Monday, May 18, 2009

Polling in Virginia

Over the weekend the Washington Post had a story about how they haven't been running any of the poll numbers on the Democratic primary for Governor essentially because they're taking the paternalistic attitude that their readers need to be 'protected' from automated polls.

Rather than being offended by this I was actually quite amused. Perhaps 10 or 20 years ago it would have been a real problem for PPP if our numbers didn't get run in the Washington Post but the fact of the matter is people who want to know what the polls are saying are finding out just fine. Every time we've put out a Virginia primary poll we've had three or four days worth of explosion in traffic to both our blog and our main website. And folks are able to see all of the data from the polls and decide for themselves what it means rather than just getting whatever the Post filtered from it. There's been wideranging discussion of each of the polls throughout the Virginia blogosphere so basically this is just another case of a newspaper ceding their once upon a time authority to the internet. No complaints from us on that front.

What I do find interesting is the lack of methodological rigor behind the Post's choice not to run automated poll results, given that's essentially what they're accusing our polls of lacking. They have no statistical evidence pointing to a lack of reliability in IVR poll results, just a couple anecdotes from campaign pollsters about how they don't use it. They say more research is needed on the validity of automated polls.

Well here are some places to go:

-The National Council on Public Polls broke out all the 2004 public polling by mode and found IVR slightly more accurate than live interviewers.

-The AAPOR report on what went wrong in the primaries last year found no difference between the accuracy of IVR and live pollsters.

-Wall Street Journal analysis of active swing state pollsters from last year showing Survey USA and PPP more accurate than equally prolific live interviewer competitors.

There's much more out there but at this point I'm not sure the media outlets that refuse to print IVR polls really care about the track record. Numerous election cycles in a row now have shown that automated polls are just as accurate as others. Kudos to Hotline and Roll Call which have looked at that and changed their policies recently, but there are some places I'm not sure are ever going to change.

What I find most disappointing is the fact that the Post isn't conducting its own polling on the race. It's a good chance for them to put their money where their mouth is and prove their superiority to us and Survey USA. They might be right, but we'll never know if they don't try.

1 comment:

DavidNYC said...


Web Statistics