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.
9 Analysis of Long and Short Screener Differences In the telephone samples for Airports 1 and 2, half of the addresses without matching telephone numbers were mailed a short screener survey that requested the telephone number (Appendix B); the other half were mailed a longer screener survey that requested the telephone number (Error! Reference source not found.) which also contained the same questions included on the Mail Survey Instrument (Appendix D). The overall telephone response rates for addresses without matching telephone numbers that were mailed the long and short screeners are very similar: 9.8% for the long screener and 9.6% for the short screener (Ï^2= 0.1, p-value > 0.30). The components for those response rates, however, differ for the two forms. Overall, 29.2% of the households mailed the long screener returned it, but only 17.9% of the households mailed the short screener returned it (Ï^2= 24, p-value < 0.0001). After the mail screener was returned, however, the pattern reversed: only 34% of the households that returned the long screener provided a telephone interview, while 54% of the households that returned the short screener provided a telephone interview (ð2 =12, p-value < 0.001). The discrepancy in response rates for the long and short screeners is largely explained by the fact that a substantial number of respondents to the long screener did not provide a telephone number (Table 12). A total of 136 out of the 194 households (70%) returning the long screener included the telephone number on the form, while 105 out of the 117 short screener households (90%) provided the telephone number. The conditional telephone response rates for the households providing a telephone number on the screener are 48% for the long screener and 60% for the short screener (ð2 =3.5, p-value = 0.06). Table 12 Number of unmatched telephone cases by screener type Long Short Number Mailed 746 746 Number Mailed Minus Number PNDs 664 665 Number Returned Mail Questionnaires 194 117 Number Providing a Phone Number 136 105 Number of Completed Telephone Interviews 65 63 There was high consistency between screener and telephone responses to the question on aircraft annoyance for the 103 households11 who returned the long screener and also responded to the telephone survey. Overall, 38 of the 46 respondents who reported being highly annoyed on the long screener also reported being highly annoyed on the telephone survey; 53 of the 57 respondents who reported not being highly annoyed on the screener also reported not being highly annoyed on the telephone survey. Note that 7 of the 12 households where there was a discrepancy in the results had multiple adults, so it is possible that different persons in the household responded to the screener and telephone questionnaires. 11 These included the original households without a matching telephone number plus additional households that were sent the long screener because their âmatchedâ telephone number was found to be invalid. 28
