Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Appendix E. Sampling Design for Airport 3 Results from the Airport 2 and Airport 1 studies were used to modify the procedures used for the Airport 3 survey. The first modification was for the sampling of vacant addresses in Airport 3.14 These were sampled and considered as eligible for the first two airports. A total of 132 of the addresses in the Airport 1 / Airport 2 sample were coded as vacant. Of those, 8 households provided a response: 2 by CATI (one each in Airport 1 and Airport 2), and 6 by mail (5 in Airport 1 and 1 in Airport 2). The low yield from the vacant addresses in Airport 1 and Airport 2 led to a decision to exclude vacant addresses from the Airport 3 sample and replace them with additional occupied households in order to increase the yield for the sample. The final telephone response rates for households without matching phone numbers were approximately the same for the short and long versions of the screener (Appendix B and Error! Reference source not found. respectively). The long screener, however, had higher response rate at the mail phase with a lower percentage of households providing a telephone number on the form and a similar overall response rate. The long screener therefore provides partial information on annoyance for the households that do not complete the telephone interview while the short screener does not include this information. Since the final response rates were approximately the same, it was decided that the Airport 3 survey would use only the long form screener for unmatched telephone numbers. This has the added benefit of providing information on the relationship between the reported annoyance on the telephone survey and the reported annoyance on the screener within the same household. The estimated relationship between annoyance and noise exposure is similar for the mail and telephone surveys in Airport 1 and Airport 2 for noise exposures up to 65 dB. Above that noise exposure, however, there are too few respondents to the telephone survey to evaluate the relationship. For the Airport 3 survey, we adjusted our sampling to counter the previous deficiency and included a larger number of addresses at the high noise exposure levels to obtain more data at those values. 14 Vacant addresses are those that are classified as unoccupied on the U.S. Postal Service Computerized Delivery Sequence Files. Sometimes those addresses are eligible housing units, particularly in high-turnover areas. E-1