Colorado Teeter-Totter Tips Ever-So-Slightly To Romney Following Denver Debate, Every Vote Vital; Amendment 64 Tightens
In an election today for President of the United States in Colorado, Mitt Romney and Barack Obama finish a whisker apart, with Romney nominally ahead 48% to 47% in a 2-way matchup, and with Romney nominally ahead 46% to 45% in a 3-way matchup that includes Libertarian Gary Johnson polling at 2%.
Today’s results reflect a 2-point shift to Romney when compared to a SurveyUSA Denver Post poll conducted 5 weeks ago. Then: Obama had a nominal 1-point advantage. Today: Romney has a nominal 1-point advantage. In between: Romney’s “47%” comments were publicized, which may have tipped the teeter-totter to the left, and Obama’s Denver debate was derided, which may have tipped the teeter-totter to the right. The 2-point shift reported here, poll-on-poll, likely masks a larger volatility. Any outcome in the fight for Colorado’s 9 electoral votes remains possible, 4 days till voting begins.
Given a 2-point shift in the overall numbers, few of the underlying demographic groups show dramatic movement. Exceptions: Among voters with a 4-year college degree, Obama had led by 5 points, Romney now leads by 5 points, a 10-point right turn. Among gun owners, Romney had led by 10 points, now leads by 22 points, a 12-point right turn. More typical: Romney is up 2 points among Independents, up 3 points among men, up 4 points among middle-income voters and among moderates, and up 5 points among voters under age 50. There is some offsetting movement to Obama, who among voters age 50 to 64, had been up 9, now is up 14, and among voters with some college education, where Obama had trailed by 10 and now trails by 2.
About 1 in 8 of those who watched the 10/03/12 Denver Presidential debate changed their mind after watching. Of the mind-changers: twice as many switched to Romney as switched to Obama. Specifically, in the pivotal group that switched candidate preference as a result of what they saw in Denver, 34% switched from Obama to Romney and another 25% switched from undecided to Romney, for a total of 59% who switched to Romney. Disproportionately, those who switched to Romney were men. 16% switched from Romney to Obama, and another 13% switched from undecided to Obama, for a total of 29% who switched to Obama. Disproportionately, those who switched to Obama were women.
Poll-on-poll, there is little movement in who voters trust to fix the economy (Romney, barely), who voters think would do better at reducing the federal debt (Romney), and which candidate is more in touch with the average working person (Obama). Voters still say that the President should focus on “jobs” ahead of all other issues, with “balancing the federal budget” still in 2nd place.
Amendment 64, which led by 11 points 5 weeks ago, leads by 5 points today. 48% vote Yes, 43% vote No. There has been an 18-point erosion in support among women, who 5 weeks ago favored Amendment 64 by 10 points, but who today oppose the Amendment by 8 points. There has been 14-points of erosion among those with a 4-year college degree, who 5 weeks ago favored the Amendment by 9 points, and now oppose it by 5 points. A similar but smaller shift occurred among upper-income voters.
Cell-phone and home-phone respondents included in this research. 725 adults from the state of Colorado were interviewed by SurveyUSA Tuesday 10/09/12 and Wednesday 10/10/12. Of the adults, 661 were registered to vote. Of the registered voters, 614 were determined by SurveyUSA to be likely to vote in the 11/06/12 election. 70% of likely voters, reachable on a home telephone, were interviewed using recorded-voice telephone calls to landline phones. Sample was provided by SSI. The youngest-male method of respondent selection was used on calls to landlines. 30% of likely voters, unreachable on a home telephone, were shown an HTML questionnaire on their smart-phone or other electronic device. Sample provided by United Sample. Responses were minimally weighted to U.S. Census targets for gender, age, race and region. Where necessary, answer choices were rotated to prevent order bias, recency and latency effects. SurveyUSA assigns to each question within the instrument a theoretical margin of sampling error, and while such error is useful in theory, and is commonly cited in the presentation of research results, sampling error is only one of many types of error that may influence the outcome of an opinion research study. More practical concerns include the inability to contact some, the refusal of others to be interviewed and the inability of still others to speak the language. It is difficult to quantify non-sampling errors. Timing: All interviews for this survey were completed prior to the Vice Presidential debate on 10/11/12.