In 1982 the Census Bureau requested the Committee on National Statistics to establish a panel to suggest research and experiments, to recommend improved methods, and to guide the Census Bureau on technical problems in appraising contending methods with regard to the conduct of the decennial census. In response, the panel produced an interim report that focused on recommendations for improvements in census methodology that warranted early investigation and testing. This report updates and expands the ideas and conclusions about decennial census methodology.

This open access book describes the differences in US census coverage, also referred to as “differential undercount”, by showing which groups have the highest net undercounts and which groups have the greatest undercount differentials, and discusses why such undercounts occur. In addition to focusing on measuring census coverage for several demographic characteristics, including age, gender, race, Hispanic origin status, and tenure, it also considers several of the main hard-to-count populations, such as immigrants, the homeless, the LBGT community, children in foster care, and the disabled. However, given the dearth of accurate undercount data for these groups, they are covered less comprehensively than those demographic groups for which there is reliable undercount data from the Census Bureau. This book is of interest to demographers, statisticians, survey methodologists, and all those interested in census coverage.

The usefulness of the U.S. decennial census depends critically on the accuracy with which individual people are counted in specific housing units, at precise geographic locations. The 2000 and other recent censuses have relied on a set of residence rules to craft instructions on the census questionnaire in order to guide respondents to identify their correct "usual residence." Determining the proper place to count such groups as college students, prisoners, and military personnel has always been complicated and controversial; major societal trends such as placement of children in shared custody arrangements and the prevalence of "snowbird" and "sunbird" populations who regularly move to favorable climates further make it difficult to specify ties to one household and one place. Once, Only Once, and in the Right Place reviews the evolution of current residence rules and the way residence concepts are presented to respondents. It proposes major changes to the basic approach of collecting residence information and suggests a program of research to improve the 2010 and future censuses.

The change in the way the federal government asked for information about race in the 2000 census marked an important turning point in the way Americans measure race. By allowing respondents to choose more than one racial category for the first time, the Census Bureau challenged strongly held beliefs about the nature and definition of race in our society. The New Race Question is a wide-ranging examination of what we know about racial enumeration, the likely effects of the census change, and possible policy implications for the future. The growing incidence of interracial marriage and childrearing led to the change in the census race question. Yet this reality conflicts with the need for clear racial categories required by anti-discrimination and voting rights laws and affirmative action policies. How will racial combinations be aggregated under the Census's new race question? Who will decide how a respondent who lists more than one race will be counted? How will the change affect established policies for documenting and redressing discrimination? The New Race Question opens with an exploration of what the attempt to count multiracials has shown in previous censuses and other large surveys. Contributor Reynolds Farley reviews the way in which the census has traditionally measured race, and shows that although the numbers of people choosing more than one race are not high at the national level, they can make a real difference in population totals at the county level. The book then takes up the debate over how the change in measurement will affect national policy in areas that rely on race counts, especially in civil rights law, but also in health, education, and income reporting. How do we relate data on poverty, graduation rates, and disease collected in 2000 to the rates calculated under the old race question? A technical appendix provides a useful manual for bridging old census data to new. The book concludes with a discussion of the politics of racial enumeration. Hugh Davis Graham examines recent history to ask why some groups were determined to be worthy of special government protections and programs, while others were not. Posing the volume's ultimate question, Jennifer Hochschild asks whether the official recognition of multiracials marks the beginning of the end of federal use of race data, and whether that is a good or a bad thing for society? The New Race Question brings to light the many ways in which a seemingly small change in surveying and categorizing race can have far reaching effects and expose deep fissures in our society. A Volume in the Russell Sage Foundation Census Series Copublished with the Levy Economics Institute of Bard College