ber 2001, and March 2003—the Bureau recommended against such adjustment, and the secretary of commerce accepted the recommendations.
The 2000 A.C.E., like its predecessors, the 1990 PES and the 1980 Post-Enumeration Program (PEP), was designed to estimate the population of the United States and major population groups by dual-systems estimation (DSE). This method is closely related to a widely used statistical methodology known as capture-recapture, which was first developed for estimating wildlife populations. The methodology requires adaptation for the census context, as described in Fienberg (2000) and Hogan (1992, 2000a,b).
The basic concept of dual-systems estimation is determining how many people counted in one survey are also validly counted in a second, independent survey. In the census context, the initial survey is called the E-sample, which consists of the census enumerations in the A.C.E. sample blocks.2 The subsequent independent survey is the P-sample, which consists of people living at addresses listed independently of the census address list in the sample of census blocks. Not every census enumeration is correct; some are erroneous (e.g., a duplicate), so the process also involves estimating how many of the records in the E-sample represent correct enumerations. This is done by visiting E-sample people who fail to match P-sample records to determine for each individual whether he or she was enumerated in the census despite being missed in the P-sample, or whether the person was enumerated in error, such as a duplicate or a fictitious enumeration.
In general terms, the P-sample and the E-sample are used to estimate two components of the formula for calculating the DSE for each of several hundred population groups, called poststrata. Poststrata are defined by categorizing people on the basis of such variables as age, sex, race, ethnicity, owner or renter, region, and others; about 400 poststrata were used for the original A.C.E. To the extent possible, the intention is to develop poststrata that group
The E-sample does not include every census enumeration in the A.C.E. blocks, for such reasons as subsampling of large blocks (see Appendix E.1).