The research project was conducted by two academics who, despite being non-Ugandan, have over a decade of experience working in Uganda. In addition to AfT, the research team collaborated with several other implementing partners, including The Department of Political Science at Makerere University, Innovations for Poverty (IPA), and Leo Africa Forum, among others. The research team signed a Memorandum of Understanding (MOU) with each of these partners at the outset of the project that delineated the roles and responsibilities for each partner. For AfT, this included providing “advice on study design, implementation, and interpretation of results,” as well as providing “strategic support in acquiring clearances and permissions.” The MOU also included details about the data ownership and publications, dissemination of findings, financial matters, and general principles of the partnership.
This field experiment assessed the conditions under which information about candidates and government performance affects voter behavior in Uganda during the 2015 primary and 2016 general elections. The study found that candidate videos in which parliamentary candidates answered questions about a set of policy issues can shape voter behavior and that information about candidates reduces uncertainty about opposition candidates. These effects may in turn lead some voters to switch to opposition candidates.
Findings from the study were disseminated in a workshop with government and civil society leaders in Uganda, which included presentations by the researchers, key note addresses by implementing partners, and a panel discussion on the potential role of candidate debates to feature in upcoming parliamentary elections in Uganda. Participants included the head of the Electoral Commission, the spokesperson of the ruling party, the President of the main opposition party, and nationally recognized journalists.
The project also produced a working paper and an EGAP policy brief:
In the sections that follow, the brief provides Rwakakamba’s responses to a series of questions asked during an interview about AfT’s decision to partner with the researchers, their working relationship, and his reflections on the dissemination of findings. It also includes his policy recommendations for future researcher/implementing partner collaborations.
Decisions to Partner with the Research Team
Unlike some collaborations, the research team already had an established relationship with AfT, especially with Rwakakamba, after working in Uganda for over a decade. Consequently, the collaboration between the researchers and AfT on this project was organic in its origins. This prior relationship was ultimately essential to working as a cohesive unit without encountering many problems throughout the research project.
When asked about the contributions of AfT prior to the start of the project, Rwakakamba noted that there was not a significant amount of consultation in terms of the research design—this was a likely consequence of the granting and project organization (the project was part of the Evidence in Governance and Politics (EGAP) Metaketa Initiative, which requires a high degree of coordination across studies), as well as the research team’s overall knowledge of Uganda’s political system. Rwakakamba did, however, appreciate the research team’s efforts to consult with AfT prior to the project, relating to, for example, potential sample locations. That said, AfT was primarily focused on project implementation rather than design.
Nevertheless, Rwakakamba wishes that prior to the start of the project there was a more sustained effort, in addition to the MOU described above, to delineate roles and responsibilities, including specific tasks for AfT and coordination with other implementing partners. He notes that this would have clarified the working relationships within and throughout the project and helped ensure smoother working conditions and relationships between the research team and implementing partners.
Working with the Team
As is often the case, AfT was essential in helping to “clear the way” for the research, i.e. in developing an on-the-ground network of support for the research project. In many ways, AfT’s presence provided a sense of “legitimacy” for the research and helped secure institutional backing such as support from the NRM Electoral Commission. As is clear from this project, implementing partners often work not only in direct support of the project, but also in the background providing support and legitimacy to the research teams. This often allows for smoother operations and problem solving as the research is conducted. AfT was also essential in helping obtain permissions from the central government, which in turn, facilitated access to local communities. This is especially important in politically sensitive contexts or when a project itself could be seen as controversial. In this case, given that the team was looking at how the electorate votes, Rwakakamba and AfT worried that politicians would fear that the research would undermine these politicians’ power.
Dissemination of Findings
Rwakakamba’s main suggestion for improving the collaboration is regarding the dissemination of findings to AfT and its on-the-ground staff. While EGAP, IPA, and the research team organized a high-level dissemination event in Kampala attended by many government and civil society leaders, for which Rwakakamba served as the keynote speaker, he believes there could be more follow-up beyond the presentation and also more resources provided post-project, beyond policy briefs. According to Rwakakamba, “there was limited capacity for syndication [of the results] with radio or television…this is a message that should have gone all over. Providing resources to help implement findings, whether through training and/or physical resources, is important for projects, like those in the Metaketa Initiative, that have not only theoretical interests but also real-world implications.” As a result, Rwakakamba notes that the results of the project “just kind of faded,” although he hopes that with the upcoming 2021 election there will be renewed interest in the results.
In this case, the project focused on the relevant issues that voters use to select primary candidates, and ultimately, to vote for them in elections. Rwakakamba suggests that it would be useful to develop a set of resources based on the results of this project that could be adapted and used to educate the electorate about the candidates in the upcoming and future elections—essentially a scaled-up version of the intervention. In addition, he believes that providing resources about research methods, especially for new types of research, will help implementing partners to further expand and build on the findings from the study. He also hopes that in the future—not only with Metaketa projects, but with research projects in general—researchers set aside time and effort post-project for some sort of transfer of skills and knowledge beyond providing papers and giving a presentation of the findings.
That said, Rwakakamba also notes that this is a more systemic problem, rather than a problem directly related to this particular project. In general, researchers are busy with multiple projects, competing deadlines, and have set incentives, especially when it comes to grant funding, and these incentives rarely include deliverables post-write up. He ultimately believes that efforts for training and transfer of knowledge should be built into grants, such that when the research on the ground ends, more can be gleaned from the results.
This brief examines collaboration between researchers and implementers who help to administer governance interventions that may be politically sensitive. Of course, any collaboration benefits from clear delineation of roles, adequate material support and training, and regular communication. Going beyond that, however, Rwakakamba recommends that researchers should appreciate that the risks that governance intervention implementers take on do not necessarily end when the fieldwork ends. Moreover, he explains that it is important to recognize the “shared responsibility that [the research team and implementing partners]” have for the research, and that the implementing partners “also have a claim, in some ways, on the findings”—that implementing partners should be able to use these findings meaningfully in their home country or organizations post-project. Both of these points imply that researchers should not view collaborations in governance interventions as short-term, “pay-for-service” arrangements. Post project, Rwakakamba encourages researchers to stay engaged with implementers and provide resources, as this project did through its dissemination activities that could be furthered with capacity building for local partners, which are important for the implementer’s benefit and also converting findings into action.