Rasmussen and Opinion Dynamics Corp. Agree: Clinton Whips All Republicans

Senator Hillary Clinton now enjoys a five-point lead over Mayor Rudy Giuliani in the latest Rasmussen Reports national telephone survey. It’s Clinton 48% and Giuliani 43%. The former First Lady also leads former Senator Fred Thompson 49% to 41%.

Both match-ups represent gains for the Democratic frontrunner. But the more modest lead, over Giuliani, is also the more dramatic news.

Two weeks ago, Clinton held a statistically insignificant one-point lead over Giuliani. Prior to that, Clinton had held an advantage over Giuliani only once in fourteen consecutive Rasmussen Reports polls.

Giuliani began the year with the advantage in this match-up, but Clinton has gained ground in the latter part of the year. And Clinton has had a particularly good week, getting an often-generous reception for her recently announced health care plan (see "Hillary's Great Week"). On Health Care, voters currently trust Democrats more than Republicans by a wider margin than on any other major issue).

Senator Clinton also continues to enjoy a very healthy lead in the national Democratic nomination race. Recent polls in the early Primary States of New Hampshire and Florida show her lead growing in both places.

http://www.rasmussenreports.com/public_content/politics/election_2008_clinton_vs_giuliani_thompson

In hypothetical presidential matchups, Clinton, the clear leader for the Democratic nomination, bests Republicans Rudy Giuliani and John McCain by 7 percentage points each and Fred Thompson by 13 points.

While independents are more undecided than others, in each case they prefer Clinton over the Republican candidate by double-digits.

Similarly, Obama tops each Republican hopeful tested on the poll, outperforming Thompson by 12 points, and barely capturing the edge over McCain by 2 points and Giuliani by 1 point.

Among independents, Obama has a 12 point advantage over Thompson and 1 point over Giuliani, but McCain tops Obama among this group by 5 percentage points.

http://www.foxnews.com/story/0,2933,298297,00.html

Shes a clone of Bush plain and simple.

...have hoped will topple her.

That doesn't make a whole lot of sense.

That would be a first.

:rofl:

I've seen lots of polls suggesting that Hillary would beat Rudy, and lots suggesting that she wouldn't.

On the other hand, more or less every poll I've seen suggests than any other matchup, including Hillary vs X or X vs Giuliani, polls consistently as a clear Democratic victory.

Clinton vs Giuliani is definately the worst case scenario. Unfortunately, it also seems the most likely.

Even IF they had fielded a good candidate they'd still be in trouble.

Hillary Clinton is the last person in the world to let up...and I don't think her supporters will either...

:toast:

are these the same polls that didn't even bother to include candidates, like Kucinich?

We're only looking at Clinton vs. the Republicans here.

BTW, Kucinich gets trashed in head-to-heads.
http://www.angus-reid.com/polls/view/kucinich_trails_giuliani_thompson_in_us_race/

only reinforces my opinion that she's definitely a loser if she manges to get the nomination....

I'm sure the Republicans are salivating over that chance.

Polls showing her having a comfortable lead means that she's going to lose?
:shrug:

and what their methodologies look like. Without a crash course in quantitative methods, I can't explain all the concepts- but what I can say is that what we typically see are methods designed to influence public opinion, not reflect it.

It's a common theme these days- even though the claim is made that they're "scientific" that doesn't make them valid or generalizable, particularly on an electoral basis.

:shrug:

Had a whole semester of it when I considered getting a minor in marketing.

Your "explanation" is off-base.

if you do some homework.

Used to be, they taught this as a required college stats course- but now it seems to have largely been relegated to grad school.

Even so, it's not inaccessible. There are plenty of resources on the net that can show you how to analyze these things.

We covered polling, of course. So you can use all the jargon on me you like.

What systematic biases exist in every major telephone poll that would give Hillary the lead over her Republican opponents?

Who's more likely to answer land lines?

Especially amongst single people under 30 the most likely demographic to use cell phones only?

Pardon me for a second but :rofl:

which is accounted for in the full polling data, and which would slightly skew the head-to-head poll pro-Republican, not pro-Hillary. Next?

Is the poster claiming that cell phone only users are a hidden massive GOP demographic?

