This paper provides a meta-analytical review of empirical studies into the impact of schooling on entrepreneurship selection and performance. We first describe the main effects found in the current entrepreneurship literature. We then explain the variance in results across the hundreds of empirical studies by means of an analysis of variance. Five main conclusions result from this meta-analysis. First, we find that the impact of education on selection into entrepreneurship is neither positive nor negative. Second, the effect of education on performance is positive and significant. Third, the return to a marginal year of schooling in terms of the income it generates is 6.1 percent in the U.S.A. Fourth, the effect of education on earnings is smaller for entrepreneurs than for employees in Europe, but equal or larger in the U.S.A.. The fifth conclusion from the meta-analysis is the most striking and pertains to the estimated effect of education on performance: all results obtained so far are potentially biased. Estimation and identification strategies used to identify the effect of education on performance have merely measured the (conditional) correlation between education and performance rather than the causal effect, which is the estimate of interest. We finally conclude that the entrepreneurship literature on education can greatly benefit from the technical sophistication used to estimate the returns to schooling for employees by means of twins studies, field experiment or instrumental variables approaches.