Better, More Reliable Polling

Jerome beat me to the punch over in Breaking Blue, but I wanted to take a moment to highlight a really awesome project being put together by the folks over at Pollster.com. As you might have noticed, I'm a pretty big fan of the site -- I cite it often here at MyDD, and I'm over there multiple times a day. Simply put, there's no better place to get information on polling.

Pollster.com is embarking on a project that could fundamentally alter the way we watch American politics. I'm not making an overstatement here. The Pollster.com Disclosure Project seeks to open up the polling process so that a genuine methodological review can be applied to the publicly available political polling in the country.

Ever think that a pollster has too loose of a likely voter screen? Ever think that a pollster is weighting its sample in an improper way? Ever just think that the numbers coming out of a particular poll just don't seem right? Pollster.com is seeking to answer these and other questions, first in primary polling from the early states and then in national polling, so that we can actually see if the polling so often cited on the evening news and the morning newspapers is actually methodologically sound. And with an ever-present eye watching the way that these polls are conducted, it's not too much to hope that we see better, more reliable polling in the future.

Head over to Pollster.com to find out more.



Display:


Re: Better, More Reliable Polling (none / 0)

This sounds like a great idea.


Join the Iowa progressive community at Bleeding Heartland.
by desmoinesdem on Tue Sep 25, 2007 at 12:20:47 AM EST

Re: Better, More Reliable Polling (none / 0)

Great idea...


by JaeHood on Tue Sep 25, 2007 at 12:28:29 AM EST

Re: Better, More Reliable Polling (none / 0)

Fortunate Son!  Rock on.


by Trey Rentz on Tue Sep 25, 2007 at 12:51:55 AM EST

good idea (none / 0)

but do you really think it will "fundamentally alter the way we watch American politics?"  I think these are relatively minor issues.  Unless you think there is evidence that polls are off, by, say 10%, or that we will spend more time discussing methodology than the results of the polls.


New Jersey politics and news
by John DE on Tue Sep 25, 2007 at 09:41:49 AM EST

As someone (none / 0)

who is completely polling obsessed, I think this is a great idea.  Two specific things I would like to see addressed:
there are two instances where methodology interests me greatly:

1.  How are they identifying likely voters in Iowa? Exit polling shows about 65% of Caucus participants are first time participants, so determining likely attendance is difficult.
2.  What is the percentage of independants in New Hampshire polling.  Exit polling says about 40% of voters in the Dem primary will be independents.  Frequently I see far smaller percentages in polls (ARG eg).

As an aside, I tend not to use the pollingreport because they charge for access to state polls, while realclearpolitics does not.


by fladem on Tue Sep 25, 2007 at 10:53:22 AM EST


You are not logged in.

In order to post a comment, you must be logged in. If you have a member account, please log in to comment.

If not, you can make an account right here. It's quick and free.