Using American panel data from the National Educational Longitudinal Study of 1988 (NELS:88) this paper investigates the effect of working during grade 12 on attainment. We exploit the longitudinal nature of the NELS by employing, for the first time in the related literature, a semiparametric propensity score matching approach combined with difference-in- differences. This identification strategy allows us to address in a flexible way selection on both observables and unobservables associated with part-time work decisions. Once such factors are controlled for, insignificant effects on reading and math scores are found. We show that these results are robust to a matching approach combined with difference-in-difference-in-differences which allows differential time trends in attainment according to the working status in grade 12.