EXECUTIVE SUMMARY

census enumerations outside the A.C.E. search area. However, the person duplication study could not determine which member of the linked pair was in the correct location. The fact that we do not know which member of the pair is correct does not matter much for national level estimates, but does matter for estimates for geographic areas. Assuming that the linked person does exist, the probability that each member of the link is in the correct location is estimated. Key Results • The estimate of the overall correct enumeration probability for E-sample cases with a duplicate link outside the search area is 0.4463. • For the P-sample nonmovers, the estimate of the overall residence probability for cases with links to census enumerations outside the search area is 0.5092. • Accounting for E-sample cases with a duplicate link outside the search area contributes to a decrease of approximately 1 percent in the estimate of correct enumerations relative to the original estimate. This is roughly 2.5 million additional erroneous enumerations due to census duplication. • Accounting for P-sample cases with a link to a census enumeration outside the searc