RE: Polls - a question for Mike

[rick.jacobi@tllrwda.state.tx.us (rick jacobi)]

         Reply to:   RE: Polls - a question for Mike 
Ruth asks a good question.  The Texas Public Interest Research Group, an
anti-nuclear group (among other things), ran a poll several years ago
where they asked: "Would you favor a nuclear dump in west Texas if you knew it
would pollute the Rio Grande?"  Not surprisingly, 82% of respondents said
they would not favor the "dump".  The PIRG issued a press release falsely
claiming that 82% of Texans were opposed to the proposed Texas waste
disposal facility.  A number of Texas papers ran the story, citing the poll
outcome, apparently without looking into the validity of the results. 
(What's really surprising is that 18% of respondents would favor a dump that
would pollute the Rio Grande!).  

Senator Wellstone, in the debate on the Texas compact in the U.S.
Congress last week, referred to the poll results in his remarks.  No one
challenged his claim, or had the time during the debate to verify the poll
results.

This is a common problem with the media.  There have been many times, in
my experience, where an AP or UPI reporter will garble a story, publish a
bogus number or quote an unfounded assertion and that will be picked up
and published in smaller, less sophisticated papers that use the AP/UPI
stories as backgrounders.  These will be published over and over again and
will surface years later as real numbers, doctrine or facts.

In fairness to AP and UPI, their reporters are usually seasoned
professionals who are able to see through these things and will take the extra time
to make sure that they aren't being misled.  And, in fairness to the
anti-nukes, I saw a nuclear industry poll a few months ago that claimed an
overwhelming number of persons in the United States supported nuclear power. 
I was a bit skeptical about the nature of that poll, as well.
-----------------------------


On 9/10/98, Ruth Weiner wrote:

     This is a serious question, and not an attempt to pull your chain.  
     You say you don't want to publish "unscientific" poll results.  How
do 
     you, or how does AP, determine whether a poll is scientific or not? 

     Suppose that an "activist group" issues a press release saying that 
     their polls show that 65% of those polled want to shut down a 
     particular nuclear power plant immediately.  Does AP investigate who

     was polled, what questions they were asked, in what sequence the 
     questions were asked, how those polled were selected, and so on?
     
     Clearly only my own opinion.
     
     Ruth Weiner
     rfweine@sandia.gov
************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html

************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html