IN-Rasmussen: Romney up by 16

[Tender Branson]

51% Romney
35% Obama
  3% Others
11% Undecided

The Indiana online survey of 400 Likely Voters was conducted July 31-August 1, 2012 by Rasmussen Reports.  The margin of sampling error is +/- 5 percentage points with a 95% level of confidence. Field work for all Rasmussen Reports surveys is conducted by Pulse Opinion Research, LLC.

http://www.rasmussenreports.com/public_content/politics/elections/election_2012/election_2012_presidential_election/indiana/election_2012_indiana_president

[MilesC56]

And the Senate race is still tied...

[pepper11]

Maybe we should reconsider looking to the Indiana results as a good indicator on election night.

[a Person]

This seems like an outlier. I don't see why Indiana would have shifted ten points in the last two months.

[MorningInAmerica]

This seems like an outlier. I don't see why Indiana would have shifted ten points in the last two months.

An outlier based on what? To judge a poll an outlier, there has to be other polls for a point of comparison. What polling data on the presidential race out of Indiana are you using to judge this poll an outlier?

[a Person]

This seems like an outlier. I don't see why Indiana would have shifted ten points in the last two months.

An outlier based on what? To judge a poll an outlier, there has to be other polls for a point of comparison. What polling data on the presidential race out of Indiana are you using to judge this poll an outlier?

The previous Rasmussen poll, maybe.

[timothyinMD]

The true outlier was the 2008 election.  That isn't happening again, not even close.  Romney will win Indiana by 15-20 pts

[pbrower2a]

On-line. Easily manipulated. I use no on-line polls for anything.

[HagridOfTheDeep]

I'm more concerned with the sample size than the online nature of the poll.

[ajc0918]

Just curious how do they do online polls? Do you have to be on an email list? Or is it just a website you go to and take the poll?

[MorningInAmerica]

This seems like an outlier. I don't see why Indiana would have shifted ten points in the last two months.

An outlier based on what? To judge a poll an outlier, there has to be other polls for a point of comparison. What polling data on the presidential race out of Indiana are you using to judge this poll an outlier?

The previous Rasmussen poll, maybe.

Still not sure how TWO polls of a race is enough to deem one of them an outlier. Usually several are required. And we don't have that in Indiana, unfortunately.

[JulioMadrid]

The true outlier was the 2008 election.  That isn't happening again, not even close.  Romney will win Indiana by 15-20 pts

An election can be an outlier?? I'd believe a real election over an online survey.

[pbrower2a]

1* If the 2012 presidential election were held today, would you vote for Republican Mitt Romney or Democrat Barack Obama?

 

2* Do you have a very favorable, somewhat favorable, somewhat unfavorable or very unfavorable impression of Mitt Romney?

3* Generally speaking, how would you rate your own personal finances these days? Excellent, good, fair, or poor?

4* Are your personal finances getting better these days, or worse?

5* A proposal has been made to repeal the health care bill and stop it from going into effect. Do you strongly favor, somewhat favor, somewhat oppose or strongly oppose a proposal to repeal the health care bill?

 

6* If Mitt Romney is elected president and Republicans win control of Congress, is the economy likely to get better, get worse, or stay about the same?

7* If President Obama is reelected and Democrats regain control of Congress, is the economy likely to get better, get worse, or stay about the same?

8* Generally speaking, is Mitt Romney’s track record in business primarily a reason to vote for him or vote against him?

9* How would you rate the job Barack Obama has been doing as president… do you strongly approve, somewhat approve, somewhat disapprove, or strongly disapprove of the job he’s been doing?

Question 9 should be asked immediately before or after question 1.  Note that questions 2-5 or 5-9  may be rotated, but #8 is clearly a leading question.  

[pbrower2a]

I'm more concerned with the sample size than the online nature of the poll.

Any on-line poll is to be rejected. Internet access is further from universal than is having a land-line telephone. Poor and elderly people are much less likely to have internet access. It could be of course that a bunch of people who know about the poll volunteer what they believe.

