Pollster.com

Articles and Analysis

 

POLL: SurveyUSA National


SurveyUSA

National
Clinton 48, McCain 46
Obama 46, McCain 46

UPDATE: SurveyUSA will also release 50 statewide surveys testing general election match-ups for McCain vs. Obama and McCain vs. Clinton later today.

 

Comments
John McIntyre:

Interesting poll results. Sounds like Obama is loosing his edge and Hillary has a comeback. Not sure, Obama's charisma survive too long ! I voted for Obama but now it makes me to think - I made a mistake. He projected himself as a Judgement wizard on everything - but lately he shows up like a pretender and clown. He is weak on details and all I saw was marketing belony. It is shame to see his involvement with Rezko, secret discussion Canadian official and his church priest involved with Farrakhan. Anyway I wasted my vote.

____________________

joshua bradshaw:

Susa has a survey for the General based on states and shows an electoral map. From my calculation based on their maps Clinton would Win 277 electoral votes and the Presidency and Obama would win 269 electoral votes and lose the presidency to McCain. However I am a little wear of this because it shows obama winning Utah and North dakota which I do not think will happen. It also shows clinton losing orgeon and washington which i do not think will happen. what are your guy's thoughts

____________________

K. Threlfall:

John, take it easy...for Rezko, it is at least comforting that Obama can admit making lousy decisions (unlike others); check the latest news concerning "Naftagate" (I love that one) and the seeming complicity of the Clinton campaign; and as for Farrakhan, Obama has repudiated his endorsement (again, unlike others). Oh yeah, and check your spelling...

____________________

re: Survey USA releasing 50 statewide surveys.... i'm too excited to speak. oh, the anticipation! and i'm not even being sarcastic. pollster.com, the gift that keeps on giving.

____________________

Shane:

"Susa has a survey for the General based on states and shows an electoral map. From my calculation based on their maps Clinton would Win 277 electoral votes and the Presidency and Obama would win 269 electoral votes and lose the presidency to McCain. However I am a little wear of this because it shows obama winning Utah and North dakota which I do not think will happen. It also shows clinton losing orgeon and washington which i do not think will happen. what are your guy's thoughts"

Well, by the survey proper, both win the presidency. I guess if you change the results of the survey, then Obama loses and Clinton wins, but, if you're going to do that, why use the survey at all? I can easily reply that "Obama will not lose PA or New Jersey in a general election."

____________________

Andrew S. in California:

First John look at the other poll. As the first poster on there said, "nothing like mixed signals." Second Joshua its way too early to make a prediction based off of now to predict who will win what states. One point though is that Obama has been far better than Clinton at taking purple states. Obama and Clinton have been pretty equal at getting purple states. I think we need to factor in a good VP candidate to balance weak areas for the ticket as well. I do think that people factor that in passively.

____________________

a. smith:

I voted for Clinton and I think I have wasted my vote big time. I was just hearing her say that she has lifelong experience and McCain has life long experience but Obama does not. After hearing that I really can't vote for Clinton in November. She is just dividing up the democratic party and destroying our chance to win in November. I also thought she would publish her tax return just to put that thing behind us but it seems she cares little about transparency and honesty. I wish I didn't vote for her. She will be no different than McCain. At least McCain can say he means what he says no matter how bad it is.

____________________

Shane:

Joshua Bradshaw,

I just ran the numbers again, and, A) It doesn't show Obama getting Utah, B) They both win the presidency by the same number of electoral votes. Where do you get your information?

____________________

joshua Bradshaw:

Actually shane by susa's map Obama would fall 1 electoral vote short of the Presidency. And I was not changing the results I was mearrly asking if you guys think that it is realist to think that obama would win North Dakota or Utah. this is simpily a question about how reliable a poll this far out can be.

____________________

Andrew S. in California:

Ugh strike that sentence about Obama getting purple states better. lol

____________________

Rasmus:

Well, maybe Utah will vote against McCain because he kicked out Romney?
Don�t know, I must crunch the numbers first...
But I love you, SurveyUSA! 30,000 interviews, that is GREAT!

____________________

Rasmus:

Well, maybe Utah will vote against McCain because he kicked out Romney?
Don�t know, I must crunch the numbers first...
But I love you, SurveyUSA! 30,000 interviews, that is GREAT!

____________________

Andrew S. in California:

Shane I think you need to learn a bit of geography. Obama won't win Utah on Susa's map he'll win Colorado which can be considered a purple state for him.

____________________

Andrew S. in California:

Also being originally from Washingtonian and a partime Oregonian :p I can easily see at least Washington going to McCain over Clinton. Our state is full of independents and of course that is McCain's bread and butter.

____________________

Shane:

Andrew,

I said that he didn't win Utah on SUSA's map. Where's the confusion? Don't insult me before making sure you're checking your information.

Josh,

I ran the numbers twice.. maybe we're using different delegate calculators?

____________________

Shane:

Josh,

Maybe you're still confusing Colorado and Utah. They show him to win Colorado, not Utah, and Colorado is a bigger state in terms of population.

