Research location, population and sampling: Eight villages were selected for this research, two from each of the four districts in Bangladesh (Mymensingh, Jamalpur, Sherpur and Netrokona). These areas were selected due to: (i) the stagnation or reduction of crop yields in recent years[4] and ii) the gradual decline of soil fertility. In addition, the average cropping intensity of the study area is 218%. This is higher than the national average for Bangladesh (181%). This indicates that farmers in the study areas cultivate their lands intensively (more than twice a year) to obtain more products to meet family needs. 598 farmers drawn from 494 families in the eight villages are the population of the study. About twenty percent of the total population (120 farmers: 39 landless, 34 marginal, 19 small, 20 medium and 8 large) were randomly sampled for the study.
Measurement of farmers’ use of ISF and nm practices: Seven soil fertility management practices were selected for the study based on the recommendations of an expert panel from the Soil Science Department of Bangladesh Agricultural University. To measure the extent to which farmers use those techniques, a four-point summated rating scale was used. The respondents were asked to indicate the extent of their use of soil fertility management techniques as ‘do not use’, ‘rarely use’, ‘occasionally use’ or ‘regularly use.’ These responses were assigned scores of 0, 1, 2 and 3, respectively. Thus, a respondent could receive a score of 0 to 21 describing their overall use of different soil fertility management techniques. The Cronbach’s alpha coefficient for this section was 0.89. Plant nutrient management, on the other hand, has two main aspects. These are the management of organic nutrients and the management of inorganic nutrients (chemical fertilizers). To collect data concerning farmers’ use of organic manures for crop production, seven sources of manure were identified (cow dung, farmyard manure, crop residues, green manure, poultry feces, oil cake and ash) and the respondents were asked to describe the extent to which such manures were used. Possible responses were ‘do not use’, ‘very low use’ (1.0-3.0 ton ha−1 ), ‘low use’ (3.1-5.0 ton ha−1 ), ‘moderate use’ (5.1-7.0 ton ha−1 ) and ‘adequate use’ (>7.0 ton ha−1 ). These responses were assigned weights of 0, 1, 2, 3 and 4, respectively[7]. Thus, a farmer might receive a score of 0 to 28 describing the extent of that farmer’s organic manure use in crop production. For this data, Cronbach’s alpha coefficient was 0.86. To measure the extent of farmers’ use of chemical fertilizers, another four-point summated rating scale was employed. Seven types of chemical fertilizers were identified (Urea, Di-Ammonium Phosphate (DAP), SSP (Single Super Phosphate), Triple Super Phosphate (TSP), Muriate of Potash (MP), Gypsum and Zn fertilizers). The farmers were asked to indicate the extent to which each of these chemical fertilizers was used with: ‘do not use’, ‘use a self-assessed dose’, ‘partial recommended dose’ and ‘recommended dose.’ These responses were assigned a score of 0, 1, 2 and 3, respectively. Thus, a farmer could receive a score ranging from 0 to 21 describing the use of chemical fertilizers. The Cronbach’s alpha coefficient for this section was 0.87. The Cronbach’s alpha coefficients indicated that the measurement scales were internally consistent.
Data collection and analysis: Data were collected from the 120 sample farmers through face-to-face interviews using a structured survey questionnaire during the period from December 2005-January 2006. Information sought included farmers’ use of soil fertility and nutrient management practices for crop production. Data were also collected from the Bangladesh Bureau of Statistics and reports on agriculture and rural development of the areas under study. The collected data was grouped, summarized and presented in tabular form. Descriptive statistics such as frequency distributions, percentages and mean and standard deviations were employed for data analysis.