Romney Leads Obama in Fox Poll; Trails in Most Others
Republican Mitt Romney leads President Obama 46-44 in the latest Fox News poll of registered voters.
Republican Mitt Romney leads President Obama 46-44 in the latest Fox News poll of registered voters.
Fox News (“Fox News Poll: Romney Edges Obama as Approval of President Drops“):
Republican Mitt Romney holds a slim edge over President Obama in a head-to-head matchup, a Fox News poll released Thursday shows. In addition, the poll finds the president’s job rating has dropped to its lowest point of the year.
In a presidential matchup, Romney tops Obama by 46-44 percent if the election were today.
As with every Romney-Obama matchup in the past six months, the race is so tight that it is within the poll’s margin of sampling error. This, however, is only the second time the Fox News poll shows Romney on top. The first time was November 2011, when he was also up by 2 percentage points.
The poll was conducted Monday through Wednesday. On Tuesday, Rick Santorum suspended his presidential bid — giving Romney a clear path to the Republican nomination.
More Republicans (42 percent) than Democrats (32 percent) or independents (34 percent) say they are “extremely” interested in the upcoming presidential election.
Even so, the strength of party support in the matchup is dead even: 85 percent of Democrats back Obama and 85 percent of Republican back Romney.
Among the highly sought after group of independents, the poll found 43 percent back Romney and 37 percent Obama. Nearly one in four independent voters (21 percent) is undecided or won’t vote for either of the major party candidates. Last month, independents split evenly between Obama and Romney at 40 percent each. In February, Romney had a 9-point advantage.
While it’s possible that the exit of Rick Santorum and the resultant shift of Romney as the perceived Republican nominee has altered the dynamic of the race, it’s more likely that this poll is flawed. Rasmussen Tracking, whose methodology most analysts consider biased towards the Republican Party, took a larger sample of likely voters during the exact same window and found a 45-45 tie. All the other recent national polls–granted, all taken while Santorum was still in the race–show a fairly substantial Obama lead:
Let’s wait to see how the other polls look over the next few days before proclaiming this an even race. While I fully expect the race to tighten over the summer–and to ebb and flow along with developments on the campaign trail and the larger news landscape–my strong sense is that Obama has a comfortable lead at the moment.
I doubt it’s flawed; if the real numbers are on the order of 47-44 Obama in the population, you’d expect a poll to come out with a slight Romney majority at least 10% of the time just due to sampling error.
I saw this this morning. In another Fox poll or it may have been the same one Obama’s appro fell by something like 8 points overnight. Yeah right.
@Chris Lawrence: Sure. But that’s a flawed result!
There’s a 5% probability that a poll will have the true results outside the reported margin of error, which in this case is 3 points. Now, this may be such a poll, or it may not.
Fox got their “Romney leads Obama” headline, I suspect they were going to get it one way or another.
“Romney
Leadsstatistically tied with Obama in Fox Poll; Trails in Most Others”FTFY
I’ve read that Fox shifted their polling demographics on this poll. They increased GOP voters 6% while decreasing Dems and independents. Why would they do this? Well we can all take a stab at it. If they felt they’ve been under polling GOP voters they should tell us that and explain why they’ve changed it.
@James: No, that’s a misunderstanding of polling. Yes, there’s a margin of error based on the sample size. But the topline numbers are the best indicator of what the poll shows, not simply random numbers within a sampling range. If Candidate A has a 3 point lead across-the-board in various polls, it would be incorrect to say “Well, he’s really just tied since they all have a margin of error of 3 or higher.”
@James Joyner: That’s a fair point. My polling professor liked to emphasize MOE over toplines. But the topline does have Romney in the lead.
@James Joyner:
But this isn’t a series of polls, this is just one poll.
With 95% certainty Obama’s result is between 41 and 47% and Romney’s result is between 43% and 50%.
@James Joyner:
I’m pretty sure the first James was correct. 46-44 +/-3 is a tie.
That is not “picking another result,” that is using the paired confidence information that comes with the estimate.
