Recovering Distributions in Difference-in-Differences Models: A Comparison of Selective and Comprehensive Schooling

Abstract We compare the effects of selective and nonselective secondary education on children's test scores, using British data from the National Child Development Study. Test scores are modeled as the output of an additive production function. An important input is the child's unobserved initial endowment, which may be correlated with the education system attended. In this model, we generalize the difference-in-differences approach and identify the entire counterfactual distribution of potential outcomes. Our results suggest that the better performance of selective schools relative to nonselective ones is essentially due to differences in pupils' composition.

[1]  Peter J. Diggle,et al.  A Fourier Approach to Nonparametric Deconvolution of a Density Estimate , 1993 .

[2]  A. C. Kerckhoff,et al.  Effects of Ability Grouping in British Secondary Schools. , 1986 .

[3]  A. Manning,et al.  Comprehensive Versus Selective Schooling in England in Wales: What Do We Know? , 2006, SSRN Electronic Journal.

[4]  James J. Heckman,et al.  Estimating the Technology of Cognitive and Noncognitive Skill Formation. NBER Working Paper No. 15664. , 2010 .

[5]  Costas Meghir,et al.  The Effect of School Quality on Educational Attainment and Wages , 1998, Review of Economics and Statistics.

[6]  J. Florens,et al.  A SPECTRAL METHOD FOR DECONVOLVING A DENSITY , 2010, Econometric Theory.

[7]  G. Imbens,et al.  Efficient Estimation of Average Treatment Effects Using the Estimated Propensity Score , 2000 .

[8]  J. Heckman,et al.  Estimating Distributions of Treatment Effects with an Application to the Returns to Schooling and Measurement of the Effects of Uncertainty on College Choice , 2003, SSRN Electronic Journal.

[9]  Susanne M. Schennach,et al.  Estimation of Nonlinear Models with Measurement Error , 2004 .

[10]  J. Angrist,et al.  Instrumental Variables Estimates of the Effect of Subsidized Training on the Quantiles of Trainee Earnings , 1999 .

[11]  D. Rubin.  Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[12]  Alberto Abadie.  Semiparametric Difference-in-Differences Estimators , 2005 .

[13]  Marianthi Markatou,et al.  Semiparametric Estimation Of Regression Models For Panel Data , 1993 .

[14]  P. Hall,et al.  Optimal Rates of Convergence for Deconvolving a Density , 1988 .

[15]  Petra E. Todd,et al.  On the Specification and Estimation of the Production Function for Cognitive Achievement , 2003 .

[16]  H. James.  VARIETIES OF SELECTION BIAS , 1990 .

[17]  A. Munk,et al.  Non‐parametric confidence bands in deconvolution density estimation , 2007 .

[18]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[19]  KyungMann Kim,et al.  Contrasting treatment‐specific survival using double‐robust estimators , 2012 .

[20]  Jianqing Fan.  ASYMPTOTIC NORMALITY FOR DECONVOLVING KERNEL DENSITY ESTIMATORS , 1989 .

[21]  Jianqing Fan.  On the Optimal Rates of Convergence for Nonparametric Deconvolution Problems , 1991 .

[22]  J. Heckman,et al.  Making the Most out of Programme Evaluations and Social Experiments: Accounting for Heterogeneity in Programme Impacts , 1997 .

[23]  J. Johannes.  DECONVOLUTION WITH UNKNOWN ERROR DISTRIBUTION , 2007, 0705.3482.

[24]  Stéphane Bonhomme,et al.  Stéphane Bonhomme Ulrich Sauder Accounting for unobservables in comparing selective and comprehensive schooling , 2009 .

[25]  A. Vignoles,et al.  The Heterogeneous Effect of Selection in Secondary Schools: Understanding the Changing Role of Ability , 2004, SSRN Electronic Journal.

[26]  W. Newey,et al.  Estimating vector autoregressions with panel data , 1988 .

[27]  James J. Heckman,et al.  Estimating the Technology of Cognitive and Noncognitive Skill Formation , 2010, Econometrica : journal of the Econometric Society.

[28]  Yanqin Fan,et al.  Partial identification of distributional and quantile treatment effects in difference-in-differences models , 2012 .

[29]  M. Lechner.  The Estimation of Causal Effects by Difference-in-Difference Methods , 2011 .

[30]  Peter Hall,et al.  Estimation of distributions, moments and quantiles in deconvolution problems , 2008, 0810.4821.

[31]  Ricardo Mora,et al.  Treatment effect identification using alternative parallel assumptions , 2012 .

[32]  Damon Clark.  Selective Schools and Academic Achievement , 2007 .

[33]  C. Meghir,et al.  Educational Reform, Ability, and Family Background , 2004 .

[34]  Petra E. Todd,et al.  The Production of Cognitive Achievement in Children: Home, School, and Racial Test Score Gaps , 2007, Journal of Human Capital.

[35]  Jason Abrevaya.  COMPUTING MARGINAL EFFECTS IN THE BOX–COX MODEL , 2002 .