Under development — community extension
A community effort to extend and potentially refine the analysis. This repository is a collective initiative to bring these reanalyses together in one place. The extension is separate from Peter Hull’s original 22-paper collection, which follows the criteria established in his twfe-replication repository.
Contributors and shared data. Pedro H. C. Sant’Anna and Victor Rangel shared additional data. Apoorva Lal, Avi Feller and Peter Hull have contributed through refinements, data, or both. We expect to add more papers over time and welcome further papers, data and refinements from other contributors.
Work in progress. Source preparation, methods, results and the eligibility log are under development. The extension criteria and study assessments are not fully vetted and may change with further review; they are not Peter Hull’s established original-selection criteria.
240 extension papers assessed · 46 currently included by the provisional screen · 36 with comparison summaries · 149 plotted comparisons.
The current working screen focuses on observational binary-treatment DiD analyses. IV second stages, randomized-assignment effects and non-binary treatment targets need different methods; their presence elsewhere in a paper does not rule out a separate observational binary analysis. New candidates require specification mapping and validation before entering extension calibration plots. Data, support or numerical limitations are separate from design evidence.
Extension comparisons: JSON · CSV · Code · Editable template
Full study inventory: CSV · Google Sheet (verified 2026-10-13) · Per-variable treatment counts
Design support by paper: CSV · Compatible DiD papers blocked: CSV · Specification-level estimator gaps: CSV
PDF audit correction: Page-cited findings · 161-row audit register. The old 161-paper exclusion total was not a full-paper PDF audit. In that October 12 snapshot, 36 targeted PDF identification reviews reopened 15 cases; 125 old exclusions were not PDF-reviewed. The October 13 Deep DiD follow-up below supersedes its administrative counts. The remaining 21 reviews corroborate selected-target restrictions, not exhaustive whole-paper exclusions.
Deep DiD community follow-up: Independent checks and reconciliation · 245-paper CSV. Pedro H. C. Sant’Anna’s contributed audit identifies 122 observational binary temporal contrasts (87 absorbing-pattern candidates). We checked all 26 challenges to exclusions and reopened them as named candidates. These are not 26 new fits; the original collection and calibration data remain unchanged.
Author-code TWFE evidence: Scoring guide and findings · 262-paper code-evidence register · Score counts. Fixed-effect structure is scored separately from DiD eligibility and replication success. Positive evidence includes IV, nonlinear, cohort and auxiliary specifications; missing matches are not whole-paper exclusions.
Template support, not full-paper replication. Across the 262-paper inventory, the provisional screen identifies 64 binary-DiD studies: 54 have comparison results and 10 have documented preparation, functional-form, sample, inference, computational or numerical blockers. After the PDF and Deep DiD follow-ups, 120 remain provisionally excluded and 78 are pending. These are administrative screen counts, not verified counts of papers without a usable DiD analysis. Unsupported IV, RD/fuzzy-RD, randomized or non-binary targets are not labeled failed source replications. The spreadsheet separates these categories and records partial estimator failures separately.
The inventory’s ‘Main collection’ column records the existing design-screen decision across both sections, not assignment to the original-collection tab.
Archive assessment coverage
All 245 papers in the supplied archive now have a source-design assessment. This does not mean that all 245 have been replicated. Together with 17 additional original studies, the inventory contains 262 distinct papers. Five archive papers overlap the original collection; the remaining 240 belong to this development section.
The latest batches cover the final 50 + 39 papers. They added no calibration points. At that release, 88 were outside the provisional template scope and one was pending; the subsequent Deep DiD follow-up reopens named targets and supersedes those historical screening counts. Existing estimates and all seven calibration panels in each section are unchanged. Treatment cardinalities are the previously measured counts, not new estimates from this audit.
Next 50: assessment register · Final 39: assessment register
Pending example: Gelber’s 401(k) eligibility study uses a second-difference specification with a possible DiD interpretation. Eligibility and time-path reconstruction need review before it can enter the adoption-panel template; this is not a confirmed design exclusion.
Individual assessment reports, source references and reasons for all remaining cases are linked in the provisional eligibility log below. Papers can be reconsidered as the template and community review develop.
Extension studies currently included
| Study | Status | Report |
|---|---|---|
| Staggered Difference-in-Differences in Gravity Settings: Revisiting the Effects of Trade Agreements | Comparison unavailable | Report |
| Wall Street and the Housing Bubble | 7 comparison summaries | Report |
| Provider Supply, Utilization, and Infant Health: Evidence from a Physician Distribution Policy | 2 comparison summaries | Report |
| Upstream and Downstream Impacts of College Merit-Based Financial Aid for Low-Income Students: Ser Pilo Paga in Colombia | 2 comparison summaries | Report |
| Taxation and International Migration of Superstars: Evidence from the European Football Market | 7 comparison summaries | Report |
| Personal Retirement Accounts and Saving | 4 comparison summaries | Report |
| The Impact of Insurance Provision on Household Production and Financial Decisions | 4 comparison summaries | Report |
| Stimulating the Vote: ARRA Road Spending and Vote Share | 9 comparison summaries | Report |
| Environmental Regulations, Air and Water Pollution, and Infant Mortality in India | Comparison unavailable | Report |
| The Labor Market Impacts of Universal and Permanent Cash Transfers: Evidence from the Alaska Permanent Fund | 7 comparison summaries | Report |
| Coase and Cap-and-Trade: Evidence on the Independence Property from the European Carbon Market | 9 comparison summaries | Report |
| When Does Product Liability Risk Chill Innovation? Evidence from Medical Implants | 8 comparison summaries | Report |
| Communication Infrastructure and Stabilizing Food Prices: Evidence from the Telegraph Network in China | Comparison unavailable | Report |
| The Effects of a Large-Scale Mental Health Reform: Evidence from Brazil | Comparison unavailable | Report |
| Of Mice and Academics: Examining the Effect of Openness on Innovation | Comparison unavailable | Report |
| Compulsory Licensing: Evidence from the Trading with the Enemy Act | 4 comparison summaries | Report |
| Giving Mom a Break: The Impact of Higher EITC Payments on Maternal Health | 7 comparison summaries | Report |
| Awarding Price, Contract Performance, and Bids Screening: Evidence from Procurement Auctions | 7 comparison summaries | Report |
| How Punishment Severity Affects Jury Verdicts: Evidence from Two Natural Experiments | 2 comparison summaries | Report |
| Family Formation and Crime | Comparison unavailable | Report |
| Trickle-Down Ethnic Politics: Drunk and Absent in the Kenya Police Force (1957-1970) | 4 comparison summaries | Report |
| Incentive Schemes for Local Government: Theory and Evidence from Comprehensive Performance Assessment in England | 2 comparison summaries | Report |
| There Will Be Killing: Collectivization and Death of Draft Animals | 8 comparison summaries | Report |
| The Economics of Rights: Does the Right to Counsel Increase Crime? | 4 comparison summaries | Report |
| The Economics of Nationalism | 4 comparison summaries | Report |
| Acquisitions, Productivity, and Profitability: Evidence from the Japanese Cotton Spinning Industry | 2 comparison summaries | Report |
| Corporate Incentives and Nuclear Safety | Comparison unavailable | Report |
| The Power of Political Voice: Women’s Political Representation and Crime in India | 7 comparison summaries | Report |
| Indirect Savings from Public Procurement Centralization | Comparison unavailable | Report |
| Rebel on the Canal: Disrupted Trade Access and Social Conflict in China, 1650-1911 | 8 comparison summaries | Report |
| The Economics of Speed: The Electrification of the Streetcar System and the Decline of Mom-and-Pop Stores in Boston, 1885-1905 | 2 comparison summaries | Report |
| The Costs of Patronage: Evidence from the British Empire | 2 comparison summaries | Report |
| Do College Graduates Serving as Village Officials Help Rural China? | 5 comparison summaries | Report |
| The Berlin Stock Exchange in Imperial Germany: A Market for New Technology? | 2 comparison summaries | Report |
| Water Pollution Progress at Borders: The Role of Changes in China’s Political Promotion Incentives | Comparison unavailable | Report |
| Does Abolishing User Fees Lead to Improved Health Status? Evidence from Post-Apartheid South Africa | 2 comparison summaries | Report |
| Only One Tree from Each Seed? Environmental Effectiveness and Poverty Alleviation in Mexico’s Payments for Ecosystem Services Program | 2 comparison summaries | Report |
| Buffalo Hunt: International Trade and the Virtual Extinction of the North American Bison | 7 comparison summaries | Report |
| Human Capital Persistence and Development | 4 comparison summaries | Report |
| The War on Poverty’s Experiment in Public Medicine: Community Health Centers and the Mortality of Older Americans | 2 comparison summaries | Report |
| Regulation, Ownership, and Costs: A Historical Perspective from Indian Railways | 2 comparison summaries | Report |
| Deregulation, Consolidation, and Efficiency: Evidence from US Nuclear Power | Comparison unavailable | Report |
| Reexamining the Impact of Family Planning Programs on US Fertility: Evidence from the War on Poverty and the Early Years of Title X | 2 comparison summaries | Report |
| Hometown Ties and the Quality of Government Monitoring: Evidence from Rotation of Chinese Auditors | 2 comparison summaries | Report |
| A Letter and Encouragement: Does Information Increase Postsecondary Enrollment of UI Recipients? | 2 comparison summaries | Report |
| Distributive Politics and Electoral Incentives: Evidence from Seven US State Legislatures | 2 comparison summaries | Report |
Calibration comparisons
Both axes use the matching TWFE standard error: x = TWFE / SE(TWFE), y = alternative estimate / SE(TWFE). Colors identify studies; circles denote static specifications and squares denote event-study averages. Capped points are marked at the boundary. Use the range selector or hover for actual coordinates and specification details.