If some other stream of polls, also with estimates below their resolution of confidence show a tilt one way or another, the top level numbers cannot be used to make a composite result. You could, possibly, by integrating the underlying individual poll answers, but no one has done that.
I really dislike this form of national polling for Presidential elections across the board since these polls are a measurement of popular vote rather than Electoral College votes.
@Gromitt Gunn:
You would have to ask a lot more voters if you wanted to get a Electoral College votes poll rather than a popular vote poll. For the same kind of confidence levels as this Fox News poll you would be looking to ask somewhere around 40-50,000.
Which obviously would be a lot more expensive.
Another factor often missed in coverage of polls is that any intelligent campaign is going to pace themselves. It does no good to “win” in April, or June, if you lose in November.
Many commentators seem to believe there is no way that campaigns could have individual strategies laid out on a calendar, with their best attempt at an arc toward November.
Say you’ve got a crony of a candidate saying something dangerous, or a clip of the candidate himself being stupid. I’d guess that you’d want to drop those at different times. Some you might want to build in the public mind, and some you might want fresh in October.
Or, you might let your opponent have a post-convention wave, and then hit when that starts to bore.
Tying this back to the present, the Republicans have had a great deal of visibility, and basically used it badly. They did enough damage to themselves (Romney having to wear other people’s “war on women”) that the Democrats could sit back.
That doesn’t mean the Democrats got nutthin’
If the difference between two candidates is within the margin of error, the poll should be RESPONSIBLY reported as a statistical tie. There is a confidence interval associated with each poll, which is related to the number of people polled… and that number of people is a subset of a much larger voting populous that will actually decide the election (and would likely vote differently than 46-44 if the election were today).
If your poll is a statistical tie, then you look at the trend in your own poll of the same question over time (not accurate because Santorum has exited the race since previous polls) or you look at similar polls done by other organizations (not done because those polls contradict Fox’s results and Fox is ridiculously partisan).
To title the story “Fox News Poll: Romney Edges Obama as Approval of President Drops” misleads casual readers and Fox trolls. Obviously the network is trying to get the base to back Romney and continue to bash Obama… but I’m getting sick of party-affiliated news organizations, especially the hyperpartisan nonsense that’s coming out of that network this election cycle.
JP,
“Another factor often missed in coverage of polls is that any intelligent campaign is going to pace themselves. It does no good to “win” in April, or June, if you lose in November.”
Another related factor is that the last year, the vast majority of election related coverage and virtually all of the televised ads have been about the Republican primary. And while there were differences between the Republican candidates, there was substantial agreement on how bad Obama is. Once Obama starts running ads and the media spends more time on Obama’s message, I suspect the numbers will change in Obama’s direction.
@PJ: Oh, I understand that the costs increase and the timeliness decreases dramatically. I just find it disingenuous whenever anyone in the media treats popular vote presidential polls as though they are a true predictor.
That Fox could only come up with a +2% edge for Romney is interesting – maybe they’re trying to appear to be “fair and balanced”?
There is a long way to go.
If you don’t believe that libs overs sample polls about Obama, with libs, give me a thumbs down.
Breaking news: water is wet.
@Mark:
It this is the case it would account for what appeared to an overnight huge shift in Obama’s appro. At the end of the day it’s a given that Fox News is a propaganda arm of the GOP so it wouldn’t surprise me.
The only “poll” that matters is the one that takes place on the 1st Tuesday in November.
@PJ:
Well, that is worded rather misleadingly.
Obama’s result (in this poll) is what it is – 44%. The MOE means that if you sample the same population with the same methodology 100 times, then the result of that set will be 95 results between 41 and 47.
There is no assurance, of course, that this has anything to do with the real world. If the methodology is flawed, then you will get results that are at odds with reality, no matter how often you poll.
@Tano:
I wouldn’t use “will be”. If I would roll a 20 sided dice 100 times a number of times, I will not get five ones each time.
You’re right, the methodology might be flawed.
I don’t know that it would be that hard . I have fumbled and critted twice in a row many times.
The idea that Obama holds a “comfortable” lead is a bit silly; few leads are comfortable 7 months out from the election. Not even that 8-point IBD lead (which is of registered voters, which tends to favor Democrats).
