Search results

The United States Census Bureau is vitally concerned with various aspects of the problem of nonresponse-its biasing effect on survey estimates, characteristics of nonrespondents, & differences between uncooperative & unavailable households. The trend toward increasing refusal rates has prompted investigation of the causes of refusals, & especially whether the increase in refusals can be traced to the implementation of the Privacy Act of 1974. A project was undertaken to study the characteristics of first-time refusals in the Current Population Survey (CPS). The CPS sample design calls for a household to be interviewed 8 times-in sample for 4 consecutive months, out of sample for 8 months, & then return to the sample for the next 4 months. Thus, the households involved are those that, regardless of month in sample, refuse to be interviewed for the first time. Information about these households was obtained from CPS interviewers on a special form used from May to Aug 1977. During the 4-month field period, 1,262 forms were completed. The information from 1,012 of these forms was matched to the data for corresponding cases, available on the CPS data tape. Selected information about first month completed households was also obtained from the data tape for comparison purposes. Results of these comparisons show that refusals in households at which interviews have not been previously attempted occur among the older & more predominantly middle class portions of the sample. Geographically, urban dwellers & westerners are most likely to refuse. Awareness of the Privacy Act & knowledge that participation is voluntary were not primarily responsible for refusals to the survey. 4 Tables. Modified AA.

The last decade has seen dramatic changes in how U.S. society views and recognizes same-sex couples. U.S. Census Bureau employees, Nancy Bates and Theresa J. DeMaio, chronicle recent efforts taken by the Census Bureau to update and improve the measurement and counting of same-sex couples.

A report on the results of a split-ballot experiment conducted in 1987 to test two alternative versions of the decennial census long form, which were randomly assigned & self-administered in group sessions (N = 515 respondents). The order of race & Hispanic origin items was experimentally manipulated: the standard long form asks race, then Hispanic origin; the experimental form reversed the order of the items to reduce perceived redundancy, & to create a more restricted frame of reference for the race item. The objectives of the context manipulation were (1) to reduce item nonresponse for the Hispanic origin item, & (2) to reduce reporting of "Other race" by Hispanics in the race item. Objective (1) was met; objective (2) was met for Hispanics born in a US state, but not for immigrants. The results are interpreted as reflecting a process of acculturation that affects how Hispanic respondents apply US racial categories "white" & "black" in the census. 5 Tables, 21 References. AA