BJS and Sun–Abraham use outline rings for cohort-homogeneity rejection at 5%, from the homoskedastic deviations F-test without multiplicity adjustment. Missing tests are not non-rejections. CS-DID has separate equal-event-time and treated-cell summaries. Multisynth results remain in paper reports and downloads, not separate calibration panels.
Sign reversals: 8.3%. Homogeneity rejected: 6/13 available tests. Downloads use the capped view; hover shows actual values.
Sign reversals: 0.0%. Homogeneity rejected: 6/12 available tests. Downloads use the capped view; hover shows actual values.
Sign reversals: 10.5%. No homogeneity-test annotations. Aggregation and donor information are in the tooltip. Downloads use the capped view; hover shows actual values.
Sign reversals: 5.3%. No homogeneity-test annotations. Aggregation and donor information are in the tooltip. Downloads use the capped view; hover shows actual values.
Sign reversals: 0.0%. No homogeneity-test annotations. Aggregation and donor information are in the tooltip. Downloads use the capped view; hover shows actual values.
Sign reversals: 6.7%. No homogeneity-test annotations. Aggregation and donor information are in the tooltip. Downloads use the capped view; hover shows actual values.
Sign reversals: 23.1%. No homogeneity-test annotations. Aggregation and donor information are in the tooltip. Downloads use the capped view; hover shows actual values.
Combined research downloads
For cross-collection analysis: all original + currently screened extension results: JSON · CSV. An optional provisional design-screen subset: JSON · CSV applies the working screen to both rosters. Neither combined download is the main-page calibration sample.
Provisional eligibility log
Under development: the criteria and classifications below are not fully vetted and may change. This is the community extension’s working log, not Peter Hull’s original selection criteria. It records current assessments for review, rather than final judgments about the papers.
The original 22-paper collection and its results are preserved in full. Working design notes about original papers are kept here, not used to remove or relabel papers on the main page. For extensions, the current screen determines which results enter the development plots pending further review.
Pending means the study-design decision is unresolved, not that the paper is necessarily ineligible. Computationally unsupported and design-ineligible are different categories. In particular, random sampling of audits is not automatically randomized assignment of the treatment being studied.
Working eligibility assessments · Membership totals · Historical, unfiltered estimates: JSON · CSV
Working notes on original papers
The following provisional classifications are open to review. All four papers remain in Peter Hull’s original collection with their available comparison results:
- Do Government Audits Reduce Corruption?: Municipalities are randomly selected for government audits. Pre/post regressions do not turn randomized assignment into observational DiD.
- Leadership in Social Movements: Deep DiD follow-up: Table 4 columns 2/4 regress local German newspapers / clubs on fourty8ersD with town and state-year effects; the main IV target is different. This supplies a named candidate for the development audit. All original study-10 reports, estimates and original membership remain untouched.
- The Rise of Fiscal Capacity: The paper includes an IV design using Imperial tax contributions to instrument fiscal centralization; excluded under the study-level IV screen even though the archive selected OLS/event-study specifications.
- The Real State: The traffic-police experiment uses randomized encouragement and IV. It is not an observational binary-treatment DiD study.
Extension studies currently outside the working screen
| Study | Reason | Evidence / archive |
|---|---|---|
| Wetlands, Flooding, and the Clean Water Act | nonbinary treatment: The flooding regressions use continuous wetland changes and related exposure terms in panel/long-difference specifications. Wetland acreage changes are not a binary treatment. | Source 1 · Historical report |
| Standard Setting Committees: Consensus Governance for Shared Technology Platforms | not did: The source regresses standards-development duration on committee variables and covariates in an id-level cross-section, clustered by working group. It does not supply an adoption panel for the focal coefficient. | Source 1 · Historical report |
| Gender Quotas and the Crisis of the Mediocre Man: Theory and Evidence from Sweden | nonbinary treatment: The deposited Table 4 Stata regressions use qb_post_quota, quota bite multiplied by the post-quota indicator. Quota bite is continuous. Replacing it with a treated/untreated indicator would change the source specification. | Source 1 · Historical report |
| Indirect Effects of Access to Finance | randomized trial: Firm access to finance is randomly assigned; interpost estimates its direct pre/post effect and inter4post captures continuous competitor spillovers. The pre/post form does not make this observational DiD. | Source 1 · Historical report |
| The Economic Impact of Depression Treatment in India: Evidence from Community-Based Provision of Pharmacotherapy | randomized trial: Depression pharmacotherapy is randomized across four arms. Treatment-by-survey-round ANCOVA effects remain experimental; no artificial adoption panel is constructed. | Source 1 · Historical report |
| Experimental Evidence on the Long-Run Impact of Community-Based Monitoring | randomized trial: Community monitoring is randomly assigned. The selected mortality regression compares treatment and control with district effects; it remains an experimental contrast. | Source 1 · Historical report |
| College Enrollment and Mandatory FAFSA Applications: Evidence from Louisiana | nonbinary treatment: The source trt is (1 minus baseline FAFSA completion) multiplied by the post-2017 indicator. The Louisiana policy applies statewide; the design identifies a continuous exposure gradient across schools. | Source 1 · Historical report |
| Quasi-Experimental Evidence on the Connection between Property Taxes and Residential Capital Investment | nonbinary treatment: The source treatment netg_taxcom98 measures the size of a municipal fiscal shock after school-finance reform. Both the scalar and year-interacted specifications are continuous-treatment designs, not binary adoption. | Source 1 · Historical report |
| From Extreme to Mainstream: The Erosion of Social Norms | randomized trial: The public-visibility by Trump-won-information contrast is a randomized 2-by-2 survey experiment. A factorial interaction is not an observational before/after DiD. | Source 1 · Historical report |
| Do Women Give Up Competing More Easily? Evidence from the Lab and the Dutch Math Olympiad | randomized trial: The paper combines laboratory experiments and an Olympiad cutoff design. Gender/session contrasts and randomized feedback do not supply an observational treatment-adoption panel. | Source 1 · Historical report |
| From Final Goods to Inputs: The Protectionist Effect of Rules of Origin | nonbinary treatment: The source is a triple long difference in import changes with continuous rules-of-origin intensity lroo3_treatedBOTH. A binary adoption regression would discard both its exposure scale and third difference. | Source 1 · Historical report |
| Targeting High Ability Entrepreneurs Using Community Information: Mechanism Design in the Field | randomized trial: Capital grants are assigned by lottery; the headline interacts winning with peer-ranked ability. The randomized treatment and heterogeneity interaction are not observational adoption. | Source 1 · Historical report |
| Cross-Region Transfer Multipliers in a Monetary Union: Evidence from Social Security and Stimulus Payments | nonbinary treatment: The source doneoff2000REW measures temporary transfer payments scaled by labor income, with repeated stimulus episodes. The coefficient is a continuous transfer multiplier, not a binary-policy ATT. | Source 1 · Historical report |
| Geographic Cross-Sectional Fiscal Spending Multipliers: What Have We Learned? | instrumental variables: Geographic spending multipliers instrument stimulus outlays with pre-recession allocation measures. These cross-state IV/long-difference coefficients are not binary-treatment DiD effects. | Source 1 · Historical report |
| The Neighborhood Impacts of Local Infrastructure Investment: Evidence from Urban Mexico | randomized trial: Habitat infrastructure is randomly assigned to urban polygons. The two-period polygon fixed-effects DiD estimates an experimental ITT, excluded by the current observational-extension rule. | Source 1 · Historical report |
| Early Retirement Incentives and Student Achievement | nonbinary treatment: Table 5 uses post interacted with the number of teachers above the retirement-experience threshold (post_above, with subject-specific variants), plus post interacted with total teacher counts. This is a treatment-intensity design, not binary school adoption. Converting teacher counts to an indicator would change the original estimand. | Source 1 · Historical report |
| Adverse and Advantageous Selection in the Laboratory | randomized trial: The laboratory study compares human/human, human/robot and feedback treatment arms. Its selection-game outcomes are experimental contrasts, not policy-adoption DiD effects. | Source 1 · Historical report |
| Professors in Core Science Fields Are Not Always Biased against Women: Evidence from France | not did: The source compares examiner discipline and candidate gender within exams, rather than treatment adoption over a panel clock. | Source 1 · Historical report |
| The Debt-Inflation Channel of the German (Hyper)Inflation | nonbinary treatment: The source Treat1917xPost_L_A is 1917 firm leverage multiplied by the post-inflation indicator. The event study also interacts year with leverage; neither is a binary treatment specification. | Source 1 · Historical report |
| The Dynamic Response of Municipal Budgets to Revenue Shocks | nonbinary treatment: The source event study uses the continuous 1987 census revenue shock interacted with event time, with optimal analytic weights. Its per-unit-shock effect cannot be replaced by a binary treated indicator. | Source 1 · Historical report |
| Expected Returns to Crime and Crime Location | nonbinary treatment: BIPhypgoldpriceOUTLA interacts high-South-Asian exposure with a continuously varying gold-price index. The binary exposure component alone is not the treatment; the source target is continuous-exposure DDD/TWFE. | Source 1 · Historical report |
