Better Together? Heterogeneous Effects of Tracking on Student Achievement

German federal-state comparison; students observed before and after the tracking transition

German lower-secondary students in states that track into separate school types at ~age 10 versus states that keep the non-academic-track student body comprehensive for two more years.

Triple-differences exploiting institutional differences between German federal states. In every state ~40% of students move to academic-track schools after primary; the remaining students are either tracked immediately into two further school types or taught comprehensively for two more years. Identification is design-based (state institutions), not student selection.

Prolonging comprehensive schooling by two years raises mathematics and reading scores, with the whole average effect coming from low achievers and nothing lost by high achievers. Early and rigid tracking therefore harms both equity and efficiency — the strongest single causal result against early between-school tracking.

Between-state institutional variation rather than selection into tracks, so the estimate is not driven by heritable differences between tracked and untracked children.

Consistent with the international diff-in-diff (Hanushek & Woessmann 2006) and with the Nordic and French detracking reforms.