Course · Intermediate · Sector Analysis
Education Programme Analysis
Enrolment, attendance, retention and learning outcomes, with the school-year calendar and the enrolment-versus-attendance distinction that trips up every dashboard.
What you will be able to do
- Compute gross and net enrolment ratios and explain why one of them exceeds 100% without any child being counted twice
- Separate average daily attendance from the proportion of students regularly attending, and say which a programme is judged on
- Follow a cohort through promotion, repetition and dropout, and compute survival to the final grade
- Read a learning assessment across two rounds whose instruments differ, and say what is comparable
- Detect a non-random assessment subsample and bound the bias it introduces
- Disaggregate by sex, disability and displacement, and state the sample each cut demands
Standards and methodologies
Education is the sector where the most-quoted indicator is the one most often computed against the wrong denominator, and where the distinction a programme lives or dies by — enrolled against attending against learning — is collapsed on almost every dashboard into a single number called coverage.
The course runs on four files. The attendance register and its roster have appeared in three earlier courses as a cleaning and joining problem; here they are analysed as attendance. Three are new: a two-year enrolment register with age, sex, disability, displacement and an end-of-year status; the school-age population that is its denominator; and a two-round learning assessment.
Three results shape the course, all computed from those files.
Gross enrolment is 109.2% and net enrolment is 97.4%. Neither is wrong and neither is the other. 36.4% of primary enrolment is above the official age for its grade and one child in nine is twelve or older while still in primary — which is what puts the gross ratio over 100% without a single child being counted twice.
Average daily attendance is 88.4% and 62.1% of students attend at least 90% of the days they were marked. A programme reporting the first is describing a system; a programme acting on the second is describing children.
Dropout is 19.5% among over-age students against 4.8% among the rest. Being behind is the strongest predictor in this register, and it compounds: 94.5% of the children who repeated a grade in 2023 are over-age in 2024, and an over-age child is four times as likely to leave.
Both Python and R throughout. You should have done module 3 — this course assumes you can defend a denominator, read a censored register and put an interval on a proportion.
Practice
Reading the lessons is not the same as having done the work. Each of these applies the course to a dataset it did not teach on.
- Lab · 180 minEnrolled, attending, learningBuild one indicator from each of the three populations, on four files and six denominators, and then write the paragraph explaining why the three cannot be drawn as a funnel.
- Exercise · 60 minThe two schools at the bottomA dashboard ranks twenty-four schools on attendance and two sit twenty points below the rest. A dropout response is being planned for them. Find out what actually happened before it is funded.
Progress
Take it offline
The whole course as a typeset PDF — every lesson, every code example, the data dictionary and the indicator definitions. Generated from the same source as this page.
The LaTeX source ships alongside each PDF, so an organisation can rebrand the handout or fold a lesson into its own training pack.