Search

This study estimates the earnings impact of one of the most extensive in-service training programs in the developing world, Colombia's SENA, through a comparison of nongraduates' and graduates' earnings profiles. The possible influence of selection is corrected through the use of an ordered probit model to determine participation in short and long courses. The correction causes a significant decline in SENA's earnings effect. The results are robust with respect to alternative specifications of the selection model. (Reprinted by permission of the publisher.)