Abstract
This study focuses on training-wage effects when training classifications are narrowly defined. The results show a strong and statistically significant positive training-wage effect when training is aggregated into broad classifications. When the training is disaggregated, however, most categories remain uncorrelated with wages. The exceptions are ‘management/professional’ training for both males and females, ‘computing’ training for females, and ‘technical/paraprofessional’ training for males.