In the 2025 KCSE, males made up roughly 49 percent of the cohort. Yet they accounted for between 63 and 69 percent of all grade A recipients, and 54 to 60 percent of all grade E recipients, the lowest grade in the system. This pattern has held consistently from 2018 to 2025. Boys dominate the top. Boys dominate the bottom. Girls cluster in the middle.
Nobody has a satisfying explanation for why. But it shapes everything else the KCSE data reveals about gender. And it means that any policy response which treats gender inequality as a single problem, affecting one sex in one direction, will fail.
This blog uses KCSE data from 2018 to 2025 to make three claims. First, national gender parity in enrolment masks opposite problems in different regions. Second, the system’s fragility is gendered: when schools close, girls disappear first, and the system has no plan for this. Third, the “both extremes” pattern in male performance is a structural feature of the KCSE that we do not yet understand and cannot yet address.
Parity Is Not Equity
At the national level, Kenya has achieved something remarkable. In 2025, approximately 499,000 female candidates and 490,000 male candidates sat the KCSE. Girls slightly outnumber boys. This is the product of decades of investment: free primary education, progressive fee abolition, targeted bursary programmes, and sustained policy attention to girls’ education.
However, this headline figure conceals two opposite realities.
In Garissa, Mandera, and Wajir, the Gender Parity Index (ratio of female to male candidates) remains below 0.7. Fewer than seven girls sit the exam for every ten boys. These counties also have the highest overage rates in the country, with more than 30 percent of candidates aged 21 or above. They are served almost exclusively by Sub-County schools, the tier with the fewest teachers, the weakest infrastructure, and the lowest examination outcomes. In these places, gender disadvantages, age disadvantages, and school-quality disadvantages are not separate problems. They are the same problem, experienced by the same students, in the same counties.
This is a contrast to Vihiga, Elgeyo Marakwet, and Kisumu. Girls make up 53 to 56 percent of candidates. This looks like a success story. It is not. In these counties, the female share is rising not because more girls are entering, but because boys are leaving before Form 4. Research consistently links this to household economic pressure. In low-income families, boys are pulled into seasonal farming, casual labor, and informal work. Boys from households earning below KES 10,000 per month face significantly higher risks of dropping out. Substance use and negative peer influence compound the problem. The result is a boys’ dropout crisis that registers in the data as gender parity.
The same national statistic, near-parity, is produced by girls being excluded in one region and boys being pushed out in another. Celebrating the headline number without distinguishing these mechanisms is not just misleading. It actively prevents the right policy response.
When Schools Close, Girls Disappear First
The COVID-19 disruption provided an unintended natural experiment. Schools closed from March to October 2020. The 2020 KCSE was pushed to early 2021. The 2021 KCSE sat on a compressed calendar in late 2021 and early 2022.
In 2021, for the first time in the panel, male candidates (418,435) outnumbered female candidates (403,147). The usual pattern reversed. Seven months out of school cost more girls than boys. The evidence from across Sub-Saharan Africa points to the mechanisms: early marriage, adolescent pregnancy, and increased domestic labour burdens that fall disproportionately on girls during extended closures.
By 2022, the female numerical advantage recovered. But the 2021 reversal carries a direct policy lesson. Extended school closures are not gender neutral. The system had no mitigation plan for the predictable, gendered attrition that closures produce. If Kenya faces another extended disruption, whether from pandemic, conflict, or climate events, the response must include gender-differentiated retention measures from the first day of closure. Waiting for the data to confirm what the evidence already predicts will cost another cohort of girls.
Poverty remains the deepest driver. A child from a poor household is more than twice as likely to drop out as a child from a wealthy one, regardless of gender (FLANA, 2026)1. But poverty interacts with gender differently for boys and girls. Boys are pulled out to work. Girls are pulled out to marry, to care for families, or because the household cannot afford to keep all children in school and chooses sons. Any mitigation strategy that treats dropout as a single phenomenon will miss both.
The Age Problem Nobody Talks About
A student who enters Grade 1 at age six and progresses without repetition should sit the KCSE at 17 or 18. In 2025, the largest age group was not 17 to 18. It was 19 to 20. The modal KCSE candidate is already one to two years behind.
This is not randomly distributed. North-eastern counties show overage rates exceeding 30 percent. Counties in Central Kenya and Nairobi county show the lowest. The geography of overage maps almost exactly onto the geography of low Gender Parity Index and the geography of Sub-County school concentration. A county where girls are underrepresented is also a county where students are overage and where schools are under-resourced. These are not three separate policy problems. They are one structural pattern, and the students caught in the overlap experience all of them simultaneously.
The Question We Cannot Yet Answer
Across all eight cohorts, boys consistently dominate both ends of the KCSE performance distribution.
At the top, this is partly explained by subject weighting. Mathematics and physical sciences carry significant influence on overall grades, and boys tend to outperform girls in these areas.
At the bottom, higher rates of disengagement, absenteeism, and near-dropout behavior among boys may play a role.
Yet these explanations remain incomplete. They describe the pattern without fully explaining why boys exhibit greater variability in outcomes.
International studies suggest similar trends in high-stakes examinations, but no definitive answer exists. This gap matters.
While the gender gap in university entry has narrowed from 4.5 percentage points in 2018 to 2.5 in 2025, closing it further requires addressing both ends of the distribution.
A policy that lifts girls into top performance must also address why so many boys remain concentrated at the bottom.
