Pollster.com

Lunctime Status Update for 12-16

Mark Blumenthal | September 17, 2008

The last 24 hours brings us 25 new statewide polls, all from one source: American Research Group (ARG). Their surveys, all conducted between September 7 and 16, involve 600 interviews in each state. This polling fete -- conducting 15,000 interviews in 10 days using live interviewers -- has produced the usual queries to my inbox? "Could they really be doing that many calls?" "Who is paying for all that polling?"

More on that below. We have also logged new surveys from Rasmussen Reports in New York, PPP in Virginia and the Field Poll in California. The bottom line is that the new ARG surveys shifted our classifications for two states:

• New Mexico moves back from toss-up to lean Obama, an apparently temporary shift resulting partly rom an outlier odd Zogby Internet survey released earlier in the week (see my discussion here).

• West Virginia moves from strong McCain to toss-up on our map. The issue here is that we have had only three general election polls all year in West Virginia, with today's new ARG survey being the first since June.

Unfortunately, the West Virginia scenario -- very few polls, big gap since the last poll -- shows up a shortcoming in the regression trend estimates we run. Ordinarily, those estimates have the virtue of mashing up results from many polls, which tends to minimize the contribution from outlier results. In this case, however, the estimate is based almost entirely on the ARG poll. Since that has McCain with just a four-point lead (49% to 45%) lead within sampling error, the map lists West Virginia as a toss-up.

The net result is that Obama gains back 5 electoral votes from New Mexico and McCain loses the 5 in West Virginia, bringing the total to Obama 243, McCain 219, with 76 in the toss-up category. (And yes, we know that our map is has been displaying the erroneous number 81 for the toss-up category. It's the result of a data entry glitch that we fixed and should be cleaned up momentarily).

Also, the daily tracking results have shown a modest shift to Obama in recent days and those new numbers now show a reversal in the recent trend, with Obama's estimate on our national chart (46.3%) now moving ever so slightly ahead of McCain (45.0%)

As for ARG's funding, here is the way ARG's Dick Bennett responded in May 2007 to our question about who pays for their surveys:

We still rely on subscriptions to what used to be commonly known as omnibus surveys.** The difference, however, is that we package each state and national survey separately. Unlike a traditional omnibus, there is no set schedule or guarantee of a survey without repackaging. We do a lot of panel building and have been fortunate in the past 6 months because we have been tracking household telephone, cable, and Internet access around the country and that allows for the addition of the political questions. Campaigns cannot be subscribers (but I know that some campaigns receive the latest numbers shortly after our subscribers and before public release). We are always looking for new subscribers willing to allow the political questions as part of the surveys. Subscriptions are not inexpensive -- no $199 for the entire race deals -- and subscribers pay proportionately for each survey.

**An omnibus survey is one that allows multiple paying clients to buy questions.

By Mark Blumenthal | September 17, 2008 1:37 PM | Permalink | Comments (11) | TrackBacks (0)