And speaking of polls did you see your guy's move in head to head vs Rudy and Thompson

http://www.rasmussenreports.com/public_content/politics/election_2008_richardson_vs_giuliani_and_thompson

what I've seen is that typical diehard Republicans tend to screen their calls (or are likely to be at work when pollsters are allowed to call).

Cell's are another interesting matter. The demographic has two tendencies: young people- and business people. My bet would be that both are seriously under represented.

Who chooses to respond to a pollster is also interesting. Again, my bet is that it's more "conventional" types of folks who aren't pressed for time.

because Republicans don't pick up their phones, and young people don't pick up their phones, and businesspeople don't pick up their phones. Which is interesting, because in the full data, they generally show proportional representation for party affiliation and ethnicity. And it's also interesting that telephone polling ends up matching actual vote tallies pretty darn well.

I also note that you've shifted away from "it's scientific, you wouldn't understand" to "I personally believe that everyone who doesn't want Hillary does not answer their phone." Sounds more like denial than math to me.

And I never said "anyone wouldn't understand" I just pointed out that there are a lot of concepts to pick up.

As to phone polls matching up with results- there are PLENTY of anomalies among the polls- some of which are chance- and some of which reflect crappy methodology. In addition, you're also looking at results that are very close in time to the actual election.

Indeed, it's laughable to cite a poll as "proof" of something almost 14 months away. Dukakis had a lead of how many points just months before the 88 election?

We're saying that this is an accurate reflection of the electorate as it stands now.

If there were just one poll, sure, I could see you saying "well, there are occasional aberrations." But this is a trend that's been going on for weeks. Every single poll shows her not only leading but pulling away from the Republicans, and every single poll shows her with 15 to 25 above Obama and Edwards.

And back to the original point, saying that these polls suggest that she's "a loser," as you put it, is downright stupid. These polls say that she has a healthy-but-beatable lead over the Republicans, and an imposing-but-possibly-surmountable lead over her Democratic challengers. You suggested there were systematic biases reflecting "an agenda" that artificially inflate her numbers; you've now retreated to saying "well, polls don't mean much this far out."

That's a far more reasonable stance than "this shows she's a loser because there are ingrained pro-Hillary biases."

Only the various ones that Hillary supporters cherry pick say that.

Ands the reason I submit is obvious as in one my earlier post- the deal is influence public opinion- NOT to reflect it.

And in many respects- that's what the people commissioning these polls want to do (particularly the one in the OP)- they WANT Hillary to be the nominee- because they know she can and will be trounced in 2008.

you have to go back over a month--to August 28--to find a single poll in which she's behind Ghoul, and that's by three points, which is within MOE. There's only one poll released that shows her behind Thompson--by one point--and that was in July. McCain hasn't led her since June, and neither has Mitt.

Here, take a look:
http://www.realclearpolitics.com/epolls/2008/president/national.html

Ah, so we're back to influencing the public. You still have yet to provide a single example of how the polls--that is, every single poll performed--are twisted to make her look better, or given a credible reason why why Gallup, Rasmussen, Zogby, MSNBC, CNN, Quinnipiac, and Newsweek have all decided that this year they're going to abandon the tradition of accurate polling that has been their moneymaker for decades and make it look like Hillary is going to win.

You claimed you had all sorts of ways that they were trying to lead public opinion. Let's hear one of them--and "Only Hillary Supporters Pick Up The Phone" was pretty weak, so let's get something solid out here.

The last poll to show her losing to Rudy were in mid July by Battleground and Gallup which had her within the MOE.

Poll Date Sample Giuliani (R) Clinton (D) Und Spread

FOX News 09/25 - 09/26 900 RV 39 46 15 Clinton +7.0
Rasmussen 09/24 - 09/25 800 LV 43 48 9 Clinton +5.0
Cook/RT Strategies 09/13 - 09/15 855 RV 43 45 11 Clinton +2.0
Rasmussen 09/10 - 09/11 800 LV 45 46 9 Clinton +1.0
NBC/WSJ 09/07 - 09/10 1002 A 42 49 4 Clinton +7.0
CNN 09/07 - 09/09 1017 A 46 50 1 Clinton +4.0

Here's every poll taken this month. Her lowest lead is 12 (in Ras a week agao which is back to 16 pts)


