The Counterfactual Self-Estimation of Program Participants

Abstract

This article proposes an innovative approach to estimating the counterfactual without the necessity of generating information from either a control group or a before-measure. Building on the idea that program participants are capable of estimating the hypothetical state they would be in had they not participated, the basics of the Roy-Rubin model are used to describe how the “counterfactual self-estimation of program participants” (CSEPP) creates a basis for estimating individual and average treatment effects. The problems confronting the approach are discussed as are the preconditions that need to be fulfilled for its use to be recommendable. Moreover, the article reports on the design and results of an empirical study conducted in order to gain first insights into the practicability and viability of the CSEPP. Finally, because there are some ambiguous findings, the strengths and weaknesses of the approach and the implications of the results for further research are discussed.