Kos: Back in the polling game! Teams up with Public Policy Polling (PPP).

[jefferson_dem]

Back in the polling game!
by kos
Thu Aug 05, 2010 at 01:22:38 PM PDT

It's been a rough month since the Research 2000 fraud was exposed, but we're now ready to jump back into the polling game.

We've settled on two separate pollsters -- one to do our election horserace polling, and one to do our weekly State of the Nation national polling.

We're still not ready to announce the latter, since we're still dealing with some last minute negotiations. But we can announce our horserace pollster: Public Policy Polling (PPP).

We obviously wanted a credible firm to excise the ghost of R2K, but we also wanted an accurate one. Well, according to Nate Silver's pollster rankings, PPP was among the most accurate. Strip out the firms that focus on a single state (like U Cincy, Field, etc), and it's top 10.

But Nate penalizes PPP for not being part of an industry-wide transparency project. Mark Blumenthal stripped away that penalty, and here were the results:

SUSA couldn't do horserace polling because of exclusivity contracts with other media organizations. ABC/Washington Post can't be hired out, obviously. Field Poll only does California. Blum and Weprin is out of business. Ciruli and BRC are regional Mountain West firms, the Ohio Poll is obviously just Ohio, Selzer owns Iowa and little else, while Montana State is obviously a Montana outfit. I'd never heard of Davis & Hibbits and I wanted to work with a firm I already knew and trusted.

So that left Mason-Dixon and PPP as the top horserace pollsters we could hire.

PPP, right off the bat, was game to releasing the raw data -- something NO other pollster has ever done. And I've got to admit, I love the work they've done.

We'll be announcing our other pollster in a few weeks. They have also agreed to release raw data. We still have to figure out HOW we'll do that, and in what format. There's also a technological challenge in integrating data in different formats from the two pollsters. So we won't have full functionality and raw data access available right away.

But long term, having access to the raw data will also allow us to do some pretty awesome shit, like creating widgets that allow you guys to insert your own weighing to reshape the numbers to reflect how YOU (or anyone else) thinks the electorate will look like. Access to raw data will also give everyone an unprecedented look at how pollsters massage data and better educate the public as to the art of polling, and its inherent limits.

I'm so excited about all of this I can barely contain myself. While the R2K mess has been a nightmare, it has opened up new possibilities -- the ability to work with some of the most accomplished pollsters in the biz, and break new ground by providing unparalleled transparency.

We'll ramp up slowly, but we hope to be back up to pre-scandal polling frequencies by September.

First up next week? Delaware. Then Missouri.

http://www.dailykos.com/storyonly/2010/8/5/890614/-Back-in-the-polling-game