District level crop yield estimation with reduced number of crop cutting experiments

Abstract

It was observed that around 1300000 Crop Cutting Experiments (CCE) were conducted in India every year to find out the crop yield estimates of several major and minor crops under General Crop Estimation Surveys (GCES). Due to shortage of manpower and huge bulk of work day by day the data quality is becoming questionable. To tackle this problem, a pilot study was conducted by ICAR-IASRI, New Delhi sponsored by Directorate of Economics and Statistics (DES), Ministry of Agriculture and Farmers Welfare (MoA & FW), Govt. of India to generate district level estimates of major crop yield from a reduced sample size of villages selected from the states. With the reduction in number of villages, the problem of no sample size in some districts were faced during the study where common design based estimates of crop yield cannot be generated. To tackle this problem Aggregate level Small Area Estimation (SAE) was used to tackle this problem. The results obtained from this study in the state of Uttar Pradesh for two major crops, i.e. rice and wheat for two seasons, i.e. kharif and rabi of Agriculture Year 2015-16 and for paddy in Assam for kharif of the Agriculture Year (AY) 2015-16 in India were discussed. The yield estimates were compared with the estimates released under GCES for AY 2015-16. It was found that the estimates obtained from reduced sample size of number of CCEs w.r.t. GCES, produced similar estimates with acceptable level of precision.