On Tue, 3 Dec 2002, Olsen Chris wrote in part: > If my random digit dialing generates numbers with equal > probability, and assuming both L's and UL's would answer the phones > with equal likelihood, would I not get "answers" from Listed and > NonListed in the same relative proportion as they are in the > population?
Yes; in the long run. But a single sample is about as short a run as you can get, and the sample proportion of listed:unlisted may be very far from the population value thereof. One can only hope that this characteristic is largely unrelated to the information one wishes to extract from the data. (This of course does not speak to any of the other quite legitimate concerns Dennis has raised.) -- DFB. ----------------------------------------------------------------------- Donald F. Burrill [EMAIL PROTECTED] 56 Sebbins Pond Drive, Bedford, NH 03110 (603) 626-0816 [was: 184 Nashua Road, Bedford, NH 03110 (603) 471-7128] . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
