Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: Rokke

Mickey Kaus should be given credit



he actually called the LA TIMES polling director last Thursday.

http://slate.msn.com/id/2102054/

"...The LAT Does It Again: Matthew Dowd, Bush's pollster, blasted the latest LAT poll (showing Kerry up 7) as a "mess." Specifically, he noted in an email sent to NBC and ABC that

Bush is leading independents by three, ahead among Republicans by a larger margin than Kerry is ahead among Dems, and we are down by seven. Outrageous. And it gets worse. They have Dems leading generic congressional ballot by 19. this means this poll is too Democratic by 10 to 12 points.

Who's right? Ask Governor Gray Davis! O.K, thats a cheap shot. But LAT-watchers have been skeptical of the Times Poll ever since it alone showed Davis closing to a virtual dead heat in the recent California gubernatorial recall--a report that virtually everyone else (including rival campaigns and the rival Field Poll) scoffed at. I've been told, however, that Times polling director Susan Pinkus is a straight shooter, so I did the irresponsible thing and postponed sniping while I called her up. [Don't let this happen again--ed] Here's what I learned:

--The party breakdown in the LAT poll was 38 % Democratic, 25% Republican, 24% Independent. That's about the same as the 38/19/26 breakdown of a year ago, but it's a big increase in Democrats since March of this year, when they were only 33 percent of the sample. Pinkus argues her latest numbers are not that different from a recent ABC poll that she said showed Democrats with a 37/27 percent edge. And she says her overall horse-race result isn't much different from the latest Gallup poll, which had Kerry up 6 in a three-way race. (That was among "likely" voters. The Times surveyed "registered" voters--and Gallup only had Kerry up by 3 in that broader group.)

--On the gigantic Democratic generic Congressional-preference lead in her survey, Pinkus said, "I don't know what's happening with that. If that's true, it's huge. ... I've seen it 5 or 6 points, but never 19, it's true." She said she stood by her poll, however. (Earlier she had noted that one out of 20 polls will be wrong, given the accepted margins of error.)

--Other commentators (such as RCP's T. Bevan) have hung their critique on Bush's much better showing in the Times' separate, more intense look at three battleground states. Can Bush really be losing nationally by 6 points and still be winning Missouri by 11 points? Seems unlikely. One possible explanation: The Times apparently used a different telephoning outfit to conduct the state-by-state polls than it used for the national poll. Might not something in the different survey techniques of the two firms have skewed the results in two directions? "I don't know. I can't answer that. That's a legitimate question," said Pinkus. If there is a difference in the results of the two survey techniques--even using the exact same questions--then which technique is more accurate? Maybe the Times' technique really does skew results to the left, no? (That would explain a lot!) Or its subcontractor's technique might skew results to the right. It could be something very simple. If--speaking hypothetically--all the Times' phone surveyors were Latinas with exacting NPR-style Spanish accents, those surveyed might try to please them by appealing to their assumed Democratic leanings. They might get a different result than would a survey conducted by men with thick Southern accents and gruff manners. One group would get it wrong.

P.S.: Note that the generic Congressional result was also much more "normal"-- and much less pro-Democratic (6-8 points instead of 19)--in the three "battleground" states surveyed by the subcontractor. Why would Missouri, Wisconsin and Ohio all happen to be almost equally less pro-Democratic than the nation as a whole? Or is the difference really a difference in the techniques of the subcontractor that did the "battleground" state surveys?

Does Pinkus plan to post the LAT's methodological numbers, as requested by ABC's Note? "Not at this time ... I guess I could but I haven't thought about it." She points out that she has given ABC a response to Dowd's charges. (Check The Note tomorrow.)
"


19 posted on 06/15/2004 7:46:57 AM PDT by Pikamax
[ Post Reply | Private Reply | To 5 | View Replies ]


To: Pikamax

So who has a REAL poll, one that reflects the mere 17%-19% admitted liberal population?

I bet the individual campaigns have the real numbers. These public polls are push or propaganda polls.


22 posted on 06/15/2004 7:51:30 AM PDT by longtermmemmory (VOTE!)
[ Post Reply | Private Reply | To 19 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson