respected figures of the community such as local administration, village heads, community group leaders, and religious leaders. As they have expertise in the local culture, traditions, and context, they could assist in resolving any issues that may arise during your research activities.

Step 3: Involve the wider community. Hold town hall meetings (in Kenya, barazas) with the community members. These meetings are important because they inform villagers about study objectives and discuss expectations, culture, and so on with the goal of a mutual understanding of the study.

Step 4: Keep an open line of communication. This nurtures the relationship. Researchers should regularly engage with community members, updating them on study events like incentive distribution and interventions to prevent misconceptions and reduce tensions. Obtain consent at each study stage (baseline, intervention, end-line) to avoid misunderstandings about procedures.

Step 5: Reflect and update. Conduct end-of-study debriefs with participants and community leaders to evaluate if initial expectations were met. Document and address any issues for reference in future research-community relationships. For instance, in our case study, debriefs strengthened our relationship with the communities, enabling Busara to conduct successful follow-up studies.Step 6: Share study results. This is not only ethical but important to the community. In recent qualitative research, participants expressed their desire to receive study results because when they agree to participate in research studies, they hope to learn something they can share with the rest of the community. Second, If researchers from the Global North simply get knowledge in exchange for participant labor and incentives, fundamental power imbalances are created, perpetuating colonial relationships and dynamics. We do not want that. The least we can do is ensure any knowledge gained from research goes back to participants in a manner that is comprehensible and useful to them.

The last steps, sharing research results and debriefing the community members, often prove to be particularly difficult to achieve. Before we conclude, we therefore want to highlight some guidelines for conducting debriefs and disseminating research findings effectively:

Plan in advance. Plan for community debriefs and sharing results during the design phase of your study. Include these activities in your protocol to hold yourself accountable. Debriefs and results sharing are resource-intensive exercises. building them into your design allows you to plan them adequately, and ensure you set aside enough resources to follow through. 

Contextualize your approach. There is no one-size-fits-all way of conducting debriefs and sharing results. Embrace dialogue when planning. Engage the community members to identify context-relevant, effective, and meaningful debriefing methods. For example, in some communities, we have an official meeting day where members meet to discuss issues affecting their communities. By engaging community members during your planning phase, you get to know such information, and you can leverage these meetings to reduce the costs of debriefing and maximize impact. 

Co-design solutions. People are experts in their own lives. They know what works and what does not work in a given context. If issues arise during the debrief sessions, ensure a clear way forward before you continue with other research activities. Be honest with what is achievable. Do not make promises you cannot keep to achieve high recruitment rates. In the long run, deceiving participants may build distrust, leading to high attrition in future study phases or resistance during future engagements. 

Learn and update. Sharing results is not only ethical but also allows you to check your interpretation of the data you collect. Use this opportunity to improve the accuracy of your findings, which will improve their reliability and credibility.Embrace transparency. Include an ethics appendix in your final publication that discusses any special ethical issues or considerations that arose over the course of your research. This enhances the replicability of your study findings by providing other researchers with better insights into the participants’ interactions with your study. Discussing ethical issues explicitly in paper appendices can also help clarify and improve ethical norms, as Asiedu, Karlan, and Udry (2021) suggested.

Conclusion

Through our experience and those of other organizations (cf MacPhail et al., 2013; Iguna et al., 2022; Wein et al., 2023), we have come to understand research with humans as an ongoing relationship between the researcher and the participants. When the researcher and researched share context and expectations, this relationship is easy to manage. However, the greater the chasm in context, the more actively that relationship must be managed, lest the research brings about unintended social harms due to misunderstandings resulting from the context gap and expectations. Sometimes, third parties that have a foot on both sides of the context chasm are necessary to eliminate these understandings. This is a role that Busara often plays – but only because we have learned through hard experience the harms that result when the chasms that separate different worlds are not effectively bridged.

References

Asiedu, E., Karlan, D., Lambon-Quayefio, M., & Udry, C. 2021. A call for structured ethics appendices in social science papers. Proceedings of the National Academy of Sciences 118(29), e2024570118. https://doi.org/10.1073/pnas.2024570118

Haushofer, J., Mudida, R., & Shapiro, J. 2020. The comparative impact of cash transfers and a psychotherapy program on psychological and economic well-being. National Bureau of Economic Research, Working Paper 28106. https://doi.org/10.3386/w28106.

Iguna, S., Getahun, M., Lewis-Kulzer, J., Odhiambo, G., Adhiambo, F., Montoya, L., … & Camlin C.S. 2022. Attitudes towards and experiences with economic incentives for engagement in HIV care and treatment: qualitative insights from a randomized trial in Kenya. PLOS Global Public Health 2(2), e0000204. https://doi.org/10.1371/journal.pgph.0000204

MacPhail, C., Adato, M., Kahn, K., Selin, A., Twine, R., Khoza, S., … & Pettifor, A. 2013. Acceptability and feasibility of cash transfers for HIV prevention among adolescent South African Women. AIDS and Behavior 17, 2301-12. https://doi.org/10.1007/s10461-013-0433-0

Schmidt, M. 2022. The gift of free money: on the indeterminacy of unconditional cash transfers in western Kenya. Journal of the Royal Anthropological Institute 28(1), 114-29. https://doi.org/10.1111/1467-9655.13655

Wein, T., Lanthorn, H., & Fischer, T. 2023. First steps toward building respectful development: three experiments on dignity in aid in Kenya and the United States. World Development Perspectives 29, 100485. https://doi.org/10.1016/j.wdp.2023.100485