Poll Date Sample Clinton Obama Edwards Richardson Spread
Rasmussen 09/23 - 09/26 750 LV 40 24 14 4 Clinton +16.0
Rasmussen 09/20 - 09/23 750 LV 40 28 15 3 Clinton +12.0
CBS News* 09/14 - 09/16 Adults 43 22 16 -- Clinton +21.0
Gallup 09/14 - 09/16 531 A 47 25 11 5 Clinton +22.0
Reuters/Zogby 09/13 - 09/16 LV 35 21 10 3 Clinton +14.0
Pew Research 09/12 - 09/16 568 RV 42 25 14 3 Clinton +17.0
Cook/RT Strategies 09/13 - 09/15 405 RV 36 23 18 3 Clinton +13.0
FOX News 09/11 - 09/12 396 RV 43 24 13 1 Clinton +19.0
AP-Ipsos 09/10 - 09/12 482 RV 43 23 13 4 Clinton +20.0
American Res. Group 09/09 - 09/12 600 LV 39 21 15 5 Clinton +18.0
NBC/WSJ 09/07 - 09/10 Adults 44 23 16 4 Clinton +21.0
CNN 09/07 - 09/09 Adults 46 23 16 5 Clinton +23.0
NYT/CBS News* 09/04 - 09/09 Adults 44 26 17 -- Clinton +18.0
USA Today/Gallup 09/07 - 09/08 500 LV 45 24 16 2 Clinton +21.0
Rasmussen 09/05 - 09/08 750 LV 43 22 16 4 Clinton +21.0
ABC News/Wash Post 09/04 - 09/07 Adults 41 27 14 2 Clinton +14.0

I'll cede that to you. For what they're worth, your polls do say that.

Vis a vis other candidates against Republicans (compared with Hillary Clinton) that's another matter entirely.

And Hillary vis a vis other Dem candidates- there are more than a few disparities.

If you're choosing a candidate based on herd mentality- OK then. But recognize that herd mentalities often run right off cliffs.

Looking at the other candidates? Sure. Hillary is tied with or beats Barack against each of the Republican big four, and runs a few points behind Edwards on Mitt and Thompson, but beats him on Ghoul and Walnuts. Not really seeing "she's a loser" here.

Now, let's look at the "more than a few disparities" in the Dem candidate field. Here are the last six polls released, regarding the primaries:

Clinton+17
Clinton+16
Clinton+21
Clinton+22
Clinton+14
Clinton+17

Yes, there is a decent amount of swing there. However, all of it shows her consistently having a strong lead.

We're not talking about choosing a candidate here. We're talking about what the polls say. It might behoove you to actually look at some polls next time you decide to argue about them.

And you do realize introducing anecdotal evidence doesn't really help much.

The biggest problem is you think "conventional" types are oversampled when election after election proves "conventional" describes a huge majority.

which is all that matters.

They know in advance what percentages of various groups they need in advance to get a representative sample...If the sample is built on an electorate that is 10% African American they are going to keep calling to they get that percentage... Ditto for any group; women, men, GOP, Indys, Hispanics, Dems, etcetera...

and so is clustering. This is particularly true with "national" polling- which doesn't mean a lot in terms of the electoral college. It's also a problem with statewide polls. How one does their clustering can (and does) change results well outside of margins of error.

Those are in addition to the problem with selection, response and non-response bias.

eom

I can't tell you how many pieces of research I've seen fouled up by weighting.

Gallup was crucified during the 04 election because some folks felt they were oversampling Republicans...

You're smarter than me ...I'll pass the baton to you...

But you don't need a PHD in Statistics from MIT to see polls have a pretty good track record... They have come within a percent of two in every presidential election since 1948...

Having once had a mailing address in Hyde Park doesn't automatically make you right, despite what practically everyone who goes there believes ;)

Because it would be nice to share if you actually did the analysis.

eom

100 polls with questionable methodology aren't necessarily going to be more accurate that any given one, although you will get some averaging.

Take telephone polls for example (a situation like the infamous literary digest poll in 1936). They exclude quite a number of people who:

Only use cell phones;

Screen their calls;

Refuse to participate

Aren't home during the time periods pollers call.

In each instance, this creates a bias- where the survey sample doesn't reflect the electorate at large.