| Aid, China, and Growth: Evidence from a New Global Development Finance Dataset | instrumental variables: The aid-growth design instruments lagged Chinese official-finance exposure with historical recipient propensities and global supply shifters. It is panel IV with continuous exposure, not binary adoption. | Source 1 · Historical report |
| Early Voting Laws, Voter Turnout, and Partisan Vote Composition: Evidence from Ohio | nonbinary treatment: The source measures turnout responses to the number of early-voting days (and changes in days in the border design). Treatment intensity is multivalued, not binary adoption; the headline also requires county-and-border clustering. | Source 1 · Historical report |
| Media Bias in China | nonbinary treatment: The source reform_cg2003 is the prefecture’s 2002 stock of party-county-daily papers interacted with post-2003; a second interaction distinguishes commercial papers. This is a continuous-intensity reform specification, not a binary policy ATT. | Source 1 · Historical report |
| Immigration Restrictions as Active Labor Market Policy: Evidence from the Mexican Bracero Exclusion | nonbinary treatment: The source treatment_frac equals the post-1965 indicator times the state’s 1955 Mexican seasonal-worker share. The original wage coefficient is a continuous-exposure effect; it is not recoded as any Bracero exposure. | Source 1 · Historical report |
| The Returns to Microenterprise Support among the Ultrapoor: A Field Experiment in Postwar Uganda | randomized trial: assigned_p1 is randomized microenterprise-support assignment. The selected endline ITT with baseline controls remains an experiment, not an observational DiD. | Source 1 · Historical report |
| Internal Labor Migration as a Shock Coping Strategy: Evidence from a Typhoon | nonbinary treatment: The source uses post-typhoon wave interactions with the fraction of village land within 5 km of the storm track (treat_aft_5), survey weights and household controls. This is continuous exposure rather than binary adoption. | Source 1 · Historical report |
| Information Asymmetries in Consumer Credit Markets: Evidence from Payday Lending | instrumental variables: The source uses RD/RK and IV variation to instrument continuous loan_amount, including a DiD-style instrument in Appendix Table 1. The loan-level analysis_file.dta is absent; an IV interaction is not an observational binary-treatment ATT. | Source 1 · Historical report |
| Media and Political Persuasion: Evidence from Russia | nonbinary treatment: The two-wave Russian election panel uses Watch_probit_, a continuous NTV exposure measure in 1999 and zero in 1995. Binary adoption methods cannot reproduce its per-unit-exposure coefficient. | Source 1 · Historical report |
| How Merchant Towns Shaped Parliaments: From the Norman Conquest of England to the Great Reform Act | not did: The selected source is a medieval borough cross-section: parliamentary representation in 1348 on prior Farm Grant status and royal-borough status. There is no pre/post outcome clock. | Source 1 · Historical report |
| Temporary Protection and Technology Adoption: Evidence from the Napoleonic Blockade | nonbinary treatment: The short-run department panel estimates the effect of log shipping-route distance lnshortestLo on spindle capacity. Exposure is continuous; there is no corresponding binary adoption variable in this source specification. | Source 1 · Historical report |
| Free Trade and the Formation of Environmental Policy: Evidence from US Legislative Votes | nonbinary treatment: TreatImp and TreatExp are continuous import/export tariff-reduction exposures interacted with the NAFTA period. The design is continuous-treatment DiD, outside this binary-treatment template rather than invalid DiD. | Source 1 · Historical report |
| When Tariffs Disrupt Global Supply Chains | nonbinary treatment: The selected import-value equation uses the continuous tariff-change interaction dltariffCHNOASIAtreat. NewTariff coding or a China indicator is not the whole treatment regressor; it is not thresholded. | Source 1 · Historical report |
| Energy Saving May Kill: Evidence from the Fukushima Nuclear Accident | nonbinary treatment: Table 2 estimates mortality responses to continuous electricity-saving targets interacted with temperature-bin counts and age groups. The reported effect is an age-share-weighted linear combination with population weights, three interacted fixed effects and two-way clustering. It is not a binary adoption ATT. | Source 1 · Historical report |
| The Effects of Import Competition on Unionization | nonbinary treatment: nexposure is standardized industry import-competition exposure in the unionization regression. Its continuous variation is retained conceptually, not recoded into treated/untreated industries. | Source 1 · Historical report |
| Market Power and Price Exposure: Learning from Changes in Renewable Energy Regulation | nonbinary treatment: The headline is a triple interaction with continuous predicted premium exposure. Binary adoption would discard the continuous-exposure component of the estimand. | Source 1 · Historical report |
| The Effect of Immigration Restrictions on Local Labor Markets: Lessons from the 1920s Border Closure | nonbinary treatment: The source QE2Xpost multiplies continuous quota exposure by the 1930 period indicator, with region-period and baseline-immigrant-share controls. Binary recoding would change the border-closure exposure estimand. | Source 1 · Historical report |
| Competition and Pass-Through: Evidence from Isolated Markets | nonbinary treatment: The focal excise coefficient measures continuous tax pass-through across three tax changes and five fuel products. Pooling these repeated price shocks into first binary adoption would change the model. | Source 1 · Historical report |
| The Human Side of Structural Transformation | nonbinary treatment: predyield1 is continuous country-specific predicted agricultural-yield exposure. Country/year effects estimate a dose coefficient, not a binary-adoption ATT. | Source 1 · Historical report |
| Improving Police Performance in Rajasthan, India: Experimental Evidence on Incentives, Managerial Autonomy, and Training | randomized trial: Police-station intervention arms and training intensity are experimentally assigned. percenttrained_post is a multivalued dose; within-station pre/post estimation remains experimental. | Source 1 · Historical report |
| Trade Shocks and the Provision of Local Public Goods | instrumental variables: The headline is long-difference 2SLS of local outcomes on Chinese import exposure d_tradeusch_pw, instrumented by other-country import exposure. The treatment is continuous, not a binary adoption. | Source 1 · Historical report |
| Are State Governments Roadblocks to Federal Stimulus? Evidence on the Flypaper Effect of Highway Grants in the 2009 Recovery Act | instrumental variables: Highway grants per capita are instrumented with lagged road-formula variables in a cross-state fiscal-response regression. Neither continuous grant exposure nor its instruments are binary adoption. | Source 1 · Historical report |
| The Effect of State Taxes on the Geographical Location of Top Earners: Evidence from Star Scientists | nonbinary treatment: The migration equation uses the continuous origin-destination log net-of-tax differential ATR_p99_diff_log. Varying tax doses are not replaced by binary adoption. | Source 1 · Historical report |
| Taxation and the International Mobility of Inventors | nonbinary treatment: logret is the continuous log top net-of-tax retention rate in inventor-location regressions. The tax elasticity is not a binary policy-adoption ATT. | Source 1 · Historical report |
| A Machine Learning Approach to Analyze and Support Anticorruption Policy | instrumental variables: All numerical source targets condition on narrow1_prediction_NS, an XGBoost prediction absent from the deposit. The supplied R script reconstructs audit timing only, not outcomes or the final dataset. Rebuilding the documented week-long ML preparation is outside this template run; no timing-only object is passed off as an estimated effect. | Source 1 · Historical report |
| Public Information Is an Incentive for Politicians: Experimental Evidence from Delhi Elections | randomized trial: Delhi councillors were randomly selected for newspaper report cards. The before/after interaction is an experimental treatment effect. | Source 1 · Historical report |
| Demand and Supply of Infrequent Payments as a Commitment Device: Evidence from Kenya | randomized trial: Randomized monthly-payment and flexibility arms in the Kenyan dairy experiment. Binary assignment with before/after outcomes is still an RCT. | Source 1 · Historical report |
| Social Networks as Contract Enforcement: Evidence from a Lab Experiment in the Field | randomized trial: The lab-in-field contract-enforcement experiment estimates game-condition interactions with network distance. Individual effects and repeated game rounds do not identify a policy-adoption DiD. | Source 1 · Historical report |
| Should Aid Reward Performance? Evidence from a Field Experiment on Health and Education in Indonesia | randomized trial: Subdistricts are randomized to incentivized, nonincentivized or control block-grant arms. The indicator-pooled health/education effects remain experimental; key analysis inputs are unavailable. | Source 1 · Historical report |
| Is It Whom You Know or What You Know? An Empirical Assessment of the Lobbying Process | not did: s_Indicator distinguishes specialist from nonspecialist lobbyists in a pooled congress-effects regression of opposite-party connections. It is a binary characteristic, not a policy adoption or pre/post treatment. | Source 1 · Historical report |
| Regulation by Shaming: Deterrence Effects of Publicizing Violations of Workplace Safety and Health Laws | instrumental variables: Press-release effects use a fuzzy RD/IV around region-period penalty cutoffs. Restricted OSHA inputs are absent; a binary release indicator would not remove the RD/IV identifying assumptions. | Source 1 · Historical report |
| The Welfare Effects of Nudges: A Case Study of Energy Use Social Comparisons | randomized trial: Home-energy-report receipt is experimentally assigned. Recipient-by-billing-cycle effects and the compliance-IV robustness lane remain experimental, not observational DiD. Utility and survey microdata are NDA-restricted. | Source 1 · Historical report |
