1# WP21 — Extraction audit (generated)23Studies extracted: 297; eligible: 242; ineligible at full text: 55.4Coefficient rows: 18831; usable in the pooled synthesis: 7693 from 149 studies.56## Confidence of extracted rows78| confidence | rows |9|---|---|10| high | 14776 |11| medium | 3941 |12| low | 114 |1314## Source of the standard error1516| source | rows |17|---|---|18| reported | 10227 |19| none | 3901 |20| |b/t| | 3794 |21| |b/z(p)| (two-sided normal) | 710 |22| (ci_hi-ci_lo)/(2*1.96) | 199 |2324## Harmonisation flags (rows excluded from pooling)2526| flag | rows |27|---|---|28| none | 13535 |29| nonstandard_x_transform | 2555 |30| unit_not_harmonisable | 2351 |31| year_built_not_age | 226 |32| implausible_magnitude | 110 |33| year_built_not_age;nonstandard_x_transform | 54 |3435## Full-text exclusion reasons (first 80 characters)3637| reason | studies |38|---|---|39| Aggregate unit of analysis: the regression (Tables 4-5, p. 9) is a monthly time- | 1 |40| Aggregate unit of analysis: dependent variable is the (log) mean price per squar | 1 |41| No hedonic coefficient table. The paper estimates a generalized propensity score | 1 |42| Aggregate unit of analysis: cross-section of 985 Italian railway station areas; | 1 |43| Aggregate unit of analysis: regressions use Seoul's 25 autonomous districts (Gu) | 1 |44| Not a hedonic regression study: the paper proposes a theoretical 'dynamic impact | 1 |45| Hedonic model (Table 5, p.11) reports only the appraisal-anchoring variables (ln | 1 |46| Conference essay reviewing anchoring effects; the only estimates (Tables 1-3, p. | 1 |47| Aggregate unit of analysis: panel of 30 Chinese cities x 10 years (2006-2015) re | 1 |48| Regression tables (Tables 4 and 6, p. 14) regress price per m2 on three expert-w | 1 |49| Aggregate unit of analysis: 1498 residential price samples averaged into 653 sam | 1 |50| Aggregate unit of analysis: the regression explains the MEDIAN house value of ce | 1 |51| Not an empirical housing-market study: the author states 'the dataset used was s | 1 |52| No regression coefficients are printed: the paper compares predictive performanc | 1 |53| No hedonic regression coefficients on property characteristics are printed. The | 1 |5455## Independent double-extraction validation5657```json58{59 "papers_double_extracted": 20,60 "eligibility_agreement": 0.95,61 "coef_recall": 0.788720246856144,62 "coef_precision": 0.7731166520957543,63 "beta_agreement_1pct": 1.0,64 "se_agreement_1pct": 1.0,65 "sign_agreement": 1.0,66 "taxonomy_agreement": 1.0,67 "coefficients_compared": 64768}69```7071Interpretation: numerical agreement on matched coefficients is the relevant accuracy measure; recall/precision below one reflect72differences in which auxiliary variables (composite indices, planning dummies, local-regression summaries) each extractor recorded.73Eligibility disagreements are listed in data/validated/validation_results.csv and were resolved by the protocol rule (duplicate-of-prior-publication → excluded).