Obama polls: Vulnerable, strong, or battered?
posted at 9:50 am on March 28, 2012 by Ed Morrissey
A slew of polls hit the wires today, most of which make no sense when compared with each other, and most of which don’t have the necessary sample data to analyze properly. That doesn’t mean we should ignore them, but let’s consider them as a group rather than individually.
FLORIDA: Obama 49 – Romney 42; Obama 50 – Santorum 37
OHIO: Obama 47 – Romney 41; Obama 47 – Santorum 40
PENNSYLVANIA: Obama 45 – Romney 42; Obama 48 – Santorum 41 …
President Obama leads former Massachusetts Gov. Mitt Romney in three key swing states, according to new Quinnipiac University polls released Wednesday, but Obama remains vulnerable in all three battlegrounds.
In Quinnipiac’s last collective look at the presidential race in the three states, in late November and early December, Obama trailed Romney in Florida and Ohio and he led by a statistically insignificant margin in Pennsylvania. Quinnipiac has polled each state since then, finding better numbers for the president. But while Obama has jumped ahead, he is under the critical 50-percent threshold in each state, and his lead in Pennsylvania remains inside the margin of error. Obama leads former Sen. Rick Santorum of Pennsylvania by larger, significant margins in each state.
Well, at least Quinnipiac includes demographic data and crosstabs, which is better than the next two pollsters we’ll review. There are a few problems with Quinnipiac’s samples, though. The D/R/I for Florida in the 2008 exit polls was 37/34/29, and in 2010 it was 36/36/29. Quinnipiac’s weighted D/R/I is 33/29/33, which is hardly predictive. They have the same D/R/I for Ohio, even though in 2008 it was 39/31/30 and in 2010 36/37/28. The Q-poll D/R/I for PA is 37/34/25; in 2008 Democrats did much better at 44/37/18, but it’s similar to 2010’s 40/37/23. So we have Quinnipiac using models closer to 2008 for Ohio and Florida (remember, these are weighted, so this is a deliberate choice), but 2010 for Pennsylvania, and that makes a difference in the results.
Plus, let’s not forget that head-to-head matchups at this stage reflect Republican division during the primaries. Obama doesn’t get to 50% in any of these polls against the Republican frontrunner, a position of weakness instead of strength. When the GOP unites behind one candidate, we’ll get a clearer picture of Obama’s relative strength.
Next up we have the Washington Post, which puts Romney and Obama head-to-head in a likeability contest — and then fails to provide the sample data:
Half of all Americans now express unfavorable views of Mitt Romney, a new high for the GOP presidential hopeful in Washington Post-ABC News polling. The deteriorating public impressions of the former Massachusetts governor foreshadow a significant obstacle for him as he tries to shift the focus of his campaign toward a potential match-up against President Obama.
Romney’s negative numbers have jumped around this election cycle, but the overall pattern is similar to his trajectory four years ago: As he became better known, his unfavorables shot up far more rapidly than his positive numbers. Negative impressions are up eight percentage points in the past week, nudging past the previous high, which occurred about the time Romney suffered a big loss in the South Carolina primary.
In the new poll, 50 percent of all adults and 52 percent of registered voters express unfavorable opinions of Romney, both higher — although marginally — than Obama has received in Post-ABC polling as far back as late 2006.
However, the biggest difference between Romney and Obama is on the other side of the ledger: 53 percent of Americans hold favorable views of the president; for Romney, that number slides to 34 percent. Positive ratings of Obama steadily improved over the course of his fight for the 2008 Democratic presidential nomination — something not evident in Romney’s ratings this time around, or last.
Er, sure. However, bear in mind that this isn’t job approval; it’s whether or not people like each of the two men. Obama’s 53% favorability is on the low end of the scale for the Post series, and more people strongly disfavor Obama (28%) than Romney (25%). But more importantly — especially for the WaPo/ABC poll — we have no idea how the sample is composed on partisan lines. Given the propensity of the WaPo/ABC poll to use samples where Democrats have double-digit advantages, the lack of transparency on this poll result renders this utterly meaningless.
More than two-thirds of Americans disapprove of the way President Barack Obama is handling high gasoline prices, although most do not blame him for them, according to a Reuters/Ipsos online poll released on Tuesday.
Sixty-eight percent disapprove and 24 percent approve of how Obama is responding to price increases that have become one of the biggest issues in the 2012 presidential campaign.
In the past month, U.S. fuel prices have jumped about $0.30 per gallon to about $3.90 and the Republicans seeking to replace the Democrat in the November 6 election have seized upon the issue to attack his energy policies.
The disapproval reaches across party lines, potentially spelling trouble for Obama in the election, although the online survey showed voters hold oil companies or foreign countries more accountable than politicians for the price spike.
The partisan split on the approval rating in this case may be moot. A majority of Democrats disapprove of Obama’s performance on gas prices (52%), but the biggest worry will be the 73% of independents now disapproving. Gas prices will continue to rise through the summer, which means that those numbers will worsen as the election draws nearer. These voters may want to blame the oil companies more than Obama, but if those prices continue to go up, Obama’s energy policies will get a lot more criticism as the summer proceeds. Thanks to the multiplying force of gas prices on the distribution chain, consumer inflation will go up and the economy will almost certainly slow down from its presently modest level, which could put a big dent in Obama’s approval ratings at the very worst time possible for an incumbent.
I suspect of all the polls presented today, this one has the most potential for predictive value.