Data collection
Data were collected through face-to-face interviews using asemi-standard questionnaire (see Appendix B). The questionnaire was divided into four sections: (1) farmers’ personal and family characteristics; (2) farmers’ perceptions of the environment in rural areas and their experience with environmental change; (3) farmers’awareness of agricultural infrastructure projects and their attitudes towards the results and implementation process of these projects;and (4) farmers’ willingness to participate in agricultural infrastructure projects, their acceptance of agri-environmental measures with and without subsidies, and the initial factors that affected their choices. In sections 2–4, respondents were asked to selecta score from a five-class scale to express their agreement with each statement. In section 4, respondents were required to answer yes or no to the agri-environmental measures to express their acceptance/disagreement. Three simple measures that could be implemented in both farming systems were provided and stated as follows: (1) plant grass/herbs or shrubs as a buffer strip along the field margin (buffer strip measure); (2) set apart some land with low or no chemical fertilizer and pesticide use (low input measure);and (3) plant grass/herbs as mulch in orchards (orchard mulching measure). The purposes of these measures were to increase the heterogeneity of the intensive agricultural landscape, preserve bio-diversity and protect the soil.A pre-survey was conducted in August 2013 in Beijing, and the final survey was conducted in September 2013 in Changsha and October 2013 in Beijing. Interviewees were randomly selected by interviewers and any comments made by the interviewees related to our topic were recorded.