When the mean impact of an assigned social program cannot be estimated due to non-feasibility of a pre-intervention baseline survey, it is possible to track the ex-participants instead. By developing a triple-difference estimation model, measured outcome changes are compared between present participants and ex-participants. With sufficient follow…

Not all policy questions can be addressed by social experiments. Nonexperimental evaluation methods provide an alternative to experimental designs but their results depend on untestable assumptions. This paper presents evidence on the reliability of propensity score matching (PSM), which estimates treatment effects under the assumption of…