How we score leaders

Why a scorecard

Debates about African leadership usually run on memory and emotion. The KemetLeads Scorecard adds a second layer: for every head of state or government we profile, we measure what can be measured, using the most credible public frameworks that exist — and we say plainly what cannot be measured.

The five criteria

1. Development delivery. What the country's people actually got during the tenure, from three long-run series that exist for every African country since 1950: GDP per capita (Maddison Project, percentage change), life expectancy at birth (years gained), and child mortality (change in the share of children dying before five, UN data via Our World in Data). Each is ranked against the rest of the continent; the criterion is the average of the three ranks.

2. Democratic trajectory. The change in the V-Dem electoral democracy index (0–1) over the tenure. V-Dem aggregates the judgments of thousands of country experts into continuous scores with published uncertainty, and its data reach back to 1789 — so the same instrument covers Nkrumah and Ruto.

3. Rule of law & control of corruption. The change in V-Dem's rule-of-law index combined with its political-corruption index (corruption inverted, the two averaged, 0–1). Chosen over the World Bank governance indicators, which begin only in 1996, so that a 1960s tenure and a 2010s tenure are measured with one ruler.

4. Rights & participation. The change in Freedom House's political-rights and civil-liberties ratings, combined into a single 0–1 freedom score (1972 onwards) — an independent cross-check built by different people, from different sources, than V-Dem.

5. Constitutional conduct & exit. The only criterion scored at the level of the individual leader, modeled on the standard of the Ibrahim Prize for Achievement in African Leadership: did the leader gain power with democratic legitimacy, govern within constitutional limits, develop the country while strengthening rights, and leave office peacefully? We code each of these four checks from the historical record as yes, partial, or no — for any era.

The honesty rule

Criteria 1 through 4 are country-level measurements. A country's score can move because of a leader — or because of war, drought, commodity prices, or the world economy. The World Bank itself warns that its indicators cannot attribute change to specific policies or people. So the Scorecard presents these numbers as change during the tenure, never as proof that the leader caused it. Criterion 5 is the exception: conduct is personal, documented, and codeable.

The comparability rule. A raw change is not comparable across eras: life expectancy rose almost everywhere after 1960, and incomes fell across the continent in the 1980s. So every cell is reported next to the median change of all other African countries over exactly the same years, and as the leader's percentile among them — P82 means the country improved more than 82% of its peers over that window. A 1960s tenure is judged against the 1960s, a 2010s tenure against the 2010s, and the same four sources are used for every leader. Windows shorter than three years are not measured. For leaders removed by coup, the final calendar year is excluded, since its annual value is partly coded under the successor regime.

Historical figures

Kings, queens, and pre-constitutional leaders — from Mansa Musa to Shaka — cannot be fairly graded against constitutional standards that did not exist in their world. They are profiled, not scored. Post-independence heads of state and government are scored on criterion 5, and criteria 1–4 show the measured change across their tenure wherever the source data covers it — the exact years used are printed next to each number. Where a tenure predates a dataset, the row says no data rather than guessing; a leader whose tenure was cut short within months is marked not assessable. For leaders removed by coup, the final calendar year is excluded, since its annual value is partly coded under the successor regime.

Sitting leaders

Leaders still in office receive a provisional score: the peaceful-exit check cannot be coded until they leave. Provisional scores are recalculated as events unfold.

What citizens said

Alongside the Scorecard, each leader's profile displays citizen-trust data from Afrobarometer, Africa's largest independent survey network. The bars show the percentage of respondents who say they trust the president, parliament, or the courts of law "somewhat" or "a lot" — country-level results from Round 9 (2021–2023), covering 36 of the countries in the KemetLeads roster.

These numbers are not part of the Scorecard. They measure public opinion at a point in time, not the performance of an individual leader. They appear on every profile whose country was surveyed, regardless of when the leader served — because trust in institutions is part of a country's story, even when it cannot be attributed to a single person.

Sources

Maddison Project Database 2023 (Groningen) · UN World Population Prospects and UN IGME child mortality, via Our World in Data · V-Dem Institute, University of Gothenburg (v16, 1789–2025) · Freedom House, Freedom in the World (1972–2024) · Ibrahim Prize (Mo Ibrahim Foundation) for the criterion 5 standard · Afrobarometer Round 9 (2021–2023), Dispatch No. 891, for citizen-trust data. The build is a script with cached source files, so every number on the site can be regenerated. Scores are editorial assessments by KemetLeads based on these public frameworks; corrections with sources are welcome at admin@kemetleads.com.