Monday, November 05, 2007

Forces Slanting Polling Trends to Rudy's benefit/Romney's detriment?

OK, "conspiracy" is probably too strong. However, consider the following figures in early state polling courtesy of Real Clear Politics' and Pollster.com's 2008 Polling sites (other resources used: Electoral Votes from FEC website; Primary calendar from About.com):

Iowa--Jan 3rd (7 electoral votes) has been polled:
  • 9 times in the last 7 weeks.
  • 27 times in the last 6 months-- 4.5X/mo
  • Romney leads RCP average by 14%
Wyoming--Jan 5th (3 electoral votes--only a fraction of state's primary delegates up for grabs on this date though):
  • No polling that I could find
  • Conventional wisdom is that Romney will win here solidly (he just got 61% of vote in a straw poll last weekend)
New Hampshire--Jan 8th (4 electoral votes) has been polled:
  • 10 times in the last 7 weeks.
  • 27 times in the last 6 months-- 4.5X/mo.
  • Romney leads RCP average by 8%
Nevada--Jan 12th (5 electoral votes) has been polled:
  • 2 times in the last 7 weeks.
  • 6 times in the last 6 months--1X/mo.
  • No RCP average calculated . . . but Romney leads Rudy by 1% over last 6 polls (though many expect Romney to do much better in the caucus format than Rudy)
Michigan--Jan 15th (17 electoral votes) has been polled:
  • 2 times in the last 7 weeks.
  • 7 times in the last 6 months--1.2X/mo.
  • Romney leads RCP average by 5%
South Carolina--Jan 19th (8 electoral votes) has been polled:
  • 6 times in the last 7 weeks.
  • 19 times in the last 6 months-- 3.2X/mo.
  • Statistically a 3 way tie for first (Romney,Rudy, Fred)
Florida--Jan 29th (27 electoral votes) has been polled:
  • 11 times in the last 7 weeks.
  • 27 times in the last 6 months-- 4.5X/mo.
  • Rudy leads RCP average by 11%

Some observations:
  1. The traditional "Big Three" early states (IA, NH, SC) have a long history of being frequently polled and this remains the case.
  2. MI and FL seem like they should be polled similarly to me (the two largest states prior to Feb 5th). However, where Romney leads and is expected to win there is a paucity of polls but where Rudy leads and is expected to win there's over a poll per week (and nearly two per week over the last 7 weeks)
  3. There seems to be a polling firm "blind spot" to WY (completely) and Nevada (relatively) when these are two of the first 4 contests . . . and ("coincidentally"???) where Romney is leading polls or expected to win.

Now, moving on to Feb 5th ("Tsunami Tuesday") some of the polling trends borderline on comical. Here are all the states going that date:
Alabama, Alaska, Arizona, Arkansas, California, Colorado (caucuses), Connecticut, Delaware, Georgia, Illinois, Minnesota, Missouri, New Jersey, New York, North Dakota (caucuses), Oklahoma, Tennessee, Utah

New Jersey (15 electoral votes) has been polled:
  • 5 times in the last 7 weeks.
  • 12 times in the last 6 months-- 2X/mo.
  • Rudy leads by OVER 40%!!.
WHY IS NJ BEING POLLED MORE THAN MICHIGAN?!?!? Rudy's never led by less than 32% in any of these copious NJ polls. Who's wasting all this money polling a state that's never been close when it's just another Feb 5th state? Something's fishy there.

California (51 electoral votes) has been polled:
  • 5 times in the last 7 weeks.
  • 15 times in the last 6 months--2.5X/mo.
  • Rudy leads RCP average by 19%
CA is the biggest prize and so I see why it's polled frequently

Moving on:

Pennsylvania--April 11th (YES, that's over 2 months AFTER Tsunami Tues!; 21 electoral votes) has been polled:
  • 3 times in the last 7 weeks.
  • 11 times in the last 6 months-- ~2X/mo.
  • Rudy leading by about 22%
Again, why in the world is this state being polled more than MI!? Don't you think the results of the early states and the results of Tsunami Tuesday just might have a little effect on how Penn votes on April 11th? That's called wasted money.

Conclusions/Questions:
So, is this ample evidence that there is an over-abundance and over-representation of polls for states where Rudy leads strongly? Or is there an ignoring of non-traditional early states where Romney is leading? Is this an attempt to get repeated positive news/headlines for Rudy and/or to downplay/diminish Mitt's potential?

Why the huge disparity between MI and FL? Why more NJ and Penn polls than MI? Why is NV being largely ignored? Am I missing something? Am I doomed to become a "conspiracy theorist?"

Thoughts? Comments?

Jeff Fuller

5 comments:

Scott said...

I'm actually in agreement with you. There are powerful forces being stirred up against Mr. Romney... we've got to stay strong.

I think people underestimate how much winning all these early states is going to do.

PolticalJunkie2008 said...

A large group of Mitt supporters are organizing an event to collect pledges and raise money online for Mitt. We are having it on the 7th of December and the website is http://www.december7thformitt.com/

Keith Steurer said...

Dems know that Romney will beat them, so they are paying the big 3 news agencies to only hire and cite the polls that will keep him out of the nomination. It;s all about information control. Information is more valuable than money in politics!

Slick-Willy said...

Along those lines: CNN showed a poll a few weeks back that had Romney down in South Carolina. Unfortunately, the same firm that had done the poll they were using had a new poll w/Romney in the lead in SC that had been taken the day before. Rather than use the good news to proclaim a Romney surge (based on recent polling) they used 2-week-old data and acted as though it was news. This allowed them to present bad news for Mitt.

I'm not sure if it was extreme incompetence or a minor conspiracy. Either way it was unfortunate.

Daniel Caldwell said...

I did some of my own research to determine why there are so many polls in the east and none in the west. The other campaigns are paying polling firms to take polls in places where they are already leading in the polls, and this pulls up their averages in the polls. Nobody has commissioned a poll in Wyoming, so there are no poll results. The Romney campaign (or any other campaign) could pay $7000 to have 10 questions asked to 500 random voters in Wyoming, but they have not.
Here are three polling firm points of contact in case you can find somebody with the funds:
KellyM@moore-info.com
spolk@mclaughlinonline.com
daliresearch2000@aol.com

I would like to see the answers to 3 questions, but I do not have the cash:
Please select one of the following even if your main choice is not listed. Of the following 6 candidates, who would be your first choice? (names rotated, with option to repeat if necessary)
Rudy Giuliani
Ron Paul
Mitt Romney
Fred Thompson
John McCain
Mike Huckabee
Of the five remaining, which one is your second choice?
Of the remaining four, which one would you find least favorable?