Project · Beginner
School attendance and dropout early warning
An early-warning analysis that flags students at risk of dropping out from attendance patterns, in time for a school year to still act on it.
PythonRKoboToolboxSustainable Development Goals (SDG)
Context
- Organisation
- Education programme covering 24 schools
- Problem
- Dropout was only measured at year end, by which point the students had already left and the programme could report the number but never change it.
- Audience
- Education programme officer and head teachers
- Decision it informs
- Which students receive a home visit before the end of term.
Datasets used
Deliverables
- Analysis notebook
notebooks/dropout-early-warning.ipynb - Dashboard
dashboard/index.html - Technical documentation
docs/feature-definitions.md - Final report
reports/term-report.pdf - Presentation slides
slides/head-teacher-briefing.pdf
Measuring something while you can still change it
Year-end dropout counts are a report. Consecutive-absence flags are an intervention.