Getting "Residence" Right
The mechanics of census-taking have changed dramatically since 1790, when marshals set off on horseback to conduct the nation's first headcount. Yet every decennial census has adopted the same basic goal of counting individuals only once, and at their "usual" place of residence.
However, "usual residence" can be extremely difficult to define and measure, especially in today's highly mobile and diverse society. Some people have ties to multiple places, such as children in shared custody arrangements, and others lack connections to any fixed one. In 2000, the Census Bureau used 31 formal residence rules to try to account for all possibilities. But those rules were too complicated -- and often hard to apply, says a recent National Research Council report. Instead, the bureau should use core principles to determine residency for the 2010 census, and the bureau should improve how it communicates them to respondents. The agency also should study ways to collect data on how individuals are connected to dwellings that are not their usual residences.
The panel that wrote the report suggested several core principles. Among them, individuals living in the United States, including non-U.S. citizens, should be counted where they live or sleep more than any other place. Also, a person's individual circumstances should be the basis for determining usual residence, instead of family relationships or group labels such as "persons in hospitals."
When living arrangements are not straightforward, the burden of deciding usual residence should be shifted from census respondents to the bureau, the report adds. Under this approach, people would provide information on ties to another residence so the agency could make the most accurate determination possible.
Another persistent challenge is collecting information from people who live in group quarters, including college students in dormitories and prisoners. Such residents should be approached and counted in the same manner as the general household population, the report says. Questionnaires should be distributed to and completed by them, or administered by enumerators. The bureau also should develop a special form to collect responses from a central administrator or from facility records when direct access to people living in group quarters is not possible or allowed.
Census data are used primarily to distribute political power through the drawing of congressional and state legislative district boundaries and to allocate funds for public programs. Having accurate census information is a component of good government, both researchers and policymakers agree. -- Vanee Vines
Once, Only Once, and in the Right Place: Residence Rules in the Decennial Census. Panel on Residence Rules in the Decennial Census, Committee on National Statistics, Division of Behavioral and Social Sciences and Education (2006, 376 pp.; ISBN 0-309-10299-5, available from the National Academies Press, tel. 1-800-624-6242; $52.00 plus $4.50 shipping for single copies).
The panel was chaired by Paul R. Voss, professor emeritus, department of rural sociology, University of Wisconsin, Madison. The study was sponsored by the U.S. Census Bureau.