DDI_BWA_1993_AgC_v01_M Development Economics Data Group 2012-02-29 NADA Version 01: (February 2012) Agricultural Census 1993 AgC 1993 BWA_1993_AgC_v01_M Central Statistics Office (CSO) Agricultural Statistics Unit NADA Food and Agricultural Organization Central Statistics Office Agricultural Census [ag/census] The 1993 Botswana Agricultural Census was the second census conducted in the country. The first agricultural census, which covered both livestock and crops, was conducted in 1982. Since 1967, the Government of Botswana has been generating national agricultural statistics through an annual agricultural sample survey on crops and livestock. The first livestock census was carried out in 1971, while the first agricultural census, which covered both livestock and crops, was conducted in 1982. The census was based on a sample larger than the annual agricultural survey with both the scope and coverage more expanded. Data presented are from the 1993 Botswana Agricultural Census (1993 BAC), which aimed to establish new data bench-mark for the agricultural sector and a frame for annual surveys to be conducted over the next ten years. The 1993 BAC was conducted under the overall responsibility of the Central Statistics Office in the Ministry of Finance and Development Planning, in close cooperation with the Ministry of Agriculture (MOA). Within MOA, the major operational responsibility for the census was given to the Agricultural Statistics Unit (ASU). Technical and financial assistance were also provided by the Food and Agricultural Organization of the UN (FAO). A total of 200 enumerators were employed to carry out the field operations. Botswana National - Agricultural Holding and Holders The statistical population for the 1993 Botswana Agricultural Census consisted of all agricultural holdings in the rural areas, including large villages considered urban in the 1991 Population and Housing Census. This universe was divided into two parts, i.e. the Traditional and the Commercial Sector. Census/enumeration data [cen] The 1982 Block Sampling Frame, used for traditional holdings in the annual surveys, was updated with information obtained from the 1991 Population and Housing Census. A total of 3471 blocks were constructed and a sample of 999 blocks (clusters) was selected. The list prepared by the CSO, and reporting information from the Register of Deeds, was used as the frame for commercial holdings. The sampling design of the 1993 BAC was a probability stratified cluster (block) two-stage sample. In the first stage, Primary Sampling Units (PSUs) or Clusters (Blocks), further classified into three strata, i.e. Village stratum (1251 blocks), Land stratum (1589 blocks), and Cattle Post stratum (667 blocks), within each of the 26 Agricultural Districts, were selected systematically and with Probability Proportional to Size (PPS) of the number of total dwellings (occupied and not occupied). In the second stage, around 8000 Holdings or Secondary Sampling Units (SSUs) were systematically selected. In traditional sector, data were collected through direct interview and objective measurement techniques were applied. In commercial sector, questionnaires were mailed to holdings, with interviewers visiting non-respondents for an interview. Face-to-face [f2f]