Because most of the firms did quite well in nailing the 2006 and 2004 elections.

http://www.pollingreport.com/2004.htm#Pollster

If they were as flawed as you suggest they would have never have come as close...As they say the proof is in the pudding...

I concede it's very hard to get a representative sample; with the advent of fax machines, voice mail, and cell phones , but those problems can be overcome and they are routinely overcome...

It's not on the report....

and another thing to consider- what explains the swings and inconsistencies seen in the polls that year? Are THAT many people really changing their minds that much?

Could it be that the farther away an election, the more important it is for certain groups commissioning polls to want skewed results?

"yes, many people frequently change their minds. It is not unheard of for a person to change their minds two or three times in a campaign season."

The answer to the second question is, "No, actually, things this far out are worthless for public opinion. Crunch time for campaigns is the final month and the final week, and that's when the polls are most precise."

And seriously, don't you think it's kind of silly to be suggesting that Newsweek, Gallup, Zogby, Rasmussen, Quinnipiac, NBC, and CNN are all part of some grand conspiracy?

Zogby aside, prevailing methods and practices often become ingrained in certain industries- regardless of their objective value.

Why does every single local "news" channel (sometimes 4 and 5 of them) have almost the exact same content? Not to mention national "news" channels. Think about it.

Moreover, I've seen way too many ridiculous things done in the consulting and polling businesses to discount the fact that companies get the results they're paid for. Sometimes intentionally- though not always. It would be really easy for me to put together "scientific" looking surveys that would show very differing results. The math would work just fine.

75% of Americans believe that angels are watching over them every day- sure, I can do that. So can the polling firms. I can also cross tab things like GSS data and garner very different results.

that you believe give Hillary an advantage are. You're long on vague complaints, and short on solid answers.

Again, because you seem to be ducking it: what exactly are every single polling company in America doing wrong? What are the wordings that are altering the answers? What's wrong with their sampling methods? What is objectively wrong with the way polling is done?

A dozen posts asking you to tell us, and you've yet to pin down anything but "um, only Hillary supporters are picking up their phones."

I did post grad work in Government at FSU and I am quite familiar with the math behind the polls...They are what they are; relatively good snapshots that give us an idea of what is happening now...

Granted that among the population (or type of population) being sampled they may be useful at showing trends over time- provided that the polls ask the exact same question to the same group or "type of group".

Unfortunately, all of the cool math and statistic controls in the world won't make up up for loaded questions or poorly drawn samples, which is the norm with these types of deals.

Good research is expensive and time consuming- which is why campaigns spend so much money on their internals (and keep the results held close to their chests).

There's also a dynamic in place...Many people just want to be on the side of the winner...It's an empirically tested proposition whether some folks like it or not...

See above.

yet when you look behind the polls- who's sponsoring them, what the sample consists of, how the clustering is done, etc. you can often find an agenda.

I Hope this holds. My biggest problem w/ HRC has been her polls in the general.
I Hope that lead gets stretched out well past a 5% spread so they can't steal it again.

But goddammit i wish she'd stop with that forced, obnoxious *laugh*(shudder).

As long as he can keep his foot out of his mouth (and get a few articles written about him in which he's saying something other than "I'm Hispanic"), he'd make a good VP choice. Personally, I think Sec'y of State would be the best place to put him.

But he needs to shine more in the debates. He has been mediocre in most of them.

I like all our candidates except Gravel... He's intentionally rude, condescending, and uncivil...

Baltimore MD true blue :)

"Sir, we have the intellectual vote locked up."

"Great," Adlai replied, "We've got the intellectual vote. Now what about the other ninety percent of Americans?"

We've got the Baltimore garbageman vote... :) Now for the rest of it!

So stop yer f*cking whining.

If she gets the nomination- you'll be singing a different tune in November 2008.

So will your state level candidates.

The Clintons, they just have no political chops.

What happened to Democrats and progressive issues from the mid 1990's until 2006?

Why was that?

Think it might have had something to do with enabling and legitimizing far right policies?

My state just gained back our legislature- and some positive things got done for the 1st time since 1994.

Have a look around- see what the benefits of "triangulation" were on the state and local level.

Do yourself a life affirming favor, fella. Cut back on the doobie.

The herbal lobotomy thing is not doing you a whole lot of good.