Bounding a Matching Estimator: The Case of a Norwegian Training Program

摘要: In this paper we evaluate a Norwegian vocational training rehabilitation program by comparing employment outcomes of trainees and nonparticipants using nonexperimental data. A matching estimator is used to calculate the effect for different subgroups sample. We demonstrate how bounding can be intrinsic uncertainty estimated effects due selection on unobserved individual characteristics. After adjustment observed into programs find that overall around six percentage points. This mainly high significant individuals with low probability participation. calculating upper lower bounds test statistics hypothesis no sensitive selection. However, result positive who are less likely participate in not bias. These also have lowest probabilities, which indicates potential harmful cream skimming sector.Copyright 2001 Blackwell Publishing Ltd