home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!!SERVICES"
- X-Organization: Concordia University, Montreal, Quebec
- X-Envelope-to: stat-l@MCGILL1.BITNET
- X-VMS-To: IN%"stat-l@mcgill1"
- X-VMS-Cc: ILFC594
- Message-ID: <877A06788000F735@Vax2.Concordia.CA>
- Newsgroups: bit.listserv.stat-l
- Date: Tue, 1 Sep 1992 14:01:00 EDT
- Sender: "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
- From: "GEOFF SELIG - CONCORDIA U. COMPUTING SERVICES"
- <ILFC594@VAX2.CONCORDIA.CA>
- Subject: Re: Election polls
- Lines: 35
-
- > I am interested in what statisticians think about the current wave of
- > election polls. Specifically, how should the media and public make sense
- > out of them, particulary in light of the fact that they widely diverge, yet
- > claim to be based on valid random sampling techniques. [...] For example,
- > the latest ABC poll shows Clinton at 56 and Bush at 36, while the Time/CNN
- > poll shows Clinton at 48 and Bush at 40. Any ideas about the statistical
- > interpretations of such data are greatly appreciated.
-
- Each poll has a level of reliability (something like +-4% 95 times out of
- 100). What this means is that if we were to repeat the poll 100 times, the
- actual (population) percentage of voters supporting Clinton would lie in
- the range 56% +- 4% (or 52% to 60%) in the first poll and 48% +- 4% (or 44%
- to 52%) in 95 out of the 100 polls taken.
-
- As these two bands (marginally) overlap, we cannot say that the two poll
- results are statistically different from each other, even though the
- magnitude of the poll result differences appears to imply this! Thus,
- there is no (statistical) difference between Clinton's popularity in the
- two polls.
-
- Misinterpreting poll results is one of the ways the (ignorant) media
- manipulates the (statistically) naive reader into believing that popularity
- differences exist between candidates when they do not. Although this is
- not the case at the moment, we have all read newspaper reports announcing
- that candidate X is far ahead of candidate B based on a 3 or 4% lead in the
- polls.
-
- Let the reader beware!!!
-
- Geoff Selig Phone (514)-848-7666 | _, /| The Cat's Meow
- Academic Services FAX (514)-848-7622 | \ o.O'
- Department of Computing Services | =(___)=
- Concordia University (H928) | U
- Montreal, Quebec H3G 1M8 | ilfc594@vax2.concordia.ca
- | ilfc594@conu2.BITNET
-