We construct a revised version of the Barro and Lee (1996) data set for a sample of OECD countries using previously unexploited sources and following a heuristic approach to obtain plausible time profiles for attainment levels by removing sharp breaks in the data that seem to reflect changes in classification criteria. It is then shown that these revised data perform much better than the Barro and Lee (1996) or Nehru et al. (1995) series in a number of growth specifications. We interpret these results as an indication that poor data quality may be behind counterintuitive findings in the recent literature on the (lack of) relationship between educational investment and growth. Using our preferred empirical specification, we also show that the contribution of TFP to cross-country productivity differentials is substantial and that its importance relative to differences in factor stocks increases over time ...