What citizens’ priorities should change in development research

Afrobarometer published What Africans Want on September 24, 2026, drawing on 50,961 interviews across 38 African countries in 2024/2025. In its results, Afrobarometer reports that 38% of respondents named health among the three most important problems for their government to address, compared with 33% who named unemployment. The finding is a prompt for development teams: when a programme claims to serve public priorities, whose account of those priorities shaped its design?

Afrobarometer says health has overtaken unemployment in its most recent survey results, while unemployment has remained a leading concern over the past decade. That change deserves attention, but not a simplistic reading in which a new ranking invalidates work on livelihoods. A careful team would ask how its proposed activities connect to both concerns, and whether the connection is visible to the people expected to benefit.

The gap is a research question

Afrobarometer says the African Center for Economic Transformation commissioned the analysis to compare citizens’ survey responses with priorities raised in development-cooperation consultations in Rabat and Kigali. Afrik describes a distinction between citizens’ immediate concerns, including health and employment, and experts’ emphasis on measures such as industrial policy and economic diversification. These are different kinds of answers to different kinds of questions, not necessarily competing visions of development.

Industrial policy might be intended to improve employment; economic diversification might be expected to strengthen the resources available for services. But intent is not evidence that citizens will experience either result. Researchers should make the proposed chain of change explicit: which near-term problems does a strategy claim to address, through what steps, and on what timetable? Monitoring plans can then test that chain instead of treating expert agreement about a policy as proof of public relevance.

National averages are not a mandate

Afrik reports that the leading concern differs by country: health in Guinea-Bissau, Uganda and Liberia, and unemployment in Namibia, Lesotho and Morocco. A regional finding can open a useful conversation, but it cannot settle the priority for a particular programme area. Teams planning work across locations should inspect the relevant country findings before turning a broad pattern into a target or a funding rationale.

Question wording matters as much as geography. Afrobarometer asked about problems respondents considered among the three most important for government to address. That is not the same as asking people to choose a preferred development project, identify the cause of a problem, or approve a proposed solution. For AI data teams preparing summaries or classifications, preserving that distinction is part of the analysis: a concise output should not quietly turn a stated concern into consent for an intervention.

Treat responsiveness as an outcome

The report also points beyond the usual debate over sectors. Afrobarometer reports that 59% favour government accountability even if it slows decision-making, while 77% expect elected officials to follow voters’ demands rather than their own ideas. Those findings suggest that the process used to choose and revise priorities matters alongside the services a programme hopes to deliver. Consultation should not be treated as a ceremonial step completed before implementation begins.

A careful monitoring-and-evaluation plan would record not only what participants said, but how their input affected decisions and what happened when it did not. Teams could revisit stated priorities during implementation, make disagreements legible, and distinguish evidence of participation from evidence of influence. That approach is especially important when a long-term reform is being justified through benefits that citizens may not yet be able to see.

The practical lesson is to keep the survey finding, the policy theory and the eventual outcome separate. Citizen responses identify problems that deserve an answer; expert proposals offer possible routes; evaluation tests whether those routes deliver. Treating any of these as a substitute for the others would make development cooperation easier to describe, but harder to hold accountable.