Education Crisis in Kenya: From Policy to School Safety

Both analyses use a unified analytical pipeline combining spreadsheet preparation, dashboarding and statistical modelling.

Data Cleaning & Preprocessing
Data cleaning and descriptive statistics performed in Google Sheets.

Exploratory Analysis
Exploratory analysis and descriptive statistics to identify unrest and readiness patterns.

Dashboard Visualization
Power BI dashboards mapping school unrest incidents, root causes and geographic risk distribution.

Statistical Modelling
Python-based correlation analysis, regression modelling and clustering of counties, supported by a Composite Risk Score (0–100) built from infrastructure quality, fire safety compliance and dormitory overcrowding.

High-Risk Counties

Across the 10 counties tracked, unrest affected 68 schools and caused 16 fatalities — all in Nakuru. Narok (composite risk score 98/100), Nakuru (95) and Taita Taveta (95.2) form a distinct high-risk tier well above the rest.

Infrastructure & Safety

Risk splits cleanly along infrastructure grade: the 5 counties graded "D" with "Low" fire safety compliance average a composite risk score of 95.3, versus 27.2 for the 5 counties graded "C" with "Medium" compliance — a complete overlap between weak infrastructure and high risk.

Overcrowding

Dormitory overcrowding ranges from 50% to 90% (average 69%) and correlates strongly with composite risk (r = 0.94) — making it the single strongest driver of unrest risk in the dataset.

CBC Readiness Link

Schools with low CBC readiness often exhibit higher safety risks, and infrastructure quality is a shared driver across both readiness and unrest — the same "Grade D" counties turn up in both the CBC readiness gaps and the school unrest risk tier.

Concentration of Risk

Risk is evenly split at the county level — 5 of 10 counties fall in the "High Risk" category and 5 in "Low Risk" — but the divide is stark: high-risk counties average 95.3 on the composite score versus 27.2 for low-risk counties, with almost no middle ground.

1. Urgent Fire Safety Audits

Conduct urgent fire safety audits in high-risk schools.

2. Reduce Overcrowding

Reduce dormitory overcrowding to safe capacity levels, addressing it as a dual risk factor for both safety and CBC implementation.

3. Improve Infrastructure

Invest in infrastructure improvements in low-readiness, high-risk schools.

4. Integrate Monitoring Systems

Integrate CBC readiness and safety monitoring into one national system, and implement a national monitoring system for school safety trends.

5. Prioritize Overlapping High-Risk Counties

Prioritize interventions in counties that overlap across both readiness gaps and safety risks, and strengthen education policy implementation tracking.

The analysis shows that CBC readiness and school safety are not separate challenges but interconnected issues within Kenya's education system. Addressing both together provides a more effective framework for improving education outcomes and student safety.

Limitations & Next Steps: This analysis uses a single-year snapshot across 10 tracked counties. With more time, I'd bring in year-over-year data to test whether the overcrowding–unrest correlation is stable or worsening, and extend the composite risk score to all 47 counties for full national coverage.