Reinterview Results for the Parent and Youth Components of the 1996 National Household Education Survey. Working Paper Series.

The National Household Education Survey (NHES) is a telephone survey of the noninstitutionalized civilian population of the United States that collects data on educational issues that are best explored through contact with households rather than with institutions. In the 1996 NHES (NHES:96) the topical components of interest were parent and family involvement in education and civic involvement. This report examines errors arising from interviewing respondents in the Parent and Family Involvement and Civic Involvement (PFI/CI) and the Youth Civic Involvement (YCI) components of the NHES:96. The estimates from these components and every survey are subject to both sampling error and nonsampling error. Sampling errors, the differences between population values and the sample estimates that arise because data are obtained from only a sample population, are generally well understood, but nonsampling errors arise from a variety of sources and are more difficult to measure. Important components of nonsampling errors for the NHES:96 include coverage, nonresponse, and measurement errors. The reinterview program was designed to identify survey questions that were not reliable, quantify the magnitude of the response variance for groups of questions collected from the same respondent at different times, and to provide feedback to improve the NHES for future surveys. A sample of 1,808 interviews (854 parent and 954 youth) was selected for reinterview. In all, 806 parent and 892 youth reinterviews were completed. Overall, both reinterview analyses indicate that the impact of measurement error on the estimates is low to moderate, as measured by the gross difference rates. In addition, the net difference rates support the use of the gross difference rates as measures of response variance. The reinterviews served their major purposes of finding questions with high error rates and providing feedback to improve the design of future questions. An appendix contains the reinterview questionnaires. (Contains 10 tables, 1 exhibit, and 7 references.) (SLD)