# Contract — F4-conf-01

Question. Extract, exactly as printed, specified treatment-effect estimates and their uncertainty from two arXiv extracts (Armendariz 2026 DCR staggered DiD, Tables 6/7; Li et al. 2026 RESEA RCT, Table 2), attach locators, derive (a) a 95% CI for the Table 6 5 km ATT from its printed SE and (b) an SE for the Table 2 wage ATE from its printed 95% CI, with a stated convention, and recognise unreported quantities (e.g. exact p-value of HIV 5 km ATT; SE of wage ATE). Estimands. The printed numbers themselves; derived quantities assume approximate normality of the estimator (Wald-type CI, critical value 1.96) and, for the wage SE, that the printed CI is symmetric Wald-type. Outputs. out/results.json (keys listed in TASK.md incl. locators, derivation_notes, task_id, feasibility, feasibility_reason, data_issues, n_input_rows, n_analysed, random_seed=null), out/report.md (6 sections), out/analysis.py (regenerates results from data/ and prints matched lines), out/claims.json, plus state files and out/exclusions.csv. Information rules. Only the two extracts are evidence; no memory; values as printed from the table the key names (Table 6 not A19 for the 5 km ATT SE); null + "not reported" locator for unprinted quantities. Invalidity. Using values from wrong table, rounding/rescaling printed values, fabricating p-values, unstated derivation conventions.