Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!apple!bionet!hayes.ims.alaska.edu!accuvax.nwu.edu!nucsrl!telecom-request
From: gwangung@milton.u.washington.edu (Roger Tang)
Newsgroups: comp.dcom.telecom
Subject: Re: NY Times Method For Conducting Phone Poll
Message-ID: <13932@accuvax.nwu.edu>
Date: 22 Oct 90 15:09:25 GMT
Sender: news@accuvax.nwu.edu
Organization: University of Washington, Seattle
Lines: 23
Approved: Telecom@eecs.nwu.edu
X-Submissions-To: telecom@eecs.nwu.edu
X-Administrivia-To: telecom-request@eecs.nwu.edu
X-Telecom-Digest: Volume 10, Issue 756, Message 1 of 12


In article <13818@accuvax.nwu.edu> oplinger@sol.crd.ge.com (B. S.
Oplinger) writes:

!cmoore@brl.mil describes the process for a NY Times/CBS News poll:

!How pray tell can they have generated 'telephone numbers ... formed by
!random digits, thus permitting access to both listed and unlisted
!numbers' and then caused them to be 'screened so that only residences
!would be called?' Is there some magic way to tell if a number is
!residential or commercial, especially the unlisted ones. Or is this
!simply a case of a newspaper article mixing facts and fiction?

![Moderator's Note: I think they made the assumption (mostly correct)
!that business phones would probably not be non-pub;

	No, they don't do this.  Based on my days as a telephone
survey taker (NOT a solicitor!), we ASKED (or added two plus two when
somebody answered, 'Joe's Pizza.').

	There's all sorts of methodological screening techniques; some of
theme are quite sophisticated.  This one, however, really just requires
the brains of an avocado.