| The Wind of Change: Maritime Technology, Trade, and Economic Development | instrumental variables: The paper uses sailing-versus-steam travel-time instruments in gravity/trade and income equations. The supplied OLS travel-time regression is not a binary-adoption DiD substitute for that design. | Source 1 · Historical report |
| Local Responses to Federal Grants: Evidence from the Introduction of Title I in the South | instrumental variables: The change in federal revenue per pupil is instrumented by the 1960 poverty share fracpoor60. Long differences and county clustering do not replace the IV identifying assumption with binary DiD. | Source 1 · Historical report |
| The Making of Civic Virtues: A School-Based Experiment in Three Countries | randomized trial: The civic-education intervention is a school-level cluster RCT. The public prepared target is descriptive; missing panelized student/teacher files do not justify treating it as observational DiD. | Source 1 · Historical report |
| The Negligible Effect of Free Contraception on Fertility: Experimental Evidence from Burkina Faso | randomized trial: The full contraceptive-subsidy arm is randomly assigned by clinic catchment. Treatment effects are experimental arm comparisons, not observational adoption. | Source 1 · Historical report |
| Rational Habit Formation: Experimental Evidence from Handwashing in India | randomized trial: Handwashing incentives and monitoring are experimentally assigned across six arms. The selected I1-versus-control comparison is randomized, not observational adoption. | Source 1 · Historical report |
| Competition and Entry in Agricultural Markets: Experimental Evidence from Kenya | randomized trial: Agricultural market entry subsidies are randomized. Both the S2 reduced form and the IV for entrant counts derive from that experiment, not observational adoption. | Source 1 · Historical report |
| Capital Cities, Conflict, and Misgovernance | not did: The headline regresses average intrastate conflict on log distance to the capital across geographic cells, with country effects. It is a spatial continuous-exposure regression, not a unit-time adoption panel. | Source 1 · Historical report |
| Up in Smoke: The Influence of Household Behavior on the Long-Run Impact of Improved Cooking Stoves | randomized trial: Improved cooking stoves are offered by randomized lottery. Rolling lottery dates and household panels do not change the experimental assignment design. | Source 1 · Historical report |
| Medicare Part D: Are Insurers Gaming the Low Income Subsidy Design? | instrumental variables: Table 5 relates premiums to continuous current/lagged low-income-subsidy enrollment shares; the paper also uses market-level IV specifications. This mixed design is not a binary-adoption ATT. | Source 1 · Historical report |
| Do Oil Windfalls Improve Living Standards? Evidence from Brazil | instrumental variables: Oil output per capita is continuous; it enters reduced-form regressions and instruments municipal revenue in the main IV design. Geographic allocation instruments do not create a binary-treatment DiD. | Source 1 · Historical report |
| The Gorbachev Anti-Alcohol Campaign and Russia’s Mortality Crisis | nonbinary treatment: The source interacts campaign and post-campaign periods with pre-campaign alcohol consumption. This is a continuous-dose intervention with a later reversal, not a binary absorbing-policy estimand. | Source 1 · Historical report |
| Coercive Contract Enforcement: Law and the Labor Market in Nineteenth Century Industrial Britain | nonbinary treatment: The source reduced form uses coal- and iron-county price interactions and the textile employment share interacted with textile prices. These are multiple time-varying industry shocks, not an absorbing binary treatment. | Source 1 · Historical report |
| Incarceration and Incapacitation: Evidence from the 2006 Italian Collective Pardon | nonbinary treatment: Table 6 is a population-weighted cross-province regression of differenced crime on pardon intensity around a single national pardon. It has continuous exposure and no unit-time adoption panel in the estimated specification. | Source 1 · Historical report |
| The Internet and Local Wages: A Puzzle | nonbinary treatment: The Table 2 source is a county cross-section of 1995–2000 wage growth on advanced-internet adoption intensity and controls. It is not the binary unit-time panel required by this template. | Source 1 · Historical report |
| The Evolution of Local Labor Markets after Recessions | nonbinary treatment: The stacked recession event studies use continuous local recession severity. The generated CBSA/CZ analysis panels are also absent; recession dates do not make the severity treatment binary. | Source 1 · Historical report |
| Monitoring in Small Firms: Experimental Evidence from Kenyan Public Transit | randomized trial: Vehicle monitoring is randomly assigned to Kenyan matatu firms. Six assignment-by-study-month effects remain experimental, not observational staggered adoption. | Source 1 · Historical report |
| Task-Based Discrimination | not did: Race-by-decade indicators trace changes in racial wage gaps with rich individual/composition effects. Race is not a policy adoption, and a collection of binary interaction columns does not create the template intervention. | Source 1 · Historical report |
| Does the Squeaky Wheel Get More Grease? The Direct and Indirect Effects of Citizen Participation on Environmental Governance in China | randomized trial: Polluting firms are randomized across public/private complaint arms and control. The seven-arm field experiment and violation-triggered interventions remain experimental. | Source 1 · Historical report |
| Collateralized Marriage | instrumental variables: House-price exposure enters OLS/IV marriage-market specifications using geographic/age-linked instruments. It is continuous exposure with an IV strategy, not binary adoption. | Source 1 · Historical report |
| DETER-ing Deforestation in the Amazon: Environmental Monitoring and Law Enforcement | instrumental variables: Lagged enforcement fine counts are instrumented by lagged DETER cloud cover. This is a continuous-intensity panel IV design, not binary-adoption DiD. | Source 1 · Historical report |
| Brexit Uncertainty and Its (Dis)Service Effects | nonbinary treatment: The Brexit specification interacts continuous tariff-risk exposure with the policy period and absorbs bilateral-quarter and bilateral-sector effects. Binary adoption methods would change the exposure target. | Source 1 · Historical report |
| When a Doctor Falls from the Sky: The Impact of Easing Doctor Supply Constraints on Mortality | randomized trial: Doctors and mid-level providers are randomly assigned to clinics. The doctor coefficient is an experimental arm contrast with randomization-stratum effects. | Source 1 · Historical report |
| Subsidies and the African Green Revolution: Direct Effects and Social Network Spillovers of Randomized Input Subsidies in Mozambique | randomized trial: Voucher-by-subsidy-period effects and network spillovers come from randomized input subsidies in Mozambique. Village-round fixed effects do not change the experimental assignment design. | Source 1 · Historical report |
| Exploiting Externalities to Estimate the Long-Term Effects of Early Childhood Deworming | randomized trial: Early-life exposure dp1 derives from the school-level randomized deworming program. Age-at-exposure and survey-cohort controls do not change that experimental source of identification. | Source 1 · Historical report |
| The Welfare Cost of Perceived Policy Uncertainty: Evidence from Social Security | not did: The selected target is a cross-sectional regression of elicited Social Security risk premia on perceived benefit uncertainty sdben and survey controls. There is no before/after treatment-comparison panel. | Source 1 · Historical report |
| One Mandarin Benefits the Whole Clan: Hometown Favoritism in an Authoritarian Regime | nonbinary treatment: The source PowerCapital treatment takes values 0 through 7, measuring cumulative political connections. The original coefficient reproduces in the source capture, but interpreting this multivalued exposure as binary adoption would change the estimand. | Source 1 · Historical report |
| Inflation Dynamics during the Financial Crisis | nonbinary treatment: Table 1 requires confidential BLS PPI microdata, Compustat inputs and their matched identifiers. The deposited package contains code but not ppi_cstat_q.csv; the supplied R script explicitly stops for missing restricted inputs. | Source 1 · Historical report |
| Resource Constraints and the Criminal Justice System: Evidence from Judicial Vacancies | instrumental variables: Judicial-vacancy effects use judge deaths and pension-eligibility-based predicted vacancy spells in OLS/2SLS. Actual court-capacity exposure is fractional, and the required case-level outcome data are not deposited. | Source 1 · Historical report |
| Skill Transferability, Migration, and Development: Evidence from Population Resettlement in Indonesia | not did: The source occupational-choice regressions use standardized agroclimatic and linguistic similarity in a resettlement cross-section, with geography and demographic controls. No unit-time adoption panel is present. | Source 1 · Historical report |
| Network Structure and the Aggregation of Information: Theory and Evidence from Indonesia | not did: The selected regression is cross-sectional village-level y on inverse network distance Dist_inv, with village clustering. It has no treatment timing or before/after comparison. | Source 1 · Historical report |
| Political Parties and Labor-Market Outcomes: Evidence from US States | not did: The gubernatorial-party design relies on close-election RD, retaining vote-margin polynomials and interactions. The Democratic-winner indicator is binary, but the identification is not the template adoption DiD. | Source 1 · Historical report |
| Kwacha Gonna Do? Experimental Evidence about Labor Supply in Rural Malawi | randomized trial: The village labor-supply elasticity is estimated using randomized wage offers. lnwage has multiple doses; village/week fixed effects do not make it observational binary DiD. | Source 1 · Historical report |
| Does Agriculture Generate Local Economic Spillovers? Short-Run and Long-Run Evidence from the Ogallala Aquifer | nonbinary treatment: The source event study interacts the continuous Ogallala aquifer share with calendar years, using county weights and extensive geographic controls. It does not identify a binary aquifer-adoption ATT. | Source 1 · Historical report |
