Tag Archives: matt larkin

Are PPP’s Alaska polls pure propaganda?

lies The Democratic polling firm Public Policy Polling, also known as PPP, is once again proving that it values publicity above all else. And in Alaska right now, the interest is the looming race against U.S. Sen. Mark Begich. Throw in an off-the-cuff statement that Sarah Palin made about considering running for Senate, and you’ve got the makings for some screaming headlines, no matter that the numbers buried under the headlines are suspect, at best.

A recent poll that the firm conducted via robo calling shows support for Begich’s job performance at about 42 percent. That might make it sound that Begich is in a little bit of trouble. However, Alaska-based pollster Marc Hellenthal, who actually knows the state and uses real-live people to do his polling, has Begich’s approval rate at about 60 percent in a recent poll.

Hellenthal’s got some other numbers which contradict PPP’s poll, but before we get into all of that, let’s get the elephant out of the room: Palin might have considered running for Senate, in the same way that I have considered taking belly dancing classes. I’m not going to take belly dancing classes. Palin is not going to run for Senate. She would lose.

According to PPP, most Alaskans want Palin to run. That might be true, in the way that dogs think they want that car that they’re chasing. But from the headlines the poll has generated, you would think that Palin is still popular in Alaska. Palin is not popular in Alaska.

In fairness, PPP’s numbers show that she’s got a high disapproval rating here. And the firm can be forgiven for not understanding the intricacies of Alaska’s politics, nor is it responsible for a lazy media. What isn’t forgivable is that the group’s polling methodology stinks. In this particular poll, Democrats are way over represented, as are women and older people. Law dictates that pollsters aren’t allowed to robo-call cell phones, so there goes about 50 percent of the population that use them all the time or most of the time. And those are just a few of the problems with using machines to do the work that humans should.

Alaska, as uber poll cruncher Nate Silver points out, “is perhaps the most difficult state in the country to poll. Its residents are in a strange time zone and keep strange schedules; it has very high rates of cellphone usage; it has highly unusual demographics.” (It should be noted here that Silver himself has used Alaska PPP polls and therefore gotten Alaska horribly wrong in the past).

Couple Alaska’s idiosyncrasies with PPP’s sloppy work, and you might as well throw numbers on a wall.

PPP isn’t a stranger to using such suspect methodology. It conducted an absurd poll in May that showed that Sen. Mark Begich actually lost support in Alaska as a result of voting against Obama’s gun control bill. In that poll, Democrats were over represented by a whopping 9 percent, women were over-represented by 12 percent, and the firm just couldn’t figure out the nonpartisan/Alaska Independent Party thing.

They got it a little better on this one. But the sampling error is still way off. Dems in this poll are over represented by six percent, women by 10 percent, and the ages are all screwy. The numbers show it.

Hellethall’s poll was conducted for a private citizen who is not involved in any of the races.

Here’s a few examples of how Hellenthal’s poll numbers compare with PPP’s:

  • Begich has 60 percent approval and 24 percent disapproval rating. Prior to the gun control vote, Hellenthal had Begich at a 53 percent positive and 35 percent negative. In other words, Begich’s vote helped him enormously, which is in direct contradiction to what PPP reported.
  • Lt. Gov. Mead Treadwell has a 30 percent positive approval rating, a 16 percent negative rating, 30 percent had never heard of him, and the remaining are neutral about him. PPP reported a 29 percent unfavorable rating.
  • As for unannounced Senate candidate (DNR Commissioner) Dan Sullivan: PPP has his negatives at 28 percent, which is absurdly high for a commissioner. Hellenthal didn’t poll him, but the only thing that could figure is that people are confusing Commish Sullivan with Anchorage Mayor Dan Sullivan.
  • Gov. Sean Parnell’s approval rating is 57percent positive and 26 percent negative.PPP has those numbers at 44 percent and 42 percent respectively.
  • Both Hellenthal’s poll and PPP’s poll show that Democratic Sens. Bill Wielechowski and Hollis French would have a way to go in name recognition if either chooses to run for governor.
  • Rep. Don Young’s wetback comment didn’t seem to hurt him much. He’s got a 56 percent positive and 28 percent negative approval rating. PPP has Young at 47 percent approval 43 percent disapproval rating.

PPP didn’t poll on her, but another interesting finding from Hellenthal is Sen. Lisa Murkowski’s numbers. According to Hellenthal, she is the most liked politician in the state with a whopping 69-24 approval/disapproval rating. This number might just quell the squirrels who are chirping that Murkowski is so unpopular that she’s not going to run again in 2016.

Another pollster in Alaska, Matt Larkin, who has taken over for long-time Republican pollster Dave Dittman, declined to take on the PPP poll, saying that he didn’t want to get into the back and forth of who was a better pollster. He did say that generally bad polls “undermine the integrity of the whole field.”

He also reminded me how wrong PPP got the 2010 Senate race. In the last poll PPP conducted before the vote, it had Scott McAdams tied with Lisa Murkowski. It showed that Miller would take it by 7 points. On election night, Murkowski had 40 percent of the vote. Miller got about 35 percent and McAdams 23 percent.

Contact Amanda Coyne at amandamcoyne@yahoo.com