1# Contract — F2-conf-012**Question:** effect of a school-level programme on student test scores, with design-valid uncertainty; decide at 5% (two-sided) whether τ≠0.3**Estimand:** τ = student-weighted mean score (treated schools) − (control schools) = OLS slope of score on treated with intercept. Meaningful as a causal effect if assignment of schools was as-good-as-random (simulated data; assumed).4**Uncertainty:** treatment varies only between schools; students within a school share school shocks; schools independent → cluster-robust SE at school level (main), iid OLS SE as naive baseline.5**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 (6 sections), out/analysis.py, out/claims.json.6**Information rules:** only data/schools.csv.7**Invalidity:** treatment varying within school, too few clusters, missing data, using iid SE for the decision.8