Estimating treatment effects for discrete outcomes when responses to treatment vary: an application to Norwegian vocational rehabilitation programs

This paper analyzes the impact of interventions on discrete outcomes when responses to treatment vary among observationally identical persons. Using a latent variable model motivated by economics, we show how to define and identify various mean treatment effects as well as the distribution of treatment effects for discrete outcomes. The framework is based on discrete choice models with unobservables generated by factor structures. Responses to treatment vary among persons who are observationally identical, and agents participate in the program on the basis of their idiosyncratic response to treatment. We apply the model to study the Norwegian Vocational Rehabilitation training program.