PRINCETON, NJ -- Of the handful of extremely close U.S. Senate races this year, the battle over the Colorado seat being defended by Democratic Sen. Mark Udall is one of the closest, mirroring the sharply divided politics of the state. Forty-two percent of Coloradans in the first half of 2014 identified as or leaned Republican and 42% identified as or leaned Democratic.
 |
|
Discover how Americans rate their state in 2014, across key metrics |
Colorado's political environment may be a bit better for the Democrats now than in the three previous years when Republicans outnumbered Democrats in the state by a slight margin. But, as is the case nationally, the environment is significantly worse for Democrats now than it was in 2008 -- the year Udall first captured his seat with 53% of the vote after a decade serving Colorado in the U.S. House. That same year, Americans overall were significantly more likely to identify as Democrats, and Barack Obama won Colorado in the presidential election, with 53.5% of the vote.
Udall faces Republican Rep. Cory Gardner, a former member of the Colorado Statehouse who won the election for Colorado's 4th congressional district seat in 2010, beating an incumbent Democrat.
Coloradans' Approval Ratings of Obama Near National Average
While Obama may have aided Udall's 2008 bid, he is widely seen as a drag on Democrats' chances across the country given his much lower approval ratings in 2014 compared with 2008. In this respect, Colorado is no exception. Obama's job approval rating in the state averaged 41% between January and June of this year -- just slightly below the 43% average seen nationally over the same time period, and 11 percentage points lower than Obama's approval in Colorado in 2009.
A majority of Coloradans, 55%, disapproved of Obama's job performance in the first half of 2014. And given the stability in Obama's approval rating nationally since then, his disapproval rating in Colorado likely remains about the same.
As is usually the case in midterm elections, turnout will be a major factor in deciding the outcome of the Senate race in Colorado, and as a result, the campaigns are battling hard to raise issues that could help spur their bases to the polls. Gardner is doing his best to tie Udall to Obama generally, as well as to the Affordable Care Act, or Obamacare, specifically.
One potentially positive angle on the situation for Udall, however, is that Obama's job approval rating in Colorado is not significantly lower than the total percentage of Democrats in the state, suggesting that Colorado Democrats are fairly loyal to the party. This contrasts with Arkansas, Louisiana, and, to a lesser extent, Iowa, where Obama's approval ratings in each state have been running several points lower than the total percentage of Democrats and Democratic-leaners in each state. Thus, it would appear that Democrats running in these states -- Sen. Mark Pryor, Sen. Mary Landrieu, and Rep. Bruce Braley, respectively -- run a greater risk of forfeiting a certain amount of support from their own party members, particularly if they appear too closely linked with Obama.
So, while Gardner's strategy of linking Udall with Obama may be an effective way to get his Republican base to the polls, it could backfire a bit with Colorado Democrats, who like the president and who would view Udall's connection to him as plus. Also, while Obama may be as unpopular in Colorado as he is nationally, the state as a whole is a bit less conservative than the rest of the country. One in three adults in Colorado (33%) describe their political views as conservative, compared with 36% of adults nationally. A quarter of Coloradans vs. 23% of adults nationally identify themselves as liberal.
These political assessments are available in Â鶹´«Ã½AV's new assessments, which present data on 14 political, economic, and social measures for each of the 50 states. Each state's ratings are presented alongside the national average for the same time period, to provide an easy way to gauge whether the state is performing at, above, or below par on each.
Democratic Tilt Among Women in Colorado Could Aid Udall
Meanwhile, Udall has placed his bets on reproductive rights as the issue that will activate women on his behalf. Udall has heavily focused his offense on Gardner's opposition to abortion, as well as Gardner's positions on birth control and "personhood" for fetuses.
The reasoning behind targeting women is clear, given women's relatively strong Democratic orientation in Colorado, which is similar to women's political leanings nationally. Women in the state tilt Democratic by 46% to 37%, nearly as strongly as Colorado men tilt Republican: 47% vs. 37%.
While national issues will undoubtedly be paramount in who voters back for the U.S. Senate in Colorado, conditions within the state could affect the race, mainly because of the gubernatorial election happening at the same time. Colorado residents appear a bit more positive about their state's economy than are residents of the 50 states on average. For instance, Â鶹´«Ã½AV polling in the last half of 2013 showed 44% of Coloradans saying it is a good time to get a job where they live, compared with the 50-state average of 40%. And 59% said their state taxes are not too high, versus 47% on average across the states. And more broadly, residents in Colorado are highly positive about their state, with 65% describing their state as the best or one of the best states to live in, much higher than the 46% 50-state average.
The Democratic incumbent, Gov. John Hickenlooper, ought to be benefiting most from these sentiments, in which case the entire Democratic ticket, including Udall, could do better. However, recent local polling in the state has shown Hickenlooper at best tied with his opponent, and at worst 10 points behind. Thus, if the relatively positive views residents have toward their state aren't helping the sitting governor, it's not clear how they would help Udall as the incumbent senator.
Bottom Line
Udall was elected to the Senate in a strong Democratic year, and Gardner was elected to the House in a strong Republican year. Now they are fighting for the same Senate seat at a time when the population of Colorado is sharply divided, but with the political profile of the state closer to what was seen in 2010 than in 2008.
Accordingly, most of the recent horserace polls conducted in Colorado, roughly two months out from the election, have found the Senate race too close to call. But, of course, given that turnout in midterm elections is typically near 40%, the political profile of voters in midterms can differ markedly from the general public. Any edge either candidate can get in turning out his base, such as by activating conservatives or women, could be decisive.
Survey Methods
Results for this Â鶹´«Ã½AV poll are based on telephone interviews conducted Jan. 1-June 30, 2014, on the Â鶹´«Ã½AV Daily tracking survey, with a random sample of 88,802 adults, aged 18 and older, living in all 50 U.S. states and the District of Columbia. Results are also based on Â鶹´«Ã½AV's recent 50-state poll conducted June-December 2013 with a random sample of approximately 600 adults per state, aged 18 and older, living in all 50 U.S. states.
For results based on the 2014 sample of 1,706 Colorado residents, the margin of sampling error is ±2 percentage points at the 95% confidence level.
For results based on the 2013 sample of adults per state, the margin of sampling error is ±5 percentage points at the 95% confidence level.
For both the Â鶹´«Ã½AV Daily tracking poll and the 50-state poll, interviews are conducted with respondents on landline telephones and cellular phones, with interviews conducted in Spanish for respondents who are primarily Spanish-speaking. Landline telephone numbers and cellphone numbers are selected using random-digit-dial methods. Landline respondents are chosen at random within each household on the basis of which member had the most recent birthday. In the Â鶹´«Ã½AV Daily tracking poll, each sample of national adults includes a minimum quota of 50% cellphone respondents and 50% landline respondents, with additional minimum quotas by region. In the 50-state poll, each sample of national adults includes minimum quotas of cellphone respondents and landline respondents based on cellphone and landline use in the respective state. Samples are weighted to correct for unequal selection probability, nonresponse, and double coverage of landline and cell users in the two sampling frames. They are also weighted to match demographics of gender, age, race, Hispanic ethnicity, education, region, population density, and phone status (cellphone only/landline only/both, cellphone mostly, and having an unlisted landline number). Demographic weighting targets are based on the most recent Current Population Survey figures for the aged 18 and older U.S. population. Phone status targets are based on the most recent National Health Interview Survey. Population density targets are based on the most recent U.S. census. All reported margins of sampling error include the computed design effects for weighting.
In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.
For more details on Â鶹´«Ã½AV's polling methodology, visit .
