Computer simulation to compare three sampling plans for health and production surveillance in California dairy herds

A computer simulation study was carried out using production and reproduction data from 1642 dairy herds for 1984 to determine an optimum combination of sampling plan and sample size for obtaining precise estimates of the population and disease parameters of the California dairy cattle production. The four variables selected were: average milk production, California mastitis test score, services per conception and average days open (mean number of days a cow is non-pregnant from calving date to the last reporting breeding date). The sampling plans compared were: simple random, stratified (by herd size) random with proportional allocation, and stratified random Neyman allocation. There were 8 sample sizes ranging from 10 to 80 in increments of 10 herds. Stratified random sampling with Neyman allocation was the most efficient sampling plan for the health and production variables considered. The mean of such production variables as average milk production, services per conception and average days open could be estimated with precision using a sample size of 80 or less dairy herds. For disease variables with variance structure similar to CMT score, precise estimates could not be obtained using sample sizes of up to 80 herds; a sample size much greater than 80 herds is required for obtaining a precise estimate of the central value.