%0 Journal Article %A Hao LI %A Shiping LI %A Ling NAN %A Xiaoqing LI %T A Meta-analysis of farmers’ environment-friendly pesticide application behavior in China %D 2018 %R 10.18402/resci.2018.01.08 %J Resources Science %P 74-88 %V 40 %N 1 %X

The main purposes of this study were to reveal key factors influencing farmers’ pesticide application behavior, and reasons for differences in research level. A total of 43 papers were selected from China and abroad. We explored the following issues: whether key factors in pesticide application in China are completely homogeneous; key factors influencing the application of pesticide in farmers under the condition of allowing the existence of heterogeneity; and trends in key factors over time. The meta-analysis method includes effect size selection, fixed or random effect model selection depending on the existence of heterogeneity, publication bias tests and correction, sensitivity analysis, cumulative meta-analysis and meta-regression analysis. The Q statistic was used to test whether there was heterogeneity. If variables are homogeneous in farmers, the fixed effect model is used, if variables are heterogenous, then the random effect model is used. Results showed that farmers’ pesticide application behavior was not completely homogeneous. There was significant heterogeneity among age, household income, knowledge of pesticide use, scale of cultivation and ratio of agricultural income. The key factors influencing behavior included gender, educational level, risk appetite, knowledge of pesticide use, farming experience, proportion of agricultural income, pesticide application training and pesticide residue detection. The trend in educational level, risk appetite, pesticide related knowledge, participation in pesticide application technical training and government propaganda support increased over time. However, variables including gender and agricultural income accounted for a significant impact, while the trend was nearly stable. The remaining variables gradually decreased. Based on these results, two suggestions spanning policy and academic are proposed.

%U https://www.resci.cn/EN/10.18402/resci.2018.01.08