| Housing Vouchers and the Price of Rental Housing | nonbinary treatment: The audited rental-housing specification regresses log rent on log voucher supply lvouch, with time-varying controls and housing-unit effects. It estimates a continuous supply elasticity rather than adoption. | Source 1 · Historical report |
| Meet the Press: How Voters and Politicians Respond to Newspaper Entry and Exit | nonbinary treatment: The source regresses election-to-election turnout changes on newspaper entry and exit. Its signed change treatment is not a binary absorbing adoption indicator. | Source 1 · Historical report |
| When the Levee Breaks: Black Migration and Economic Development in the American South | nonbinary treatment: The source Table 2 event terms interact county flood intensity with post-1927 decades, retaining state-year effects and geographic controls. The target is a continuous flood-dose response, not a binary flood indicator. | Source 1 · Historical report |
| Making Yourself Attractive: Pre-Marital Investments and the Returns to Education in the Marriage Market | instrumental variables: Marriage-market exposures are instrumented with predicted immigrant shares and sex ratios. Fixed effects and first-stage regressions do not provide a binary-adoption DiD target. | Source 1 · Historical report |
| The Great Diversification and its Undoing | not did: The target links aggregate volatility to fundamental volatility var3 in a country panel. There is no policy treatment, adoption date or treated-versus-comparison intervention. | Source 1 · Historical report |
| Cracks in the Melting Pot: Immigration, School Choice, and Segregation | instrumental variables: The headline instruments changes in low-English Hispanic enrollment share with the predicted immigration shock inst_sca. This is a cross-sectional IV in long differences, not binary policy adoption. | Source 1 · Historical report |
| The Willingness to Pay for a Cooler Day: Evidence from 50 Years of Major League Baseball Games | nonbinary treatment: The target is a temperature response curve represented by eleven temperature-bin indicators. A vector of dose bins is not a binary treatment, and games are not absorbing policy adopters. | Source 1 · Historical report |
| Who Benefits from State Corporate Tax Cuts? A Local Labor Market Approach with Heterogeneous Firms: Reply | nonbinary treatment: The Reply reports a nonlinear firm-owner incidence share built from four continuous business-tax-shock coefficients and structural parameters. Substituting a binary tax-change indicator would change both treatment and estimand. | Source 1 · Historical report |
| The Rising Value of Time and the Origin of Urban Gentrification | nonbinary treatment: The selected occupation-level first-difference regression uses dval, the continuous change in the long-hours wage premium. It is not a binary intervention or adoption panel. | Source 1 · Historical report |
| Estimating Social Preferences and Gift Exchange at Work | randomized trial: The source gift-exchange specification uses post-gift rounds and multiple randomized gift arms with baseline output as a covariate; it is not the template’s binary-adoption design. | Source 1 · Historical report |
| A Precinct Too Far: Turnout and Voting Costs | not did: The source Table 2 coefficient is continuous distance to the polling station in the 2012 turnout cross-section, with boundary fixed effects. It is not a temporal adoption design. | Source 1 · Historical report |
| A Passage to America: University Funding and International Students | nonbinary treatment: The source Table 2 uses log state appropriations per student to explain log international-student enrollment. It is a weighted funding elasticity, not a binary adoption coefficient. | Source 1 · Historical report |
| Temperature and Decisions: Evidence from 207,000 Court Cases | nonbinary treatment: The asylum-grant regression uses continuous case-day temperature temp6t4 with rich fixed effects. Temperature shocks are not binary absorbing adoption. | Source 1 · Historical report |
| Trade and Income: Exploiting Time Series in Geography | instrumental variables: Trade growth is instrumented by time-varying geography/predicted trade measures in GDP-growth regressions. Continuous trade exposure is not binary policy adoption. | Source 1 · Historical report |
| Labor Drops: Experimental Evidence on the Return to Additional Labor in Microenterprises | randomized trial: The wage-subsidy voucher is randomly assigned. Voucher-by-round dynamic effects remain experimental even though the data contain pre/during/post periods and firm effects. | Source 1 · Historical report |
| Consumer Spending during Unemployment: Positive and Normative Implications | not did: The public targets are unemployment-exhaustion consumption profiles and structural model/calibration results, not a treated-versus-control DiD coefficient. Confidential household transaction data are also absent. | Source 1 · Historical report |
| The Optimal Timing of Unemployment Benefits | not did: The empirical target uses the unemployment-benefit schedule kink and DD-RKD; the public R target reproduces structural calibration artifacts. Neither is the template binary-adoption ATT, and confidential register microdata are absent. | Source 1 · Historical report |
| Wealth Heterogeneity and the Income Elasticity of Migration | nonbinary treatment: The two-wave village panel estimates effects of rainfall and rice-price shocks. Both are continuous time-varying exposures, not adoption indicators. | Source 1 · Historical report |
| Racial Discrimination in the Sharing Economy: Evidence from a Field Experiment | randomized trial: The Table 2 field experiment randomizes guest names and estimates the guest_black coefficient across booking inquiries. There is no before/after panel in the source regression; treating randomized names as adoption dates would be invalid. | Source 1 · Historical report |
| Report Cards: The Impact of Providing School and Child Test Scores on Educational Markets | randomized trial: Report cards are assigned experimentally. The selected school-fee regression is a treatment/control ANCOVA with baseline fees and district effects, not observational policy adoption. | Source 1 · Historical report |
| Ready for Boarding? The Effects of a Boarding School for Disadvantaged Students | randomized trial: The selected source is a randomized boarding-school offer effect on years of attendance, estimated in a weighted post-assignment cross-section. It is not a pre/post adoption panel. | Source 1 · Historical report |
| The Judge, the Politician, and the Press: Newspaper Coverage and Criminal Sentencing across Electoral Systems | instrumental variables: The audited two-stage specification uses continuous newspaper congruence and its interactions with judicial selection systems. There is no binary absorbing treatment in the focal media coefficient. | Source 1 · Historical report |
| The Effect of the TseTse Fly on African Development | not did: The audited headline is a cross-sectional regression of precolonial outcomes on the continuous TseTse Suitability Index, with province clustering. It has no adoption clock or pretreatment panel. | Source 1 · Historical report |
| Stimulating Local Public Employment: Do General Grants Work? | instrumental variables: Cost-equalization grants are instrumented by the grant-formula discontinuity Dforcing1. The grant-amount effect is RD/IV, not the effect of binary adoption. | Source 1 · Historical report |
| Health and the Political Agency of Women | instrumental variables: The paper uses close-election/IV specifications for women politicians and child mortality. The supplied binned first-stage relationship is not an observational binary-adoption ATT. | Source 1 · Historical report |
| Mafia and Public Spending: Evidence on the Fiscal Multiplier from a Quasi-Experiment | instrumental variables: The fiscal multiplier instruments public-investment growth G with compulsory-administration shocks CD_S1 and L1CD_S2 after province/year residualization. Binary instruments are not the continuous endogenous treatment. | Source 1 · Historical report |
| Development Effects of Electrification: Evidence from the Topographic Placement of Hydropower Plants in Brazil | instrumental variables: Topographic hydropower potential instruments electricity infrastructure. The supplied first-stage coefficient and main IV effects cannot be reinterpreted as a binary electrification ATT. | Source 1 · Historical report |
| Limited Insurance within the Household: Evidence from a Field Experiment in Kenya | randomized trial: The source estimates repeated randomized income shocks to household members, not a one-time absorbing adoption. First-shock recoding would change the experiment. | Source 1 · Historical report |
| Late Budgets | nonbinary treatment: Table 4 regresses budget delay on the continuous year-over-year unemployment-rate change unemp_diff, with state effects. Binary political controls do not turn that exposure coefficient into a binary-policy ATT. | Source 1 · Historical report |
| The Economic Impacts of Climate Change: Evidence from Agricultural Profits and Random Fluctuations in Weather: Reply | nonbinary treatment: The source treatment is continuous weather exposure and its polynomial interactions, despite the inventory’s binary tag. No binary threshold is imposed. | Source 1 · Historical report |
| Government Advertising and Media Coverage of Corruption Scandals | nonbinary treatment: The target is continuous monthly government advertising expenditure in a newspaper/month fixed-effects regression of corruption coverage. No binary exposure is substituted. | Source 1 · Historical report |
| Immigration and the Neighborhood | instrumental variables: The selected house-value regression uses continuous changes in immigrant shares in decadal differences, and the paper also reports IV specifications. This mixed design is not the binary-adoption target. | Source 1 · Historical report |
Extension designs pending review
| Study | Reason | Evidence / archive |
|---|---|---|
| When Discounts Raise Costs: The Effect of Copay Coupons on Generic Utilization | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The deposited coupon-offer panel is present, but the IMS/NHCHIS utilization outcomes needed for the regression are not deposited. | Source 1 · Historical report |
| Leverage and Beliefs: Personal Experience and Risk-Taking in Margin Lending | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The supplied R entry point requires analysis_dataset.rds, which is absent. Loancontracts and other raw Stata inputs are deposited, but their linked contract/lender-exposure preparation has not been reconstructed for this template. | Source 1 · Historical report |
