Abstract:
|
The Accuracy and Coverage Evaluation (A.C.E.) uses the dual-system estimation method. The dual-system estimation method assumes there are two independent lists of the population. The first list is the original census enumerations, and the second is a list of those covered by the sampling frame for the A.C.E. sample. The independence assumption can fail due to causal dependence, or conditioning of Census 2000 data collected in A.C.E. block clusters. Conditioning occurs when the event of an individual's inclusion or exclusion from one list affects the probability of his/her inclusion in the other list.
This study examines if the A.C.E. conditioned census data collected in A.C.E. blocks. To determine the potential existence of conditioning, we aggregate census data in A.C.E. blocks to the national, evaluation poststrata, or regional and Type of Enumeration Area (TEA) levels. Then, we compare census data in A.C.E. blocks to census data in non-A.C.E. blocks to see if significant differences exist.
|