Back to Portfolio

Education Outcomes

How did bicycle access affect school attendance, academic performance, and retention across 346 students over 6 years?

📍 Hwange District, Zimbabwe 📅 2019–2024 (longitudinal panel) 👥 346 students · 17 wards 🔒 Data anonymised
-71%
Absenteeism reduction
12.6 → 3.6 days/year (2019–2024)
+31%
Academic improvement
Overall average: 38% → 50%
p < .001
Pre–post attendance test
Welch's t-test, two-sided
346
Students tracked
Longitudinal panel, 6 academic years

Attendance trend

Mean days absent per academic year (lower = better)

Academic performance

Overall average mark (%) across all subjects

Pre–post bicycle comparison

Mean days absent before bicycle distribution (2019–2020) vs after (2021–2024)

Pre-bicycle (2019–20)
9.9
mean days absent / year
Post-bicycle (2021–24)
4.4
mean days absent / year
-5.5 days · 56% reduction · p < .001

What students and teachers said

"I managed to complete my school just because of the bicycle. The bicycle made us to be at school always... If I had no bicycle, it would be easy for me to hide in the bushes and abscond lessons."
FGD Out-of-School Alumni · Retention & attendance
"I used to take two hours walking to school... now the same distance takes only 45 minutes."
FGD Student · Travel time reduction
"The fatigue of walking 14 kilometres every day made some students miss lessons... because of the bicycles, they are not missing out."
KII School Head · Attendance & fatigue
"Before the bicycles... I used to find lessons have commenced... After: both my attendance and pass rate has improved."
FGD Student · Punctuality & performance
"The issue of dropouts... has been reduced since most of the children have the chance of coming to school with the use of those bicycles."
KII District Education Official · Dropout reduction
"Some learners who had already dropped out started coming to school... it improved even the issue of truancy."
KII School Head · Re-engagement

Go deeper into the evaluation

This dashboard shows the headline findings. Explore the rigorous methods behind them.

Difference-in-Difference Impact Analysis Qualitative Evidence Synthesis R Analysis Pipeline
Methodology note: Student data comes from a longitudinal panel of 346 students tracked from 2019–2024 across 17 wards in Hwange District. Attendance records are annual days absent from school records; academic performance is the overall average mark (%). Pre–post comparison uses Welch's two-sample t-test (unequal variance). The Difference-in-Difference analysis exploits the phased rollout to estimate causal impact. Qualitative findings are from focus group discussions (FGDs) and key informant interviews (KIIs); see the full qualitative synthesis. All data are anonymised for public display. See full R pipeline for code and reproducibility details.