____________________

SUSA is showing Obama winning Colorado not Utah. That sounds relistic.
By the maps both win the Presidency
Obama 278
McCain 260

Clinton 275
McCain 263

____________________

Mark K:

Since the Republicans selected McCain, who is seen as more independent-minded by many voters, a good number of Democratic voters might switch over and vote for McCain in the general election rather than the other Democratic candidate if their preferred candidate doesn't win. Among the Democrats, so far it seems that many younger voters and independents (especially in the west) might rather vote for McCain than Clinton in the general election. At the same time, many older voters (especially in the rust-belt states) and Latinos might rather vote for McCain than Obama in the general election. I would be really interested to see how a combined ticket of Clinton-Obama and Obama-Clinton would compare when paired against McCain. Of course, it would be hard to conduct such a national poll and get fair results until McCain selects his running mate, but I would be very interested in seeing how they match up. Just some thoughts ...

____________________

K. Threlfall:

Concerning the Survey USA maps...according to my math Obama would win 278 - 260 over McCain, while Clinton would win 276 - 262 over McCain. The map shows Obama apparently winning the swing states of Colorado and Nevada, but NOT Utah. Anyway, it's easy to error when all these numbers are involved, but this is what I find...

____________________

You gotta love this quote from Survey USA's web page:

"It will take us a couple of hours this morning to get the data linked to the maps, but the maps are real, and show exactly what happens in an election 'today.'"

"Exactly." Riiiiight. No reported margin of error. No likely voter model (which would be screwy this far in advance, anyway). This is an interesting diversion, but don't put a lot of stock in it.

____________________

Rasmus:

"which would be screwy this far in advance, anyway"
I cite your citation of SUSA:
"...what happens in an election 'today.'"

They don�t need to add a MoE, they said that the sample size is 600 per state, so the state numbers have a MoE of 4%.

It would be interesting to see their total numbers weighted for the state population (take all 600 california respondants and add random respondants of the smaller states comparing the smaller states population with that of CA)

You�d get a sample size of about 5,000 with that, with a MoE of 1,4%- that would be pretty interesting

____________________

Rasmus:

To the election map: I don�t think Obama will lose NJ and PA; I also don�t think that Clinton can win FL and WV.

It�s important, in my view, that Michigan would go Obama today, but not CLinton.

It will be really interesting to see the TX, OH and MO numbers- and all the others, too.

Hm, do your HTML-tags work?
does not in my preview...

____________________

Andrew S. in California:

My bad it was directed at Joshua. The quoting is what's getting confusing :p

____________________

John - Spokane, WA:

John, At least your man enough to admit you made a mistake voting for Obama - I admire you for that. A. Smith, Never voted for Hillary - thats just a rhetoric post. This will be a learning experience for Obama. Perhaps by 2016 or so, he may make a good President - But not right now, he has more growing to do AND he better stay away from people like Rezko, Farrakahn and the like. Just like I said before on this site - The ALL TALK EXPRESS can only go so far, he must bring something of his own to the table. It looks like FL & MI will hold their "DO-OVERS" in June. There is NO RULE thats says that cannot do that. This will put Hillary over the top along with the Supers. The Texas Caucus didn't have enough teeth to help Obama - Caucuses are a joke overall, evrybody knows that ( Caucus from the Spanish root word CACA). They never reflect true results due to the vast disorganization, alienation of certain voters and so on.

____________________

Shane:

People, can we discuss the polls and the arithmetic rather than moronically throwing catchphrases at each other?

I'm looking at you John, and, to a lesser extent, A. Smith.

____________________

John - Spokane, WA:


Shane,

Polls require explanations, thats the message - Im sorry you dont get that.

____________________

Rodney:

About those Clinton tax returns. The obvious reason Hillary wants to wait as long as possible to publish them is that, with those huge book contracts she and Bill both have had and with Bill's income from speeches, they are quite wealthy. It might deter contributors, if they think she is in the position to fund a good portion of her campaign. It's just practical to keep their income out of the public eye, as long as possible, to help her raise funds. This is a non-issue, and a weak one that will just make Obama look foolish when they publish their returns after April 15, just before the April 22 Pennsylvania primary. Bill and Hillary are so much better skilled. They are just letting Obama go on and on with this non-issue, so that he looks like he has nothing when they publish the returns. This is another one where Obama's inexperience and lack of political skill shows. Let him come back in four to eight years, with some seasoning, and we'll fall in love with him all over -- but for better reasons.

____________________

John - Spokane, WA:


Rodney,

Your a man of sensablilty and wisdom, very well written post and your absolutely right.

____________________

Bree:

Good one Rodney! I laughed pretty hard at that one! Too bad Hillary has no chance in 8 years or even now. HAHAHAH!

____________________

joshua bradshaw:

Okay I have tried to post serval times but they are not going up. I made a mistake and switched nevada and utah for some odd reason. I believe that i may have old electoral numbers could someone post the number of electoral votes for each state or a link to where I can find them please. I am sorry for mixing up nevada and utah.

