Is Public Policy Polling Cooking the Polls in Missouri Senate Race?

Public Policy Polling came out with a new poll in the Missouri U.S. Senate race between Senator McCaskill and Rep. Todd Akin. The poll shows McCaskill with a 6 point lead.

Raleigh, N.C.-- PPP's newest poll of the Missouri Senate race finds Claire McCaskill expanding her lead to 6 points. She's at 46% to 40% for Todd Akin and 9% for Libertarian Jonathan Dine. On our last poll of the race, in late August, McCaskill had led by only a single point.

When you look at the data from the August poll you do find that McCaskill was leading by 1 point.

Raleigh, N.C. – PPP's newest poll of the Missouri Senate race finds that Todd Akin is weathering the storm and the contest remains a toss up. Claire McCaskill leads 45-44, just a small change from our poll last week which found Akin ahead by a 44-43 margin.

The difference between the August poll and the October poll is Libertarian Jonathan Dine was included in the October poll. The Libertarian candidate according to Public Policy Polling takes away more votes from Todd Akin than Claire McCaskill.

Q9 The candidates for Senate this fall are Democrat Claire McCaskill, Republican Todd
Akin, and Libertarian Jonathan Dine. If the election was today, who would you vote for?
Claire McCaskill .............................................. 46%
Todd Akin........................................................ 40%
Jonathan Dine................................................. 9%
Undecided....................................................... 5%

Q10 (Asked only to Dine supporters:) If you had to choose between Democrat Claire McCaskill and Republican Todd Akin, who would you vote for?
Claire McCaskill .............................................. 19%
Todd Akin........................................................ 33%
Not sure .......................................................... 48%

The August poll question was:
  

Q6 The candidates for Senate this fall are Democrat Claire McCaskill and Republican
Todd Akin. If the election was today, who would you vote for?
Claire McCaskill .............................................. 45%
Todd Akin........................................................ 44%
Undecided....................................................... 11%

As you can see the question was set up differently by Public Policy Polling. So to say McCaskill is expanding her lead is misleading in that Dine was not included in the previous poll.

Moreover, the line of questioning in the October polling differs from the August polling. In August the question was asked:

Q8 Do you accept Todd Akin’s apology for the comments he made last week, or not?
Accept his apology.......................................... 53%
Do not ............................................................. 40%
Not sure .......................................................... 7%

In the October poll two leading questions were asked:

Q13 Do you consider Claire McCaskill to be ‘ladylike,’ or not?
She is.............................................................. 46%
She is not........................................................ 27%
Not sure .......................................................... 27%

Q14 Do you think women can become pregnant from being raped, or not?
They can......................................................... 93%
They cannot .................................................... 4%
Not sure .......................................................... 4%

It is evident that a majority are not sure or do not consider McCaskill to be "ladylike" while most think you can become pregnant if raped.

However, if the question on the possibility of getting pregnant from being raped was asked with the qualifier that 99% of women use a contraceptive, the results would change. The point being that the percentage or odds that a woman becoming pregnant from a rape would be low.

You can manipulate the data with qualifying questions.

As for the demographics in the October polling:

Q15 Would you describe yourself as very liberal, somewhat liberal, moderate, somewhat
conservative, or very conservative?
Very liberal ...................................................... 11%
Somewhat liberal ............................................ 14%
Moderate......................................................... 29%
Somewhat conservative.................................. 27%
Very conservative ........................................... 20% 

Q16 If you are a woman, press 1. If a man, press 2.
Woman ........................................................... 47%
Man................................................................. 53%

The demographics in the August polling:

Q10 Would you describe yourself as very liberal, somewhat liberal, moderate, somewhat
conservative, or very conservative?
Very liberal ...................................................... 8%
Somewhat liberal ............................................ 12%
Moderate......................................................... 33%
Somewhat conservative.................................. 29%
Very conservative ........................................... 19%

Q11 If you are a woman, press 1. If a man, press 2.
Woman ........................................................... 55%
Man................................................................. 45%

Bottom line in the Missouri race is voters in Missouri want to see the U.S. Senate change from a Democratic majority to a Republican majority, a majority think McCaskill is not "ladylike" or not sure she is "ladylike" and Romney will win the "Show Me State" but has lost some to Obama.

August:

Q3 If the candidates for President this year were Democrat Barack Obama and Republican Mitt Romney, who would you vote for?
Barack Obama................................................ 41%
Mitt Romney.................................................... 53%
Undecided....................................................... 6%

October:

Q3 The candidates for President are Democrat Barack Obama and Republican Mitt Romney. If the election was today, who would you vote for?
Barack Obama................................................ 45%
Mitt Romney.................................................... 51%
Undecided....................................................... 4%

The question has to be asked why the change in demographics and the addition of Jonathan Dine?

Update: Public Policy Polling responded to this thread via twitter:

PublicPolicyPolling‏@ppppolls

@tlamb775 This one skewed too? http://www.rasmussenreports.com/public_content/politics/elections/election_2012/election_2012_senate_elections/missouri/election_2012_missouri_senate …

 

View conversation Hide conversation

• Reply RetweetedRetweet Delete FavoritedFavorite

 

The fact of the matter is Rasmussen's previous poll had McCaskill up by 10 points however as it was pointed out by the folks at the Huffington Post:

 

The numbers from Rasmussen are in sharp contrast with a poll taken Monday by the Democratic firm PPP, which showed Akin still ahead by a point a day after he said women were unlikely to become pregnant from "legitimate rape."

The reply given by the folks at Public Policy Polling is odd given that the latest Rasmussen poll has Akin closing the gap to within 6 points. This trend is in direct contradiction to the trend found in the latest polling data put out by PPPPolling.

Moreover, according to RCP, the Senate race in MO went from McCaskill to toss up. The PPPPolling should be considered an outlier when you look at trends. Adding to the trend is the story that polling done by Wenzel Strategies shows Akin ahead.

So again, is Public Policy Polling cooking the books on the Missouri U.S. Senate race?