Inference with Imperfect Randomization: The Case of the Perry Preschool Program
28 Pages Posted: 18 Apr 2011
There are 3 versions of this paper
Inference with Imperfect Randomization: The Case of the Perry Preschool Program
Inference with Imperfect Randomization: The Case of the Perry Preschool Program
Inference with Imperfect Randomization: The Case of the Perry Preschool Program
Abstract
This paper considers the problem of making inferences about the effects of a program on multiple outcomes when the assignment of treatment status is imperfectly randomized. By imperfect randomization we mean that treatment status is reassigned after an initial randomization on the basis of characteristics that may be observed or unobserved by the analyst. We develop a partial identification approach to this problem that makes use of information limiting the extent to which randomization is imperfect to show that it is still possible to make nontrivial inferences about the effects of the program in such settings. We consider a family of null hypotheses in which each null hypothesis specifies that the program has no effect on one of several outcomes of interest. Under weak assumptions, we construct a procedure for testing this family of null hypotheses in a way that controls the familywise error rate ヨ the probability of even one false rejection ヨ infinite samples. We develop our methodology in the context of a reanalysis of the HighScope Perry Preschool program. We find statistically significant effects of the program on a number of different outcomes of interest, including outcomes related to criminal activity for males and females, even after accounting for the imperfectness of the randomization and the multiplicity of null hypotheses.
Keywords: multiple testing, multiple outcomes, randomized trial, randomization tests, imperfect randomization, Perry Preschool Program, program evaluation, familywise error rate, exact inference, partial identification
JEL Classification: C31, I21, J13
Suggested Citation: Suggested Citation
Do you have negative results from your research you’d like to share?
Recommended Papers
-
Stepwise Multiple Testing as Formalized Data Snooping
By Michael Wolf and Joseph P. Romano
-
Improved Nonparametric Confidence Intervals in Time Series Regressions
By Joseph P. Romano and Michael Wolf
-
By James J. Heckman, Seong Moon, ...
-
Exact and Approximate Stepdown Methods for Multiple Hypothesis Testing
By Joseph P. Romano and Michael Wolf
-
Balanced Control of Generalized Error Rates
By Joseph P. Romano and Michael Wolf
-
Control of the False Discovery Rate Under Dependence Using the Bootstrap and Subsampling
By Joseph P. Romano, Azeem Shaikh, ...
-
Hypothesis Testing in Econometrics
By Joseph P. Romano, Azeem Shaikh, ...
-
Primate Evidence on the Late Health Effects of Early Life Adversity
By Gabriella Conti, Christopher Hansman, ...
-
By Ashok Kaul and Michael Wolf
-
Optimal Testing of Multiple Hypotheses with Common Effect Direction
By Richard M. Bittman, Joseph P. Romano, ...