Two-stage sampling designs applied to disease prevalence surveys of animal populations involve randomly selecting herds, flocks, etc. followed by randomly selecting a subsample of animals to be tested for the presence of the particular microorganism of interest. In the present study, data from an actual mastitis survey of dairy cattle was used in evaluating three unbiased two-stage sampling designs in terms of their applicability in disease prevalence estimation. Their performance was also compared to that of simple cluster sampling in which all animals in the selected herds are tested. The study demonstrated that these two-stage sampling designs are excellent alternatives to simple cluster sampling, being preferred in those cases where the variability in the prevalence among the herds is larger than the within-herd variability. The paper identifies some considerations that help to determine the appropriate balance between the number of herds to sample and the number of animals to subsample within a selected herd.
ASJC Scopus subject areas
- Animal Science and Zoology