| Conflict and the Persistence of Ethnic Bias | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source is a triple-difference in appellant and judge ethnicity around a conflict. A binary policy is not constant within court-year cells, so the adoption template does not preserve this design. | Source 1 · Historical report |
| Religion, Division of Labor, and Conflict: Anti-Semitism in Germany over 600 Years | design review: PDF audit reopened this assessment; no new comparison is admitted. Main city-by-century specification interacts Protestant location with post-Reformation periods, with city/time effects. Separate mechanism IV does not eliminate this observational DiD. Review the chosen analysis, not the entire paper as one IV estimand. | Source 1 · Source 2 · Historical report |
| Investment Banks as Corporate Monitors in the Early Twentieth Century United States | design review: PDF audit reopened this assessment; no new comparison is admitted. Table 6 Panels B/C explicitly estimate author-defined high-underwriting versus other railroads before/after the Clayton Act; Figure 3 is also binary. Existing template selected that author-defined group. A continuous-exposure or IV headline label is not grounds to claim no binary DiD analysis. | Source 1 · Source 2 · Historical report |
| Propagation and Insurance in Village Networks | design review: Deep DiD follow-up: The source uses Post x Treatment with household, calendar month and event-time effects. The stack construction retains controls shocked more than 24 months later; absorbing status is within the chosen health-shock window, not a lifetime state. Preserve stack IDs, reused controls and household inference. | Source 1 · Source 2 · Historical report |
| Immigration, Employment Opportunities, and Criminal Behavior | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source Table 2 regression is a triple difference using immigration exposure and policy-phase interactions. A binary two-way adoption comparison would omit the third difference and alter the estimand. | Source 1 · Historical report |
| The Costs of Job Displacement over the Business Cycle and Its Sources: Evidence from Germany | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The replication package omits the restricted German IEB employment records; there are no deposited analysis data files. | Source 1 · Historical report |
| Clean Identification? The Effects of the Clean Air Act on Air Pollution, Exposure Disparities, and House Prices | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The selected appendix specification is a summer-by-NBP-region-by-post triple difference with county-season, county-year and season-year controls. It requires a triple-difference adapter, rather than relabeling that interaction as ordinary treatment adoption. | Source 1 · Historical report |
| Do Consumers Distinguish Fixed Cost from Variable Cost? “Schmeduling” in Two-Part Tariffs in Energy | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The deposited study code reads rawdata1/rawdata2 and analysisdata1/analysisdata2, none of which ship in the ZIP. Only the ancillary pollution datasets and software-example data are present. | Source 1 · Historical report |
| Understanding the Changing Structure of Scientific Inquiry | design review: Deep DiD follow-up: The authors define top-10 and top-5 Soviet-field indicators and interact them with afterdummy. Explicit year/field dummies are present. xtreg lacks the fe option and defaults to random effects despite explicit dummies; do not silently relabel its estimator. Poisson is a separate specification. Check omitted columns and authors coefficient before mapping. | Source 1 · Source 2 · Historical report |
| Trade Liberalization and Markup Dispersion: Evidence from China’s WTO Accession | design review: Deep DiD follow-up: The authors split industries at the 2001 tariff median and estimate separate group time profiles. This authenticates the binary graphical comparison, not a reported joint DiD ATT or its variance. Any new pooled scalar model must be labeled an extension, not a reproduction of an authored coefficient. | Source 1 · Source 2 · Historical report |
| Crossing Party Lines: The Effects of Information on Redistributive Politics | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source compares local versus national candidates within survey respondents, interacting candidate type with radio access and coverage. This is not a pre/post adoption clock; interpreting candidate type as time would change the design. | Source 1 · Historical report |
| Corruption Dynamics: The Golden Goose Effect | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source models multiple enforcement shocks and interactions, with two-way clustering by official and day. A single absorbing treatment would change the source design. | Source 1 · Historical report |
| Women’s Suffrage and Children’s Education | design review: Deep DiD follow-up: Figure 5 uses city-level outcomes, event-time indicators and baseline-characteristic interactions, population weights and state-clustered inference. Binary suffrage events differ from the paper main childhood-exposure measure. Preserve event bins and evaluate the authored heterogeneous contrast, not an arbitrary pooled average. | Source 1 · Source 2 · Historical report |
| Expansions in Maternity Leave Coverage and Children’s Long-Term Outcomes | design review: PDF audit reopened this assessment; no new comparison is admitted. Compares birth dates around maternity-leave reforms to the same dates in non-reform years: DiD/difference-in-discontinuities. Restricted inputs and birth-date/cohort mapping are execution issues, not proof of a non-DiD paper. | Source 1 · Source 2 · Historical report |
| Labor Supply and Directed Technical Change: Evidence from the Termination of the Bracero Program in 1964 | design review: Deep DiD follow-up: Table A5 estimates treatment_binary in ppmlhdfe with crop/year effects; gen_patents_crops.do lines 69 and 86–92 define above-median 1964 exposure x post-1965. Genuine authored binary DiD, but the selected target is Poisson, not a linear ATT. Retain link function and specify the effect scale. | Source 1 · Source 2 · Historical report |
| The Long-Term Effects of Management and Technology Transfers | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The firm-level data_productivity_program.dta is restricted and absent. Deposited event-study coefficient files are outputs, not the outcome panel needed for re-estimation. | Source 1 · Historical report |
| Telecracy: Testing for Channels of Persuasion | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source is a spatial-border regression of the 2010-minus-2005 vote-share change with distance polynomials and turnout weights. Applying panel adoption estimators to this first-differenced cross-section would change the design. | Source 1 · Historical report |
| Effective Health Aid: Evidence from Gavi’s Vaccine Program | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The supplied R entry point requires coverage_analysis_data.dta, absent from the archive. Raw vaccine, coverage and mortality sources are present; the country-vaccine-cohort merge and source three-way effects need a preparation adapter. | Source 1 · Historical report |
| No Child Left Behind: Subsidized Child Care and Children’s Long-Run Outcomes | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The deposit contains no analysis data. The child-care outcomes require restricted Statistics Norway registers. | Source 1 · Historical report |
| Payroll Taxes, Firm Behavior, and Rent Sharing: Evidence from a Young Workers’ Tax Cut in Sweden | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. Only CPI and industry-classification lookup data are deposited. The Swedish administrative worker/firm panel needed by the payroll-tax regressions is absent. | Source 1 · Historical report |
| Fighting Crime in Lawless Areas: Evidence from Slums in Rio de Janeiro | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source jointly estimates temporary police intervention and subsequent pacification, with unit-specific trends. A single absorbing treatment does not preserve its two-stage policy estimand. | Source 1 · Historical report |
| When Losses Turn into Loans: The Cost of Weak Banks | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The package’s Data directory is empty. The source firm-bank regressions require confidential BPLIM data. | Source 1 · Historical report |
| The Effect of High-Tech Clusters on the Productivity of Top Inventors | design review: PDF audit reopened this assessment; no new comparison is admitted. Table 2 Panel B follows the same inventors in 1996 and 2007, comparing baseline Rochester residents with other cities. This is a separate binary-shock DiD from the main continuous-cluster IV analysis. Selected Table 2 data remain unavailable in the recorded package. | Source 1 · Source 2 · Historical report |
| New Evidence on Information Disclosure through Restaurant Hygiene Grading | design review: PDF audit reopened this assessment; no new comparison is admitted. Table 3 has author-labeled binary policy timing, but its notes retain fractional boundary-ZIP exposure and its target is a triple interaction. Do NOT equate the binary label with a verified 0/1 analysis column. Review county-level alternatives and a faithful DDD mapping; selected continuous-share failure remains valid. | Source 1 · Source 2 · Historical report |
| Familiarity Does Not Breed Contempt: Generosity, Discrimination, and Diversity in Delhi Schools | design review: PDF audit reopened this assessment; no new comparison is admitted. School-by-cohort natural-experiment DiD uses a poor-classmate exposure indicator. Lab/field experiments measure outcomes, rather than randomly assigning the integration policy. Separate peer-adjacency IV is another analysis. Cohort timing and sample/inference mapping still need checking. | Source 1 · Source 2 · Historical report |
| The Use of Violence in Illegal Markets: Evidence from Mahogany Trade in the Brazilian Amazon | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source distinguishes three successive mahogany-trade regimes (1999–2001, 2002–2008 and 2009 onward). Pooling these into a single absorbing treatment would replace its multi-regime estimand. | Source 1 · Historical report |
