Supply Predicts Finishing Better than It Predicts Starting: School Provision and Educational Outcomes

Abstract

Where upper secondary schools are scarce relative to primary schools, fewer young people finish upper secondary — and the association is considerably firmer than the association between the same supply measure and whether they are enrolled. Using Central Statistics Agency figures for 2025 and 2023 across the Indonesian provinces, this article relates the ratio of villages with a primary school to villages with an upper secondary school to two outcomes. Against upper secondary completion the Pearson correlation is −0.5189 and the Spearman −0.4882, a gap of 0.0307 indicating a relationship that holds throughout the data. Against participation at ages sixteen to eighteen the Pearson correlation is −0.6039 and the Spearman −0.3791, a gap of 0.2248; removing two provinces takes the Pearson value to −0.4419. The tertiary supply measure behaves differently again: villages with a higher education institution per hundred with a primary school correlate with upper secondary completion at 0.5343 on Pearson and 0.6391 on Spearman, the rank coefficient exceeding the product-moment one, which indicates a monotone but curved relationship rather than an outlier artefact. The article argues that supply measures should be validated against completion rather than enrolment, and that the direction of the Pearson-Spearman divergence distinguishes two quite different problems.