Can social networks facilitate smallholders’ decisions to adopt climate-smart agriculture technologies? A three-level meta-analysis

The influence of social networks on the adoption decision for climate-smart agriculture (CSA) technologies and the relative effects of different network types remain controversial. To verify those claims, a three-level meta-analysis including 26 empirical studies and 150 effect sizes was conducted in this study. The results indicate a valid, modest yet positive correlation (0.065) between social networks and smallholders’ CSA technology adoption decisions, with no publication bias in this field. The subsequent heterogeneity test and subgroup analysis show that social network type is the main factor causing significant variation in effect sizes, with friendship and kinship networks having the greatest impact. In addition, various robustness tests were performed to verify the correctness of the model setting and moderator chosen and the stability of the heterogeneity test results. In conclusion, this study testifies to the efficacy of social networks’ roles raised by the diffusion of innovation theory. Policy implications lie in the extension of CSA technologies, which should be more ambitious, and more attention could be paid to the kinship and friendship networks.