| The Effect of a Universal Child Benefit on Conceptions, Abortions, and Early Maternal Labor Supply | design review: Deep DiD follow-up: Table 5 includes distinct local-discontinuity models and an all-sample post indicator with birth-time polynomials, birth-month and interview-quarter effects (line 56; employed outcome line 65). Do not conflate them. Reconstruct cohort/seasonal comparison and preserve polynomial controls before claiming an adoption ATT. | Source 1 · Source 2 · Historical report |
| Green Infrastructure: The Effects of Urban Rail Transit on Air Quality | design review: Deep DiD follow-up: Table 8 directly estimates opening x Taipei against Kaohsiung, with authored 30-day windows, trends/weather in controlled columns and week clustering. This is separate from the main time-discontinuity specifications. Only two cities: generic many-unit clustered inference is not a faithful replacement. | Source 1 · Source 2 · Historical report |
| Increasing Organ Donor Registration as a Means to Increase Transplantation: An Experiment with Actual Organ Donor Registrations | design review: PDF audit reopened this assessment; no new comparison is admitted. Table 2 compares California with other states before/after the 2011 registration-question change. A separate laboratory experiment does not make that policy rollout randomized. Verify final-version/table alignment before admitting a new target; historical source and template results already exist. | Source 1 · Source 2 · Historical report |
| Strengthening State Capacity: Civil Service Reform and Public Sector Performance during the Gilded Age | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The supplied R script expects a merged analysis file containing treat_post, base_n_tot, sid and city controls. The deposited raw errors files lack those columns; the merge/preparation and multi-way slope-effects adapter are required. | Source 1 · Historical report |
| School Spending and Student Outcomes: Evidence from Revenue Limit Elections in Wisconsin | design review: Deep DiD follow-up: The code uses repeated operational/capital referendum wins, their leads/lags, vote-share polynomials and district/year effects; reported effects are linear combinations (lines 82 onward). This is a binary temporal contrast, but not a single absorbing policy. Preserve the RD controls, repeated treatments and authored aggregation. | Source 1 · Source 2 · Historical report |
| E-governance, Accountability, and Leakage in Public Programs: Experimental Evidence from a Financial Management Reform in India | design review: PDF audit reopened this assessment; no new comparison is admitted. Separate nationwide staggered eFMS rollout is analyzed by DiD; Table 8 uses district/year effects and policy-on indicators. This is distinct from the randomized Bihar intervention. Existing source target must be matched to the observational rollout, not described as an RCT effect. | Source 1 · Source 2 · Historical report |
| The Effects of Rent Control Expansion on Tenants, Landlords, and Inequality: Evidence from San Francisco | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The deposited code requires Infutor tenant/property panels, which are absent from the ZIP. Public census and housing-inventory files do not supply the headline tenant outcome panel. | Source 1 · Historical report |
| Illegal Immigration, State Law, and Deterrence | design review: Deep DiD follow-up: Table 5 includes Arizona passage and subsequent injunction interactions, with different city or date effects by column. Preserve both phases and the exact sample; collapsing to first adoption would change the estimand. | Source 1 · Source 2 · Historical report |
| Buy the Book? Evidence on the Effect of Textbook Funding on School-Level Achievement | design review: Deep DiD follow-up: The code collapses spending by year and affected-district group and plots group means around 2004. It does not estimate a scalar DiD in this file. Recover pre-collapse district data; do not call a newly chosen scalar effect an original replication. Displayed bands are not a DiD standard error. | Source 1 · Source 2 · Historical report |
| Moderating Political Extremism: Single Round versus Runoff Elections under Plurality Rule | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. No treated unit has an observation in the calendar year immediately before adoption. The irregular municipal-election panel has no minus-one baseline on the requested annual clock; election years are not silently compressed into common annual waves. | Source 1 · Historical report |
| Unintended Effects of Anonymous Résumés | design review: Deep DiD follow-up: The control-firm sample compares minority versus other candidates during versus after experimental monitoring, with BEFORE_D, BEFORE and ZouI, weights and vacancy clustering. It is a secondary observational interaction, not randomized policy assignment or an absorbing adoption panel. | Source 1 · Source 2 · Historical report |
| Does Vertical Integration Decrease Prices? Evidence from the Paramount Antitrust Case of 1948 | design review: PDF audit reopened this assessment; no new comparison is admitted. Paper defines theater integration as binary. Library R preparation averages vi to theater-year (R/112_gil_replicate.R), creating fractional annual exposure; headline vit is a post-divestiture trend interaction. Reconstruct appropriate timing/target before fitting; do not threshold annual means or call the policy intrinsically continuous. | Source 1 · Source 2 · Historical report |
| Are Credit Markets Still Local? Evidence from Bank Branch Closings | design review: PDF audit reopened this assessment; no new comparison is admitted. Exposed/control tracts around merger-driven branch consolidation supply a reduced-form DiD. The IV effect of branch closures is a different estimand. A binary exposure comparison may be a candidate, but cannot be relabeled the IV effect; recorded numerical checks still apply. | Source 1 · Source 2 · Historical report |
| Citizenship, Fertility, and Parental Investments | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The deposit contains no analysis microdata. The source requires restricted SOEP, German Microcensus or KIGGS inputs. | Source 1 · Historical report |
| Taxes Today, Benefits Tomorrow | design review: Deep DiD follow-up: D and T are explicitly defined and used to form treated/control before/after bunching cells (lines 228 onward). This is genuine DiD structure for distributional statistics; raw earnings are not an interchangeable outcome. Preserve first-stage bunching construction and uncertainty. | Source 1 · Source 2 · Historical report |
| Creating Moves to Opportunity: Experimental Evidence on Barriers to Neighborhood Choice | design review: Deep DiD follow-up: The authors separately estimate KCHA 2016 and SHA 2018 policy DiDs (lines 54–64 and 130–140); the latter explicitly drops experimental treatment families (lines 87–91). This is not the main randomized mobility intervention. The historical administrative data require access; no successful run or ready adapter is implied. | Source 1 · Source 2 · Historical report |
| Learning to Coordinate: A Study in Retail Gasoline | design review: Deep DiD follow-up: Published page 597 independently reports Perth-versus-other-cities DiD 3.49 cpl (SE 0.39). The cited code only computes Perth pre/post means; it does not reproduce that multicity estimate. Obtain the exact comparison-city series and specification. Endogenous coordination timing and source-data access remain separate questions. | Source 1 · Source 2 · Historical report |
| The Introduction of the Income Tax, Fiscal Capacity, and Migration: Evidence from US States | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source uses stacked local projections of forward changes in revenue, with early/late response indicators and stack-year effects. These transformed outcomes are not the template’s level-outcome adoption panel. | Source 1 · Historical report |
| Adaptation to Environmental Change: Agriculture and the Unexpected Incidence of the Acid Rain Program | design review: Deep DiD follow-up: The binary ever-installed group is interacted with installation-relative time, with boiler and year effects. Earlier lines 241–278 define actual installation and never-treated timing. Outcome preparation carries forward sparse early emissions and fills missing emissions with zero; preserve and disclose these authored conventions rather than silently using raw annual outcomes. | Source 1 · Source 2 · Historical report |
| Pre-Event Trends in the Panel Event-Study Design | design review: PDF audit reopened this assessment; no new comparison is admitted. Methods paper contains several applications. Selected newspaper-count target is nonbinary, but SNAP participation is binary in another application. Its proposed confounding correction uses IV; data, target and identifying-assumption compatibility remain unresolved. One application cannot characterize the whole paper. | Source 1 · Source 2 · Historical report |
| Tax Me, but Spend Wisely? Sources of Public Finance and Government Accountability | design review: PDF audit reopened this assessment; no new comparison is admitted. Program participation indicator PMAT identifies program-on reduced forms with municipality/time effects, alongside tax-revenue IV estimates. A PMAT policy DiD and the effect of instrumented revenue are different targets. Preserve that distinction and verify the selected specification. | Source 1 · Source 2 · Source 3 · Historical report |
| Does Early Life Exposure to Cigarette Smoke Permanently Harm Childhood Welfare? Evidence from Cigarette Tax Hikes | design review: Deep DiD follow-up: Model 5 uses event indicators, whereas model 6 uses continuous tax09. The setup file drops hikes below 25 cents and limits overlapping hikes and window coverage. The threshold is authored, not invented here; retain variable hike sizes, original windows and microdata-access requirements. | Source 1 · Source 2 · Historical report |
| How Financial Incentives Induce Disability Insurance Recipients to Return to Work | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The code-only deposit omits di_data.dta, the restricted Norwegian disability-insurance panel. | Source 1 · Historical report |
| Do Expert Reviews Affect the Demand for Wine? | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source is a distributed-lag model of repeated wine reviews and review valence, not a single absorbing treatment. | Source 1 · Historical report |
