Agricultural extension services have been dominated by development programs to improve the productivity of crops and enhance farmers’ income. The virtues and limitations of these programs ignite a debate among scholars from distinct strands of research. It is still unknown how effective extension services are in improving the agricultural productivity. We aim to unpack the relationship between these two constructs, and more specifically explore the main determinants driving systematic variations in the technical efficiency estimates from all relevant crop farming studies. A meta-regression analysis is conducted by using 335 observations from 199 farm level studies to review the effect of agricultural extension activities on farm performance. While the implementation of extension programs is likely to be non-randomly distributed in our sample, we employ the inverse probability of treatment weighting to correct for potential selection bias. Evidence for the absence of a publication bias in farm studies used in the meta-analysis is identified. Consonant with the theory of agricultural extension, we find that extension significantly improves technical efficiency by 4.8% to 7.6%. Farm productivity significantly differs in country level characteristics, type of crops and model specifications. Our empirical findings are robust when replacing missing observations with imputed values obtained from multiple imputation method.