# Contract (P0)

Question: average effect of a school-level programme on student test score, with design-valid uncertainty, and 5% two-sided test of tau=0. Estimand: tau = student-weighted mean score (treated schools) − (control schools) = OLS slope of score on treated with intercept. Meaningful as a causal effect under (assumed) random assignment of schools; as a descriptive difference otherwise. Uncertainty: treatment varies only between schools; students within school are correlated; schools independent -> cluster-robust SE at school level (CR1), inference with t(G-1) reference. Naive iid OLS SE reported as baseline only. Outputs: out/results.json (task_id, tau_hat, se_iid, se_cluster, p_value, tau_ci_low, tau_ci_high, effect_significant_5pct, icc_hat, n_schools, n_treated_schools, n_analysed, feasibility, feasibility_reason, data_issues, n_input_rows, random_seed), out/report.md (Question; Data and exclusions; Method; Results; Uncertainty; Limitations and open issues), out/analysis.py, out/claims.json (incl. claim on 5% significance). Information rules: only data/schools.csv. Invalidity: treatment not constant within school, too few clusters, missing/duplicated data, non-reproducible output, using iid SE for the decision.