Abstract
The aim of this study was to determine farmers' perception of the harmful effects of using agro- chemicals in crop production with regard to environmental pollution. An attempt was made to explore relationships of 25 independent variables with the farmers' perception. The study also investigated the contribution as well as direct and indirect effects of the selected variables to farmers' perception. Data for this study were collected by interviewing from 302 randomly selected farmers of Kumargata union under Muktagacha upazila of Mymensingh district during January to May 1998. The findings revealed that more than one half (58.6 percent) of the farmers in the study area had low perception of the harmful effects of agro-chemicals while 35.8 percent had medium and only 5.6 percent had high perception. Chi-square test indicated that out of 25 variables, 20 independent variables had significant relationships with farmers' perception. Out of these 20 variables, 19 variables had significant positive relationships with farmers' perception. These are education, family education, knowledge on agro-chemicals use, information sources use, farm size, annual income, level of fertilizer use, economic status, social participation, cosmopoliteness, opinionatedness, innovativeness, level of aspiration, self-confidence, dynamism, attitude towards judicious use of agro-chemicals, agricultural problem awareness, environmental pollution awareness and risk orientation. The only variable which had significant negative relationship with farmers' perception was fatalism. Results of stepwise multiple regression analyses dictate that 10 variables were the best predictor of farmers' perception. These variables include education, knowledge on agro-chemicals use, agricultural problem awareness, opinionatedness, risk orientation, family education, farm size, economic status, dynamism and annual income. These 10 variables combinedly explained 87.7 percent of the total variation in farmers' perception. Out of these 10 variables 3 were found important predictor. These are education (79.10 percent), knowledge on agro-chemicals use (4.60 percent) and agricultural problem awareness (1.60 percent). The remaining seven variables contributed 2.4% in the prediction of farmers' perception. The findings of the path analysis revealed that eight variables namely, economic status, education, knowledge on agro-chemicals use, risk orientation, agricultural problem awareness, opinionatedness, dynamism and family education had positive direct effects and while the remaining two variables namely, annual income and farm size had negative direct effects on farmers' perception. It was also observed that all the 10 variables had positive and substantial total indirect effects on farmers' perception. However, it was also found that six variables namely, economic status, education, knowledge on agro-chemicals use, farm size, annual income and risk orientation were more important in general and the three variables namely, economic status, education and knowledge on agro-chemicals use were, in specific, most important in determining the farmers' perception.