Mark Blumenthal | December 4, 2007
Topics: 2008 , Divergent Polls , The 2008 Race
According to the release describing the Iowa State University poll we linked to earlier today, that poll "presents a much different picture [of] the race than other recent polls." More specifically, four recent surveys from the Des Moines Register, ARG, Rasmussen Reports and Strategic Vision all show Barack Obama and Hillary Clinton running within the margin of sampling error of each other, with John Edwards just a few points behind. The ISU survey shows Clinton (at 30%) with a comfortable lead over Edwards (24%) and Obama (20%). Why the big difference?
As we noted earlier today, the poll was conducted from November 6-18, which makes its results older than the four most recent surveys on our Iowa chart. However, according to ISU political science professor Jim McCormick, who directed the poll, "the biggest explanation for that is the volatility that still exists among those people who are likely to caucus." A better explanation is the poll itself, which is very different than other recent Iowa caucus surveys.
One difference involves the sample sizes on which the results were based. These numbers were omitted from the ISU release. According to Sarah Nusser, director of ISU's survey center, the results were based on interviews with 287 likely Democratic and 241 likely Republican caucus goers. Those are much smaller samples than used on other recent polls. Smaller sample sizes make for greater random sampling error, a form of "volatility" that is about the poll's design, not the voters.
The biggest difference involves the sample. It was drawn from the Iowa Secretary of State's list of registered voters, but unlike every other Iowa poll that I'm aware of, ISU sampled only registered Democrats and Republicans, excluding the 39%
36% of Iowa voters with no party registration. [Update: Pollster and frequent commenter Nick Panagakis emails that the most recent numbers from the Iowa Secretary of State put the percentage of "no party" Iowans at 39% of active registrants and 39% of all registrants].
Here's why that omission is important: In the recent CBS/New York Times survey conducted in early November, registered independents were 19% of Democratic likely caucus goers and 13% of Republicans (and 19% of 2004 Democratic caucus goers, according to the network entrance poll). Among likely Democratic caucus goers, the independents were twice as likely to favor Barack Obama (37%) as registered Democrats (19%).
The survey release also includes a highly misleading passage:
Poll projects likely caucus night turnout
Applying the percentage of respondents who said they "definitely will attend" their party caucuses on Jan. 3 to the number of registered voters in each party. The ISU Poll projects a Democratic turnout of perhaps as high as 150,000 attendees, with a 95 percent confidence interval of 130,000 to 175,000. As many as 88,500 are projected among the Republicans, with a 95 percent confidence interval ranging from about 74,000 to 103,000.
Those statistics project favorably to past turnout and imply that the reported survey results represent those "definite" attendees. Unfortunately, the survey results that ISU reported in their release were not based on the respondents who said they will "definitely attend" their party caucuses. They were based on the larger subgroup of respondents who said they will definitely or probably attend. That larger group projects to 273,00 Democrats and 176,000 Republicans.
To be fair, as a percentage of all adults, the ISU samples of Democrats (12%) and Republicans (8%) are in line with most of the other recent surveys that have disclosed their incidence statistics to Pollster.com. But 20% to 25% of the respondents to those other surveys were independents. The omission of independents combined with the already dated field period helps explain why Obama scores lower in this survey.