I would certainly say that Obama is favored at this point, but there are quite a few things out there that could trip him up; poor job growth and high gas prices being just the most obvious. And if the election is a referendum on the last four years (as reelection campaigns tend to be), he’s going to have some problems.
Well that’s just it, isn’t it? The fact that, despite all these issues, the President is still favored to win reelection speaks volumes about the GOP in general and Mitt Romney in particular…
It’s worth remembering that it was Fox that called Florida for Bush in 2000, and that the guy who made the call is for Fox, John Ellis, is Bush’s cousin. He was also a vocal critic of Gore’s. Ellis spent much of that evening on the phone with GW & Jeb Bush.
No chance of a bias there, eh? Other networks quickly passed this on, and perception rapidly became reality.
I guess all these Republicans did not have any Who albums when they were growing up. If they had, they would understand the concept of “We won’t get fooled again”…
Of course polls are just a snap-shot of the time-frame in which they are taken.
It’ll be interesting to watch how this kind of $hit, from Fred Barnes and the WSJ, plays out over the course of the election:
The Etch-a-Sketch is shaking…
@anjin-san:
This is a really, really odd example in that all the networks erroneously reported very early in the evening that Gore had won Florida, leading most of us to assume that Gore would be president. Bush in fact won the initial count and the selected recount required under Florida law. Additionally, he also came out ahead in the non-binding media recount conducted by the New York Times and others. So . . . Fox was right.
@James Joyner:
There you edge into the broader implications of statistical ties.
In a statistical tie I don’t believe the “winner” should get all electoral votes. Fix that by direct vote, or proportional electoral counts. In this regard, Nebraska and Maine are rational.
Fox called it first. The other networks immediately started reporting that “Fox News has called FL for Bush”. In short order, everyone in the country was talking about the Bush win in FL. Fox was the first domino to fall for Bush.
And I think its really, really odd that you seem to have no problem with having a RELATIVE OF ONE OF THE CANDIDATES being the guy who makes the call.
What seems to get missed for the most part is that most of these polls are within the margin of error. In other words, they are statistically a dead heat. Further let’s remember that not so very long ago Rick Santorum was in a virtual tie with Obama.
What this comes down to is that those numbers are not going to change a great deal. Election day and the vector thereof will depend almost entirely on whether or not Obama is bad enough to make the challenger look good….At least, from an image perspective. I think the facts are that just about anybody looks better than Obama. But as we all know it all comes down to the image.
Obama supporters will feel much better knowing this is what you think. Some of us remember your perfect track record on the 2008 election – every prediction you made was not just wrong, but wildly, almost cosmically wrong.
Did you enjoy the “Democrat civil war in Denver”? The one that was going to be “worse than Chicago in ’68”. Of course we all remember your bold statement “Obama can’t win.” Then there were your breathless pronouncements of an 11th hour McCain surge that was goint to send him to the White House – you remember, your “special inside information that most people don’t have access to.”
The GOP clown car posse does not have to look far for a chief jester… they have you.
@anjin-san:
Your comment was the first I’ve heard of that and, given that it happened almost a dozen years ago, I don’t much care. If true, it’s an odd call, to be sure. But, again, he was right: Bush won Florida.
I say that thinking it quite possible that more people who voted in Florida INTENDED to vote for Gore than Bush (but didn’t because they got confused by the ballot used in a few counties) and that it’s a virtual certainty (because of the Nadar vote) that more people who voted in Florida preferred Gore to Bush.
JJ- go back and reread the media Florida recount- they counted applying 7 different standards-Gore came ahead in 4 Bush in 3.
Well, Fox plays a central role in the party you belong to, and I doubt there has been much in the way of internal reforms there, so I find it a bit curious that you simply don’t care. As for it being an established fact that Bush won in FL – there are quite a few millions of people in this country who might disagree.
We will never really know. Many parties are at fault, I also blame Gore for a too hasty concession and ham-handed attempts to skew the recount in his favor. That being said, I think “What happened then does not really matter and Bush won anyway”, is, at the very least, taking the easy way out.