Review of the 2001/2 Ethiopian agricultural sample enumeration methodology
2005
Biratu Yigezu | Eskindir Tenaw | Christopher Whitaker
In order to provide a better design for future agricultural sample surveys and censuses, 15 weredas from the Ethiopian Agricultural Sample Enumeration of Oromia, Amhara and SNNP regions, - the three regions with the largest total of cropland area and with the most diverse agricultural practices - have been considered. The analyses are based on three major crops (teff, maize and sorghum) and cattle population. Coefficient of Variations (CVs), design effects, intra-cluster correlations and between and within variances are calculated and evaluated. The average sample sizes required per wereda to achieve a specified CV (10%, 15% and 20%) are computed. Four different methods of allocating samples to strata are also compared for the 15 weredas. The level of precisions for most of the variable estimates are found to be fair, however, the design effects are high and most of the intra-cluster correlations are also low positive. Higher percentages of variation are observed within clusters (Ethiopian agricultural sample) than between Ethiopian agricultural sample. Results for the cattle population are found to be different from the results of the three crops analysed. Taking into account precision and resource, an average of about 660 households and 22 Ethiopian Agricultural Sample per wereda is found to be satisfactory in achieving a 20% CV for most of the variables. Of the four methods of sample allocations considered, power allocation is noted to be the most appropriate in providing relatively precise estimates for each wereda. For computing estimates of cropland area based on unbiased and ratio estimators, six weredas of the four major regions (Tigray, Amhara, Oromia and SNNP) are considered. Both unbiased and ratio estimates for the five major crops (maize, teff, sorghum, barley and wheat) are computed and compared. Ratio estimates calculated from each crop resulted in substantial improvement in precision.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by Ethiopian Institute of Agricultural Research
