Telephone Undercoverage Bias of 14- to 21-Year-Olds and 3- to 5-Year-Olds. Contractor Report. National Household Education Survey Technical Report.

The National Household Education Survey (NHES) was conducted for the first time in 1991 to collect data on the early childhood education (ECE) experiences of young children and participation in adult education. Because the NHES methodology is relatively new, field tests were necessary. A large field test of approximately 15,000 households was conducted during the fall of 1989 to examine several methodological issues. This report analyzes data from the Current Population Survey to identify the extent of telephone undercoverage for 14- to 21-year-olds and 3- to 5-year-olds and bias related to undercoverage for estimates of school dropouts and ECE program participation. Methods for adjusting survey estimates to reduce this bias partially are developed and evaluated. Recommendations are given to improve sampling accuracy for both populations. For estimation of 14- to 21-year-olds in the NHES, it recommended that the mean adjusted poststratified estimator be used because it incorporates an additional smoothing over the within-cell adjusted estimator. Poststratification variables that are more closely related to household income should be considered for the NHES estimation phase, and the use of tenure in addition to or in place of some of the other poststratification variables may be useful in this respect. For estimation of 3- to 5-year-olds in the NHES, the poststratified estimator appears to perform reasonably well for the range of statistics available, and it is recommended for use with this target population. Problems concerning undercoverage bias due to households without telephones were not substantial. Fourteen tables and eight figures present field test findings. An appendix discusses the source and reliability of estimates. (SLD)