No abstract

An experimental test was made of whether completion rates for a census questionnaire could be improved by offering the option of calling a toll-free number and providing the requested information to an interviewer as an alternative to mailing it back. Conducted on a national probability sample of households, five treatment panels were created to examine the effect of delivering, in different ways, the invitation to respond by telephone or mail. Offering such invitations did not improve completion rates except when included with a follow-up letter that did not also include a replacement questionnaire.

As the public policy uses of U. S. census data have expanded in recent decades, census undercount has become a contentious public issue. Concern centers on the fact that persons that are economically and socially disadvantaged are omitted at higher rates than others. In this paper we outline some of the contributions which sociologists can make to the undercount debate. First, the uses of census data are reviewed, with emphasis on how coverage errors affect social science research. Next, a conceptual model of the census enumeration process is offered, and its social system and census process components are described.

No abstract

No abstract