Lottery-Based Instrumental Variable Identification Strategy
Unobserved youth characteristics (such as motivation, conscientiousness, or family background) confound observational comparisons between employed and unemployed youth, making unadjusted comparisons unreliable.
Picture this
Imagine trying to test if wearing running shoes makes people faster, but fast people are naturally more likely to buy running shoes. Using a lottery to hand out running shoes completely breaks that confounding connection, ensuring the only difference between the two groups is pure chance.
What the evidence says
First-stage F-statistics exceed standard weak-instrument thresholds, showing that winning a lottery increases SYEP participation by over 50 percentage points, isolating clean local average treatment effects (LATE) free from selection bias.
- Who
- N = 163,447 applicants randomized across hundreds of local provider lotteries in New York City (2005–2008).
- How
- Two-Stage Least Squares (2SLS) model using randomly assigned lottery win status as an instrument for actual SYEP program participation, conditioning on provider-year-age lottery pool fixed effects.
What to do
1. Utilize administrative randomized selection lotteries as instruments when evaluating oversubscribed public interventions to isolate unconfounded causal parameters.
From the source
"Because provider-level lotteries determine which applicants are selected for the program, we use lottery win status as an instrument for participation in SYEP."
The_Effects_of_Youth_Employment_Evidence_from_New_York_City_Summer.pdf
Tagged
Nearby findings
- Community Incentive and Distance Instrumental Variable MethodEconomics (general)
- Instrumental Variables Correction for Management Measurement ErrorEconomics (general)
- Administrative Social Security Data Linkage for Earnings MeasurementEconomics (general)
- Agnostic Machine Learning Heterogeneity Detection in Field TrialsEconomics (general)