____________________

Ramus:

600 registered voters per state. Again, Riiight. Did SUSA stop polling at 600 respondents to hit 600 exactly on the spot for each state? I hope not since that would suggest possibly severe sampling bias from non-response. More likely, the 600 is an overall overage per state, with some variation among states.

We also know that all registered voters will not vote, and many other people will register between now and November. So, a registration-based sample is not "exactly" (their words) reflective of what will happen in November nor what would happen in the counterfactual that the election is held today.

This is the sort of information this website hopes to convey to interested people: how to critically consume the polling information. In contrast, most of the posts here are focusing on the horserace and hypothetical projected Electoral College results. Just take a deep breath. This survey is fun to look at, but it is not what is advertised. It is not an exact snapshot of what would happen if the election were held today, much less November. Indeed, it looks mostly like the same contours of the red-purple-blue states we have seen in the past two presidential election cycles.

____________________

RS:

@Michael McDonald:
While I agree that much can happen between now and November, some of those comparisons are quite significantly different. For instance, Senator Clinton leads or ties Senator McCain in Arkansas, Tennessee, West Virginia and Florida. Senator Obama has an definite edge (sometimes relative to Senator Clinton) over Senator McCain in Colorado, Nevada, Iowa, Virginia, Washington State, even the Dakotas and Carolinas.

While Senator Obama may not carry the Dakotas or Carolinas, there's enough differences - youth and independent support, maybe racism, possibly sexism - apparent in the cross-tabs to suggest the match-ups may not be that seen in 2000 or 2004.

I agree RVs are not LVs. SUSA weights the results by census numbers, and then lets the results fall where they may... Not sure if (a lack of) that is what you mean by "non-response bias", or just a failure to lift the phone - but then, these would be applicable to any poll, not just this one.

SUSA's methodology:
http://www.surveyusa.com/client/methodology2.aspx

____________________

Rasmus:

"600 registered voters per state. Again, Riiight. Did SUSA stop polling at 600 respondents to hit 600 exactly on the spot for each state? I hope not since that would suggest possibly severe sampling bias from non-response. More likely, the 600 is an overall overage per state, with some variation among states."

What do you misunderstand on the SUSA poll comment:
"
600 registered voters interviewed in each of the 50 states"

This poll IS a snapshot of what happened when the election would be hold today.
For sure, there is a margin of error, and we still need the exact numbers to weigh in these polls, but such a poll is exactly the thing SUSA advertised it.

OK, the numbers are out.
Whew, McCains FL lead on Obama is down to 2 points, but Obamas lead in CA down to 12 points.
There is a good chance for Obama to win Alaska, he�s just 5 points behind there.

Michigan is a toss-up, also Nebraska and New Hampshire and New Jersey, McCain leads PA by 5 points.

North Dakota and North Carolina are also very close, same with Virginia and Texas-
so a LOT of battleground states in a Obama-McCain race.

Clinton-McCain:
She leads Arkansas, CA just by 10 points, Delaware is also close race but Democratic-territory, but she leads FL with 9 points. McCains lead in Iowa and hers in Hawaii are also below 5 points, and she leads IL with just 11 points, compared with 29 for Obama.
On the other hand, she leads NY with 22 points, Obama with 14.
Minnesota and Michigan are toss-ups, Missouri is close (McCain-Obama +6, McCain-Clinton +4)
New Jersey and New Mexico are also just slightly leaning Democratic/Toss-Up, same with PA, she just leads with 1 point there.
Washington state, WV and Wisconsin are also close.

Remember that this was a registered voter survey, so the "real numbers" should favor McCain in a McCain-Clinton matchup and Obama in an Obama-McCain matchup.

Nevertheless, these numbers are good for all Democrats and a little bit more for the Obamanians.

____________________

Steve:

Obama comes within 1% of McCain in Texas!

TEXAS!

____________________

cinnamonape:

Folks! You may wish to look at the data before you post comments about the maps. The spreads down below show the State-by-State stats, and one can imagine that a poll based on about 600 respondants/State would have an error rate of between 3-4%. But this measure is so far out it's of little value except in showing where states are likely to be "held", and those that are in play.

Still the survey does show some very interesting differences between Obama and Clinton in their match-ups against McCain. Hillary does well in traditional Democratic strongholds, and wins Arkansas.

But what surprised me was Obama does better against McCain in some States she won in the Primary (e.g. Obama wins NH; and Clinton loses that State). Obama wins Virginia and Colorado and makes a race of it in North Carolina.

I don't see where Survey USA calls" several races for the candidates, either Clinton v. McCain, or Obama v. McCain when they are within a point or two (or even tied). That makes no sense...but perhaps they did it to fluff up their data to predict the ultimate delagate counts...trading one state for another with similar delegate counts.

What one should do is take any State within a 5% spread and say it's within the SE and put it up for grabs. Some might wnt to increase that number to 6% or even 7%.v Even doing this I suspect you'll find that Hillary loses many States that Obama makes competitive.

http://www.surveyusa.com/

____________________

Rasmus:

I just looked again on Michigan- THIS is really a toss-up.
McCain vs. Clinton: McCain leads by 1 (!) respondant
McCain vs. Obama: Obama leads by 10 respondants

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR