Will TV News Producers Resist The Temptation?

http://news-releases.uiowa.edu/2007/october/102907poll-methodology.pdfThe University of Iowa's new poll doesn't seem to be all that much of a useful proxy for figuring out the shifting preferences of caucus goers. It's not a _bad_ poll -- just not very orthodox. Here's a refresher course. Some news outlets in Iowa, like Radio Iowa, won't cover U of I polls at all.)

But some enterprising Obama ally made sure to send this morning's results to Matt Drudge, who, through no fault of his own, probably doesn't read too much into methodology.

Each television network has a polling unit. Presumably, producers will consult their in-house experts before broadcasting this poll without caveats.

If the U of Iowa poll becomes a talking point for a cable TV host to say "the race is tightening in Iowa," then score one for Obama's communications nexus. The fact is, the race is already pretty tight there, as all three leading campaigns will acknowledge. Tighter than a tick, an old CBS News anchor might have said.

Comments

But some enterprising Obama ally made sure to send this morning's results to Matt Drudge, who, through no fault of his own, probably doesn't read too much into methodology.

When did Drudge ever let the facts get in the way of linking to a story? But this is a clever move that will only help the Obama camp--whether the tip came from the campaign or not, is another matter. Still clever.

Posted by Cindrella Ferret | October 29, 2007 12:01 PM

marc, what about the last fox news poll that showed a national gap of only 17 points between HRC and BHO. I didn't see too much coverage of it but was wondering if its methodology was significantly different/ flawed. If not, it might be the first data point of a turn around.

Posted by pt | October 29, 2007 12:01 PM

"It's not a _bad_ poll"

Very nonjudgmental of you, Marc. But, of course, it really is a _bad_ poll.

Polling the caucus is all about how you deal with likely voters, since turnout is so miniscule. And the U of I polls deals with LV's in the least sophisticated manner possible.

That said, as an Edwards supporter, I'm happy to have Iowa expectations get lower for him.

Posted by Petey | October 29, 2007 12:08 PM

Again, if you are going to be fair, most polls of caucus voters have methodology issues as it is hard to figure out who will caucus, the whole delegate thing, second choice thing, etc. But, the media generally only raise methodological issues when it's a good poll for Obama. They did the eaxct same thing when an ABC poll showed Obama a little ahead in Iowa. The basic media strategy is this: dissect Obama's good polls methodologically but do not for HIllary. Put Obama's polls in context, not HIllary's. Polarizing candidates who are scandal-prone are good for business and that is why the media shills for Hillary.

Posted by RKA | October 29, 2007 12:28 PM

Pulse Opinion Research

Iowa

Survey of 1,000 Likely Voters

October 23, 2007

1* I'd like you to compare two Presidential tickets. The first ticket is Hillary Clinton as President and an unidentified Southern male as Vice President. The second ticket is Al Gore as President and Barack Obama as Vice President.

Would a Clinton Presidency be more effective, equally effective, or less effective than a Gore-Obama Presidency at securing global cooperation on climate change with India, China, and Russia?

14% More effective
26% Equally effective
41% Less effective
19% Not sure

2* Is a Clinton White House more likely, equally likely, or less likely than a Gore-Obama White House to invade Iran without reason?

16% More likely
40% Equally likely
23% Less likely
20% Not sure

3*Great, is a Clinton Presidency more likely, equally likely, or less likely than a Gore-Obama Presidency to shut down detainee facilities in Guantanamo Bay?

17% More likely
48% Equally likely
14% Less likely
21% Not sure

4* Does a Gore-Obama Presidency have a better chance, equal chance or a worse chance than a Clinton Presidency of reducing global proliferation of loose nuclear weapons?

26% Better chance
43% Equal chance
12% Worse chance
18% Not sure

5* Fine, is a Gore-Obama Presidency more willing, equally willing, or less willing than a Clinton Presidency to risk alienating special interests by reforming Social Security?

23% More willing
42% Equally willing
13% Less willing
23% Not sure

6* Is a Gore-Obama Presidency more likely, equally likely, or less likely than a Clinton Presidency to reduce American dependence on foreign oil?

35% More likely
36% Equally likely
12% Less likely
17% Not sure

7*Okay, would a Gore-Obama Presidency be better, worse, or just as good as a Clinton Presidency for the environment?

46% Better
14% Worse
27% Just as good
14% Not sure

8* Is a Gore-Obama Presidency more likely, equally likely, or less likely than a Clinton Presidency to reduce the total number of abortions in the United States?

12% More likely
53% Equally likely
13% Less likely
22% Not sure

9* Next, would a Clinton Presidency have a greater chance, an equal chance, or a lesser chance than a Gore-Obama Presidency of changing America for the better?

19% Greater chance
36% Equal chance
29% Lesser chance
16% Not sure

10* Is Al Gore less honest, more honest or just as honest as Hillary Clinton?

31% More honest
14% Less honest
43% Just as honest
13% Not sure

NOTE: Margin of Sampling Error, +/- 3 percentage points with a 95% level of confidence.

The telephone survey of 1,000 Likely Voters was conducted by Pulse Opinion Research on October 23, 2007. Pulse Opinion Research, LLC is an independent public opinion research firm using automated polling methodology and procedures licensed from Rasmussen Reports, LLC.

Posted by Suetonius | October 29, 2007 1:05 PM

"Again, if you are going to be fair, most polls of caucus voters have methodology issues as it is hard to figure out who will caucus"

The U of I poll LV screen assumes there will be more caucus-goers than there are adult Democrats in the entire state.

Put another way, the poll is assuming that instead of the normal 5% caucus turnout, there will be caucus turnout in excess of 100%.

That's not methodology issues, that's methodology madness.

Posted by Petey | October 29, 2007 1:09 PM

Great boys
a11de90adb790f8b4a81eeac5ab74b75

Posted by Hello people | February 1, 2008 1:48 AM