Impact Assessment of Farmer Field School Using A Multi-Period Panel Data Model

This paper presents an application of a two and three stages difference in differences model to measure environmental and economic impacts of Farmer Field School on crop and pest management practices of rice in Thailand. Panel data from 241 farm households were collected three times over a period of four years in five riceproducing provinces of Thailand. Data included socio-economic data describing household and farm characteristics, rice input and output data including detailed accounts of pesticide use and other pest management practices and farmers' knowledge of crop management and agro ecosystem factors. Using the concept of environmental impact quotient parameters on the health and environmental consequences of a change in pesticide use was computed. Data analysis was performed in two steps. First, statistical analysis using t-test was applied to detect differences in performance indicators, assumed to capture the influence of the training. These included farmer knowledge, yield, pesticide use, gross margin, pesticide health costs and environmental consequences of pesticides. In the second step, changes in farm performance, which were found to be significant in the linear statistical test were further analysed using a two and a three periods growth model. Results showed that trained farmers significantly reduced pesticide use on the short term. It was also found that they retain their reduced pesticide use practices several years after the training. However no significant change in rice gross margin could be detected.