| The Impact of Shrouded Fees: Evidence from a Natural Experiment in the Indian Mutual Funds Market | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source includes two fee-regime changes and closed-fund interactions. A single adoption effect would replace its multi-regime estimand. | Source 1 · Historical report |
| Strike Three: Discrimination, Incentives, and Evaluation | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source targets race-match by QuesTec interactions in pitch-level strike calls after batter, pitcher and umpire demeaning. This is not a one-time unit-level treatment adoption panel. | Source 1 · Historical report |
| Populist Leaders and the Economy | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The R entry point requires the merged ple_dataset.dta, which is absent. The ZIP contains raw macroeconomic and political inputs, not the completed multi-source country panel. | Source 1 · Historical report |
| School Inputs, Household Substitution, and Test Scores | design review: Deep DiD follow-up: Binary discretionary-funding receipt is used in test-score gain regressions (lines 99 and 105), separate from the randomized India sample. Preserve continuous rule funding as a control; construct baseline/endline and resolve receipt selection. This is not permanent policy adoption. | Source 1 · Source 2 · Historical report |
| Product Innovation, Product Diversification, and Firm Growth: Evidence from Japan’s Early Industrialization | design review: Deep DiD follow-up: Code has binary expansion vert_exp x common binary output-cut sotan in Poisson with half-year effects, not firm FE. vert_exp denotes current expansion (not fixed ownership); both group composition and output cuts change. This supports the source specialized comparison, not an absorbing-adoption ATT. Do not strip the link function or add firm effects and call it replication. | Source 1 · Source 2 · Historical report |
| Carbon Taxes and CO2 Emissions: Sweden as a Case Study | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The supplied numerical replication is a Swedish aggregate time-series regression of log gasoline consumption on tax levels, a carbon-tax indicator, trend and macro controls with Newey-West inference. It is not a unit-by-time adoption comparison. The paper’s synthetic-control analysis would require a separate source mapping. | Source 1 · Historical report |
| The Dynamic Consequences of State Building: Evidence from the French Revolution | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The selected Table 9 specification is a cross-section of cadastral survey completion by 1815 on capital status and historical controls. It has no pre/post outcome dimension; the paper’s separate event-study uses a different outcome panel and two-way clustering. | Source 1 · Historical report |
| The Role of Electoral Incentives for Policy Innovation: Evidence from the US Welfare Reform | design review: Deep DiD follow-up: The source fits binary lameduck and its interaction with continuous electoral margin, with state/year effects. A standalone lameduck coefficient is conditional on the interaction normalization, not a generic ATT. Preserve reversible spells and margin controls; no first-adoption recoding. | Source 1 · Source 2 · Historical report |
| Consumer Spending and the Economic Stimulus Payments of 2008 | design review: Deep DiD follow-up: The authors regress consumption changes on binary iesp and calendar-period effects separately from amount regressions and amount-IV models. Timing-based administrative variation is not an investigator-assigned RCT label. Preserve first differences, receipt window and timing-only samples; one-time receipt is not permanent treatment. | Source 1 · Source 2 · Historical report |
| Calorie Posting in Chain Restaurants | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The ZIP supplies nutrition characteristics but not the proprietary restaurant transaction/store-day outcomes required for the treatment comparison. | Source 1 · Historical report |
| Property Rights and Gender Bias: Evidence from Land Reform in West Bengal | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source estimand is a land-reform-by-first-son interaction. First-son status varies within household-year cells; the template’s unit-level adoption does not identify that triple difference. | Source 1 · Historical report |
| The Local Political Economy Effects of School Construction in Indonesia | design review: Deep DiD follow-up: Table 3 column 1 estimates binary post92 with village and year effects. Column 3 separately interacts post92 with binary INPRES exposure. These are not the same target as continuous construction intensity. Verify election timing, sample and exact column before attaching any saved fit. | Source 1 · Source 2 · Historical report |
| Manipulation of Procurement Contracts: Evidence from the Introduction of Discretionary Thresholds | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The supplied R script requires 393_analysis_sample.csv and a Stata-generated result file, neither included. Raw procurement inputs are present, but the Stata sample-construction adapter has not been reproduced. | Source 1 · Historical report |
| Do Consumers Exploit Commitment Opportunities? Evidence from Natural Experiments Involving Liquor Consumption | design review: PDF audit reopened this assessment; no new comparison is admitted. Primary estimand uses permitted sales hours, but authors also estimate binary sales-ban specifications in Table 5 column 4. They explicitly caution that these discard relevant variation and are low-powered. Consider only that author-defined alternative, not an invented threshold or equivalent headline estimand. | Source 1 · Source 2 · Historical report |
| Trade and Investment under Policy Uncertainty: Theory and Firm Evidence | design review: PDF audit reopened this assessment; no new comparison is admitted. Table 1 has an EC10/Spain-by-post-1986 policy DiD, separate from continuous tariff/uncertainty and structural analyses. The existing template chose an author binary accession specification. Recheck that target on its own merits. | Source 1 · Source 2 · Historical report |
| Commercial Imperialism? Political Influence and Trade During the Cold War | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source uses Driscoll–Kraay/HAC inference with lag 40. The common template currently supports one-way cluster inference, not this cross-sectional/time dependence correction. | Source 1 · Historical report |
| The Consequences of Radical Reform: The French Revolution | design review: PDF audit reopened this assessment; no new comparison is admitted. Earlier paper explicitly defines binary revolutionary/Napoleonic occupation measures and reports outcome regressions alongside duration and reform IV. Published-package target is duration of French presence. Binary alternative is a lead requiring published-version/data alignment, not a confirmed replacement replication. | Source 1 · Source 2 · Historical report |
| Electoral Accountability and Corruption: Evidence from the Audits of Local Governments | design review: The focal contrast is first-term versus term-limited mayors; randomly sampled audits measure outcomes, rather than randomly assigning this treatment. The source is nevertheless cross-sectional in its main corruption target; the secondary grant-panel specification needs a separate study-design decision. | Source 1 · Historical report |
| Intergenerational Mobility in India: New Measures and Estimates | design review: Deep DiD follow-up: Code defines late-status x post-1970 cohort and fits it with jati-region, region-cohort and jati-cohort effects (lines 83–94), weighted and multiway-clustered. It is a binary observational comparison with richer cohort structure, not a simple calendar-year adoption panel. | Source 1 · Source 2 · Historical report |
| Flooded Cities | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The source targets flood occurrence in city-years, allowing repeated floods, rather than first flood as absorbing adoption. A first-flood recoding would change the treatment. | Source 1 · Historical report |
| The Effect of Pollution on Worker Productivity | design review: Outside the main collection until the source identification and original treatment coding are verified; missing data or an unsupported adapter alone does not establish ineligibility. The README explicitly excludes CTrip worker records. Deposited pollution and weather data alone cannot reproduce the productivity outcome panel. | Source 1 · Historical report |
| The Decline in Long-Term Unemployment and SNAP Take-Up | design review: Deep DiD follow-up: The code estimates eight individual policies jointly and separately (lines 42–50), not only their average index. The source flags this as an absorbing candidate but its own pattern says absorbing status unverified. Code alone does not establish 0/1 support or no reversals: stateTakeup.do recodes selected fields, constructs veh by addition and noShortCert with a missing-value rule. Restore external inputs, count each actual policy and check transitions; choose one authored target and retain other-policy controls. | Source 1 · Source 2 · Historical report |
| Virtual Classrooms: How Online College Courses Affect Student Success | design review: Deep DiD follow-up: Appendix Figure 3 has authored opening dates and campus/term effects, separate from course-taking IV. Existing campuses are coded open_term=100, the same numeric event category as treated opening term; a conventional never-treated sentinel would not reproduce the exact design matrix. Resolve that normalization and pre-opening support. The supplied ZIP has code/readme, no analysis microdata. | Source 1 · Source 2 · Historical report |
| How Do 401(k)s Affect Saving? Evidence from Changes in 401(k) Eligibility | design review: The source regresses second-difference log assets on temporary 401(k) ineligibility. This is plausibly DiD-related, but the collapsed target is not a levels-on-adoption ATT. Reconstruct and review the original eligibility/time paths before admitting a template comparison; do not infer permanent ineligibility from the descriptive label. | Source 1 · Historical report |
| Wintertime for Deceptive Advertising? | design review: Deep DiD follow-up: Publisher Appendix A8 independently confirms coverage x post-launch x weekend with lower-order effects and resort/day clustering. The inspected deposit contains the main weekend regressions, not A8 implementation. Coverage is defined by at least ten post-launch reports; preserve that definition and flag its identification implications. Do not substitute the recurring weekend indicator for policy adoption. | Source 1 · Source 2 · Historical report |