Estimation of Crop Yields for Small Areas

Abstract

SUMMARY While the random sampling crop cutting surveys developed by I.C.A.R. are being followed in the States to provide crop yield estimates with a reasonable degree of precision at the district level, there is now a pressing demand for estimation of crop yields for each development block, which is a much smaller area. The simple crop-cutting approach is not feasible for this purpose, as requiring too elaborate an effort on the part of the field staff and excessive cost. A new technique involving two-phase sampling is, therefore, being tried by the Institute of Agricultural Research Statistics. In these surveys the yield of crop is obtained by eye-appraisal from a large sample of fields, while from a sub-sample of these fields the yield is obtained by the objective method of crop-cutting. The investigation of efficiency of this procedure, using the data collected during the survey on wheat crop in two districts in 1962-63 and 1963-64, showed that considerable reduction in the number of crop-cuttings and in the total cost can be achieved by adopting this procedure. The random sampling crop-cutting surveys developed by the Indian Council of Agricultural Research [1951] are being followed at present in most States to provide estimates of crop yield with a reasonable degree of precision at the district level. With the progress of planning in agriculture there is now a pressing demand for estimation of crop yields for each community development block, which is a much smaller area, there being about 5,000 such blocks in the country or an average of 15 blocks per district. If the simple crop-cutting approach is to be adopted directly for this purpose the present number of crop-cuttings will have to be increased greatly. The organisational and financial implications of such a huge programme will impose an unmanageable burden on the State Governments. Therefore, a new technique needs to be developed which could be adopted as an annual measure for estimation of block yields, which will not necessitate incurring heavy expenditure and is capable of being handled by the available local field agency with marginal additions to provide for requisite supervision of the field work and for processing the data. With this object in view, the Institute of Agricultural Research 374