An EOS was conducted on 1440 farms representing the dairy industry of Bangladesh in four agroecological zones, based on land type, soil pH, temperature and rainfall. The zones were Old Brahmaputra floodplain, Ganges tidal floodplain, Karatoya-Bangali floodplain and Chittgong coastal plain. The areas belonged to the administrative districts of Mymensingh, Khulna and Satkhira, Pabna and Sirajgonj, and Chittagong, respectively.
In Mymensingh, crop production is the main source of livelihood and per capita income is low. Dairy farming is mostly of subsistence type because capital investment is not possible. The soil of Khulna and Satkhira has a sodium ion concentration of 2.56–5.12 g/L. Owing to close proximity to the Bay of Bengal, the area is flood prone and the infrastructures are poorly developed. Socioeconomic conditions are similar to those of Mymensingh, except in Kulna Metropoliton area, where living standards are slightly better. The land of Sirajgonj and Pabna is submerged under floodwater during the monsoon season. A major part of this area has access to the Bangladesh Milk Producers’ Cooperative Union (BMPCU) Ltd. ‘Milk vita’ is the official trade name of BMPCU. Dairy farming is the major means of livelihood here because most farmers can cultivate only one crop per year (usually, three crops are cultivated per year in other areas of Bangladesh). Many farmers cultivate high-yielding fodder. Chittagong is the largest commercial city of Bangladesh. Living standards are higher than in the other areas studied. Dairy farming is commercial and intensive here.
Preparation of the survey schedule The field survey was conducted from March 2001 to June 2002. The survey form was designed by the School of Veterinary Medicine, University of Wisconsin Madison, USA and referred to as the Economic Opportunity Survey (EOS) form. A Microsoft Excel spreadsheet was programmed to enter and record data from the EOS form. The programming was done in collaboration between the Field Fertility Clinic, Bangladesh Agricultural University, Mymensingh, Bangladesh and the School of Veterinary Medicine, University of Wisconsin Madison, USA. The survey form had five sections: farm milk production on the day before the visit; expenses of cattle health care and management; inventory of herd culls and death; calf production per cow; and summary of herd management.
Data collection Because the project focused on the development of the small-scale market-oriented dairy industry in Bangladesh, farms that sell the majority of the milk produced were selected. They were visited once, and information was recorded by interviewing the farmers. If available, farm records were reviewed to obtain additional information and, if required, the animals were examined. All the prices and expenses were recorded in Bangladesh taka (BDT); however, after calculation, the figures for the economic opportunity and milk production cost were converted to US dollars (USD) (1 USD = 58 BDT).
Setting the targets The 20th and 80th percentiles of the management indices listed were calculated using the data from surveyed farms of the individual regions. Then the performance targets were identified for each region using the 20th or 80th percentile from farms depending on the index. Finally, the performance targets were entered in individual farm data to calculate the economic opportunity for different regions according to the methods described in the next section.
Calculation of management indices Age at first calving. The age at first calving was recorded in months and calculated by subtracting the birth date from the first calving date. The economic opportunity for age at first calving was calculated.
Statistical analysis The collected information was recorded utilizing Microsoft Excel 2000. A custom-designed workbook was created to record the data from five sections of the EOS form on four sheets. A macro was designed to transfer the data from the farm data workbook to a summary worksheet containing multiple farm data. We calculated the 20th, 50th and 80th percentiles of the variables studied. Because the data were nonparametric, Kruskal–Wallis test was used to test the null hypothesis that the districts did not differ from each other with respect to the parameters studied (Altman, 1991; Anon., 2000b). Mann–Whitney test was used to determine whether the differences between sources of economic opportunities were significant (Altman, 1991; Anon. 2000b). Data were plotted taking the milk production (litre/cow on one day) and cost of milk production (USD/litre) as X and Y axes, respectively.