The paper examines the impact of farmers' educational attainment on agricultural productivity. More specifically, it evaluates how farmers with vocational training perform compared to those with traditional educational training. A stochastic production frontier and inefficiency effects model is estimated using nationally representative household survey data to analyze the relationship between farmers' educational attainment and agricultural productivity in Vietnam, while controlling for factors such as gender and farmers' health status. The results indicate higher returns to vocational training in terms of its impact on raising agricultural productivity, as compared to primary and secondary education. Our findings confirm that significant productivity and welfare gains can be achieved through the promotion of education schemes tailored to the specific technical needs of smallholder or poor farmers. The lack of impact from primary and secondary education signals the need to adjust the curricula of nontraditional educational programmes in rural areas to respond to the technical and other skill needs of farmers. In other words, one general curriculum for everyone may not reap the highest returns to primary and secondary education investment in the context of countries with large farming populations. The originality of the paper resides also in the use of disaggregated education data in terms of formal and non-formal education. In addition, unlike previous studies, the production frontier function and the inefficiency segment are jointly estimated using a one step maximum likelihood estimation (MLE) procedure which guarantees both consistency and efficiency for estimated parameters. (Contains 3 figures, 6 tables, and 1 note.)