This paper explores issues that arise in the evaluation of social programs using experimental data in the frequently encountered case where some of the experimental treatment group members drop out of the program prior to receiving treatment. The standard estimator for this case and the identifying assumption upon which it rests are begun with. Then, the behavior of the estimator when the dropouts receive a partial dose of the program treatment prior to dropping out of the program is examined. In the case of partial treatment, the identifying assumption is typically violated, thereby making the estimator inconsistent for the conventional parameter of interest: the impact of full treatment of the fully treated. A test of the identifying assumption underlying the standard estimator is developed and whether exclusion restrictions produce identification of the mean impact of the program when this assumption fails to hold is considered. Finally, alternative parameters of interest in the presence of partial treatment among the dropouts are discussed and it is argued that the conventional parameter estimating the effect of full treatment of the fully treated is not always the economically interesting one. These methods are applied to data from a recent experimental evaluation of the Job Training Partnership Act (JTPA) program. It is concluded from a sensitivity analysis of the data that the empirical consequences of the failure of the key identifying assumption can be quite substantial.
Bibliography: The Review of Economics and Statistics. February 1998. Vol. 80(1): 1-11
You should be able to view web pages in your web browser (Internet Explorer, Netscape, etc.)
