The National Agricultural Statistics Service of the U.S. Department of Agriculture has been using the delete-a-group jackknife (DAGJK) in an increasing number of its surveys. This article discusses the theory behind the DAGJK when the first-phase of sampling is stratified and there are a large number of sampled units per stratum, which is the case for many list-based surveys. It goes on to propose an extension of the DAGJK for use when the number of sampled units per stratum is less than the number of jackknife replicates.
