Tangible Goods Simulation for Field BDM Auctions
Practice rounds enabled successful implementation of 4-contract BDM elicitations across 1,978 rural subjects, resulting in zero instances of participants refusing to complete winning transactions.
Tag
experimental_design
22 cards
Becker-DeGroot-Marschak Wage Elicitation Mechanism
Over 60% of participants accepted low piece rates (10 MWK/unit, yielding ~70 MWK daily earnings plus 50 MWK participation fee). Ex-post refusal rates were under 3% and declined over time, confirming that workers learned and adhered to truthful wage revelation.
Between-School Generalizability Limit of Within-School RCTs
Random assignment was strictly bounded within individual schools, grade levels, and subjects, preventing statistical evaluation of whether teacher value-added metrics hold predictive validity for cross-school teacher transfers or district-wide comparisons.
Actuarially Equivalent Cash Promises as Control Design
Methodological isolation ensured that observed increases in cash crop allocation (6 percentage point increase in probability, p = 0.041) [7] were strictly attributable to state-contingent risk transfer rather than income/wealth expansion [5, 6].
Caste-Stratified Market Penetration Randomization
Generated exogenous cross-village coverage variation ranging from 0% to 53% for cultivators and 0% to 100% for landless laborers, with zero statistically significant correlation to village size, caste concentration, or number of castes after conditioning on eligibility shares.
Caste-Stratified Village Identification Strategy
Conditioning on eligibility population shares eliminates endogeneity bias, reducing correlations between insurance marketing proportions and village characteristics (such as caste concentration and total households) to statistically insignificant levels (p-values > 0.33).
Casual-Labor Field-Laboratory Screening Proxy
Replicated real-world day-labor market conditions where 91 percent of participants were active labor force participants and 45 percent had made real job referrals, achieving a 72 percent return rate (N = 407 referrals) with robust experimental control over financial and screening incentives.
Covert Market Choice Experiment Framing
Covert framing successfully measured baseline discrimination where control group participants sacrificed goods worth nearly twice the median daily per capita food expenditure to avoid selecting transgender workers, while eliminating experimenter demand bias.
Cross-Community Audit Methodology for Bias Detection
Standardized multi-community audit visits successfully isolated seller pricing differences (-12% for Marwari sellers vs. +5% for Andhra sellers) and buyer-seller match effects independently of product quality or negotiation script variations.
Double-Sided Randomization Design
Placing 1.398 formal apprentices per firm displaced only 0.318 traditional apprentices, creating a net gain of 1.080 total apprentices per firm (p < 0.0001); net job creation was bounded between 74.1% and 77.3% of formal positions placed.
Effective Wage Differential Bundle Mechanism
The logit intercept divided by the slope directly identifies the marginal value of time, proving that effective wage variation cleanly isolates willingness to supply incremental work hours without income effects.
Intent-to-Treat (ITT) Reduced-Form Effect Estimation
The intent-to-treat reduced-form estimate yielded a coefficient of ~0.280 to 0.326, which reflected the true treatment impact (~0.955) diluted by an overall compliance rate of approximately 29.3%.
Non-Reallocation Resource Isolation
Zero observed instances of teacher or student reassignment across standards occurred because automatic promotion rules and fixed classroom counts removed principal discretion, ensuring control grades remained valid comparison groups.
Pre-Randomization Compliance Screening Filter
Retained 4,458 out of 8,248 initial recruits (54% retention rate) while maintaining statistical balance across observable baseline demographics between retained and non-retained cohorts (e.g., 74% female in retained vs 66% in non-retained).
Within-School Exchangeable Rosters and Randomization Blocks
668 randomization blocks were successfully formed across 284 schools. Standard deviation of baseline student sorting across classrooms within these blocks was 0.382 to 0.520 prior to randomization, demonstrating that blocked randomization successfully eliminated substantial pre-existing within-school student sorting.
Sanctuary Control Group Protocol
Uninterrupted linear trends in control firm contacts confirmed counterfactual stability, while treatment firms experienced a 488% jump in counselor visits and a 152% increase in phone calls without distorting control baseline behavior.
Side-Payment Neutralization Protocol
Guaranteed pre-task bonus disbursement successfully decoupled referral task motivation from referrer incentives, isolating pure peer screening from candidate effort responses.
Staggered Treatment Activation Protocol
Established a uniform 5-week pre-treatment baseline window per respondent, enabling precise subsample categorization between low and high pre-treatment remitters.
Stratified Cluster Randomization Design
Stratification achieved complete baseline balance across treatment and comparison groups, keeping all initial pretest score differences below 0.10 standard deviations.
Two-Stage School and Classroom Cluster Randomization
Disentangled direct treatment impacts from intra-school spillovers, proving zero statistical spillover to untreated classrooms (0.011 SD knowledge gain, 0.022 SD attitude gain).
Within-Firm Spillover Dilution Control
Removing multi-establishment corporate branches prevented internal information spillovers, maintaining orthogonal treatment variation across 7,438 independent establishments.
Worst-Case Variance Bounded Sampling
Setting lambda = 0.00125 guarantees a worst-case standard error of 0.05 for treatment contrasts and a statistical power >= 0.80 for detecting effect sizes of 0.124 across N = 4,000 participants, enforcing a minimum assignment probability floor of 0.05 per treatment arm.