Combining Experimental and Structural Methods

What happens when researchers combine experimental randomization with structural modeling? This video explores a recent school voucher study in India that demonstrates why the "reduced form vs. structural" debate misses the point entirely. In this final installment of our structural econometrics mini-series, we examine how researchers used a two-stage randomized experiment in rural Andhra Pradesh to estimate the true welfare benefits of school choice. The twist? Their initial structural model failed to predict voucher take-up rates. But the good news is that this shortcoming led to crucial discoveries about household search behavior and school supply responses. Scientific Papers Cited: Peter Arcidiacono, Karthik Muralidharan, and John D. Singleton, "Experimentally Validating Welfare Evaluation of School Vouchers," NBER Working Paper 32968 (2024), https://doi.org/10.3386/w32968. Karthik Muralidharan, Venkatesh Sundararaman, "The Aggregate Effect of School Choice: Evidence from a Two-Stage Experiment in India," Quarterly Journal of Economics, Volume 130, Issue 3, August 2015, Pages 1011–1066, https://doi.org/10.1093/qje/qjv013 Topics Covered: Two-stage randomization design in development economics Out-of-sample validation of structural models Why the initial model under-predicted voucher take-up by 15-25 percentage points Hidden mechanisms: household search behavior and sibling enrollment effects Computing welfare impacts and marginal value of public funds Why this program generated $3 of household surplus for every $1 of government spending Tyler Ransom is an Associate Professor of Economics at the University of Oklahoma. Subscribe for more videos on data science, econometrics, and research methods! Editing credit: @neiljohnmanllios3064 #structuraleconometrics #reducedformmethods #schoolvouchers #developmenteconomics #randomizedcontrolledtrials #welfareanalysis #consumersurplus #Indiaeducation #experimentaleconomics #causalinference #econometricmethods #policyevaluation #educationeconomics #researchmethods #economicmethodology #RCT #fieldexperiments #discretechoicemodels #revealedpreference #credibleresearch