Abstract: Based on survey data from 1 004 households in rural area of five cities in Hubei Province, the Ordered Probit Model was employed to explore the impact of social interaction on farmers' adoption of green technologies. With instrumental variable conditional mixed process estimation (CMP) to solve endogenous problems, conclusions have been drawn that: (1) Social interaction influences farmers' adoption of green technologies through endogenous interaction and situational interaction. By endogenous interaction, farmers could obtain relevant information and learn new knowledge, thus improving the confidence and skills of green technologies adoption. Situational interaction can improve farmers' enthusiasm to adopt green technologies through the demonstration effect. (2) In terms of technology heterogeneity, endogenous interaction extensively influences farmers' adoption of various green technologies, while situational interaction mainly affects their adoption of green farming and green pesticide application technology. (3) Rural households in hilly areas are more affected by social interaction than those in plain areas. Therefore, in promoting agricultural green technologies, full play should be given to the role of social interaction, the channels for farmers to obtain information should be broadened, and more attention should be paid to the radiation and driving role of science and technology demonstration households. Meanwhile, measures should be taken to create a good atmosphere and establish a social trust system.

Key words: social interaction, adoption of green technology, Ordered Probit Model, the CMP method, Hubei Province