Randomness of a sample is essential to valid polling.

[a Person]

This seems like an outlier. I don't see why Indiana would have shifted ten points in the last two months.

An outlier based on what? To judge a poll an outlier, there has to be other polls for a point of comparison. What polling data on the presidential race out of Indiana are you using to judge this poll an outlier?

The previous Rasmussen poll, maybe.

Still not sure how TWO polls of a race is enough to deem one of them an outlier. Usually several are required. And we don't have that in Indiana, unfortunately.

Well, there's also the one poll that's actually in the database.

Speaking of which, why is the other Rasmussen poll not in the database? Is it because of the Internet thing?

[Senator North Carolina Yankee]

I'm more concerned with the sample size than the online nature of the poll.

Any on-line poll is to be rejected. Internet access is further from universal than is having a land-line telephone. Poor and elderly people are much less likely to have internet access. It could be of course that a bunch of people who know about the poll volunteer what they believe.

Randomness of a sample is essential to valid polling.

I disagree. Yougov has made great strides at doing online polling, for instance. There are ways to do online polling accurately.

[HagridOfTheDeep]

I'm more concerned with the sample size than the online nature of the poll.

Any on-line poll is to be rejected. Internet access is further from universal than is having a land-line telephone. Poor and elderly people are much less likely to have internet access. It could be of course that a bunch of people who know about the poll volunteer what they believe.

Randomness of a sample is essential to valid polling.

I understand all that. I'd be inclined to believe a lot of those folks who would miss out on online surveys would actually vote Republican. So the online poll doesn't concern me much, as a Republican (not as a statistician).

What does concern me is the sample size, because it makes for a much bigger margin of error. What's the confidence level for this +/-5%? There's a reasonable chance that Obama is doing much better in this state that the poll would suggest.

[Jayhawker]

Romney will win Indiana by around 10 points. The reason it will swing more than the nation is that Obama won't waste time here in 2012, unlike when he blitzed the state in 2008.

[Bandit3 the Worker]

Maybe we should reconsider looking to the Indiana results as a good indicator on election night.

Maybe we should reconsider looking to Rasmussen as a good indicator on election night.

[mondale84]

The true outlier was the 2008 election.  That isn't happening again, not even close.  Romney will win Indiana by 15-20 pts

Umm, no.

[BigSkyBob]

Obama spent more in Indiana to run up the score in 2008. McCain spent his money trying to win. If he had lost Indiana, he was going to lose anyway. Given the 2008 was a one point affair, a tied race would start at a base of a six point advantage before adjustments. First adjustment, is the absense of an imbalance of resources in favor of Obama. The second adjustment is for the fact Romney is gaining among Whites more than the electorate as a whole. Since Indiana has more Whites than average Romney ought to do better than the uniform shift.

[Senator North Carolina Yankee]

I think about 56% is the ceiling for Romney in any reasonable scenario then can occur at this point. He isn't going do as well as Bush for various reasons.

I wouldn't be surprised to see GA, AZ and IN go about the same this time. Maybe even put MO and MT in that category as well. If gets 55% or 56% in all or most of them, he is winning. If it is 52%-53% then he is probably losing.

[Senator MaxQue]

I'm more concerned with the sample size than the online nature of the poll.

Any on-line poll is to be rejected.

Again, in Canada, in the last elections, the online polls were often the more exact.
Given than the internet access is pretty much the same in USA than in Canada, that isn't a reason for rejection. And pretty much all polls now here are online, now and they are still good.

Depends of the methodology used, in fact.

[timothyinMD]

Gallup's Obama approval tracking released this week shows Obama at 38% in Indiana. That's abysmal.  In the scheme of presidential elections, 2008 was an outlier for Indiana because it is not the start of some trend toward Indiana becoming a swing/D state.  It was a one hit wonder

[realisticidealist]

In the scheme of presidential elections, 2008 was an outlier for Indiana because it is not the start of some trend toward Indiana becoming a swing/D state.  It was a one hit wonder

Probably true, but Obama's not going to do worse than Kerry (39%) here.