Polling for caucus participants is challenging for a variety of reasons including the fact that such a small percentage of the voting age population actually participates. Also, with such small numbers of actual participants, campaign organization techniques can have a significant impact on the actual turnout. Weather and other factors can also influence the final participation and outcome.

With this in mind, it is important to review all polling on caucuses with caution and an understanding of who is being polled.

At Rasmussen Reports, we begin in the same way that we poll for general elections by calling a randomly-selected group of households in a manner that includes all geographic regions of the state. For the Iowa caucus poll, we then asked three screening questions.

First, we ask about voting history and only include those who say they vote all of the time or most of the time.

Second, we ask how closely the respondent has followed the Presidential debates and other campaign activities. Those following the campaign Very or Somewhat Closely are included..

Third, we ask specifically about how likely each person was to participate in the caucus. Those who are Very or Somewhat Likely to participate are included in the final sample.

While these three questions established the overall sample, we reviewed the data with a variety of alternative screening assumptions. For example, we looked at data only for those who are following the Campaign Very Closely and are Very Likely to participate in the Caucuses.

Return to Democratic Results.

Return to Republican Results.

We also looked at an even tighter screen by including by asking the following question… Many times, things come up that make it impossible for someone to participate in the caucuses. This year, the event will be held much closer to the holidays and winter weather is also a possibility. Are you certain that you will participate in the Iowa caucuses or is it possible that something might come up to prevent you from attending?

As we examined each scenario, the results shifted modestly but the overall results remained quite consistent. Clinton was ahead in every Likely Caucus Participant model of the Democratic race with Obama and Edwards close to each other in second. On the Republican side, Romney was ahead in every Likely Caucus Participant model of the Republican race with Thompson and Huckabee close to each other.

Overall, the sample for Democrats represents approximately 15% of Iowa’s voting age population. The tightest screening model we used represents approximately 7% of the voting age population.

For Republicans, the sample included 11% of the state’s voting age population. The tightest screening model represented just under 5% of the voting age population.

For both the Democratic and the Republican samples, the very tightest screening model showed the frontrunner with a modestly smaller lead than in the overall sample.

As part of the overall study, we also asked some people who were screened out of the final sample their preference. When looking at a slightly larger pool of voters, we again saw the same overall dynamic in both the Republican and Democratic campaigns.

The final sample of 1007 Likely Democratic Caucus Participants was 51% female and 49% male. Sixty-eight percent (68%) were over 50 and 8% under 30. Forty-nine percent (49%) say they are politically liberal while 31% describe themselves as moderate. Ten percent (10%) describe themselves as Evangelical Christians. Sixty-six percent (66%) are married and 22% have children living at home. Twenty-two percent (22%) are first-time caucus participants.

For Republicans, the sample of 650 Likely Caucus Participants was 64% male and 36% female. Fifty-seven percent (57%) were over 50, 9% under 30. Eighty-one percent (81%) were politically conservative. Thirty-five percent (35%) 35% identified themselves as Evangelical Christians while another 35% said they were Protestant. Seventy-seven percent (77%) were married and 32% have children living at home with them. Twenty-eight percent (28%) of the sample are first-time caucus participants.

Return to Democratic Results.

Return to Republican Results.