Thursday, May 10, 2012

Obama is losing, so the Drive-By Media Cooks Some Polls

It's happening, friends. The Drive-By Media is releasing cooked poll after cooked poll to show President Obama as stronger than he really is in head to head polls against Mitt Romney.  We talked about it in late April when the offender was NBC.  This time we have two offenders: Reuters/Ipsos and Tipp Online. Both have Obama beating Romney. Both have skewed poll samples to favor Obama.

Important to note is what these two polls do to the Real Clear Politics average.  (For those of you from Palm Beach County, FL, what Real Clear Politics (henceforce referred to as RCP) does is take all polls in a given election and give a straight average, whether the poll is of adults, registered voters, or likely voters.) What these two polls have done is given the President a slim lead in the RCP average of 46.2% to Romney's 46%. (So close it's a statistical tie.) However, this includes our two doctored polls. More on this later.

Before we go into the two polls, let's once again set the baseline in this country for conservatives, moderates, and liberals. As I've said so many times before, this is far more accurate than naming party affiliation, because so many people in the South especially are still registered as Democrats but vote conservative...holdovers from a past where a conservative Democrat was a real thing. However, it is reasonable to presume that conservatives will vote Republican and liberals will vote Democrat, leading the moderates as the group who may or may not vote down their party line or be a swing voter.

So, in 2011, Gallup did, as in past years, conduct a series of polls to establish the ideological demographics of this nation. In January 2012 Gallup published their National Ideology Poll.  At that time, 40% of Americans self-identified as conservative, 35% self-identified as moderate and 21% self-identified as liberal.  So a legitimate assumption would be to presume 40% of Americans to lean Republican and only 21% to lean Democrat. Even if we assume 20% of those moderates are true swing voters (as conventional wisdom proclaims) and split the remaining 15% with the two parties as moderate Democrats and moderate Republicans, that means 47.5% Republican leaners and 28.5% Democrat leaners. (Translation, they need ALL the true swing voters to win.)

The first offending poll is Tipp Online.  Tipp shows the President beating Mitt Romney 46% to 43% amongst registered voters (not likely voters, registered voters).  Even Tipp admits that "Race Tightens, Obama Lead Over Romney Narrows To 3 Points" in their headline.  But there's a problem with their polling sample. They gave Democrats a 7% advantage! You have to look at the very bottom of this poll to find it, but when Tipp asked the individuals polled which way lean politically, 38% self-identified as Democrat, 31% self-identified as Republican, and 27% self-identified as independent. 

That's a 7% advantage simply handed to the that does not exist in reality!  The only way to justify this poll is to claim, erouneously, that 65% of all moderates lean Democrat while only 35% lean Republican.

The second offending poll is Reuters/Ipsos. The poll shows Obama with a mildly comfortable lead of 49% to 42% over Mitt Romney (not likely voters, registered voters). There's just one problem...they went and handed a 9% advantage to Democrats in this poll! Whooops!  Their polling sample was 47% Democrat leaners to 38% Republican leaners! Holy cow! Once again, this does not match up with the real electorate and their political ideologies!

The only way to justify this poll is to eroneously claim that 74% of self-identified moderates lean Democrat. Nice try, but no dice. This does not add up to any reality.

So back to that RCP average that we talked about. Let's take the other polls involved, including two 3-day rolling average polls in Gallup and Rasmussen; as well as two polls of likely voters rather than registered voters in Rasmussen and Democracy Corps (ironically, these guys are a Democrat polling organization yet they're being more honest).

Rasmussen has Romney defeating Obama 49% to 44%, a 5% margin of victory (2% above the margin of error). Gallup has Romney defeating Obama 47% to 44%, a 3% margin of error (exactly the margin of error).  Democracy Corps has a tie of 47% to 47% for both candidates.

These three polls have an average of Romey 47.75% to Obama 45.5%.  This means Romney averaging a +2.25% margin of victory.  Compare this to the RCP average including the cooked Tipp and Reuters polls, which shows Obama beating Romney 46.2% to 46%.  So by publishing cooked polls, these two organizations push the RCP average from Obama losing to Obama winning.

Friends, the Drive-By Media likes to do this to dishearten conservatives, to make us give up on winning the White House. They want to make you believe that Obama will be re-elected so you won't work hard to defeat him. But what polls like this show is that they are very worried. They are trying to convince you (and themseleves) that Obama is a stronger candidate than reality shows. We must not let him. We will defeat Obama in 2012!

No comments:

Post a Comment

All posts will be reviewed subject to the Rules for Commenting. Any post that does not abide by these rules will not be posted, entirely at the discretion of the blog editor.

Commenters who repeatedly violate these rules will be permanently banned from commenting, and thus none of their comments, regardless of content, will be posted.