Final Des Moines Register Poll: Obama And Huckabee Ahead
The final Des Moines Register poll before the Iowa caucus shows Barack Obama and Mike Huckabee leading their respective parties. Meanwhile, Rudy Giuliani is in sixth place behind Ron Paul, having lost more than half of his support since last month's poll:
Democrats:
Obama 32% (+4)
Clinton 25% (+0)
Edwards 24% (+1)
Richardson 6% (-3)
Biden 4% (-2)Republicans:
Huckabee 32% (+3)
Romney 26% (+2)
McCain 13% (+6)
Thompson 9% (+4)
Paul 9% (+2)
Giuliani 5% (-8)
Comments (187)
CalD wrote on January 1, 2008 3:35 AM:I'm still expecting Romney to win the Republican contest.
kjoe wrote on January 1, 2008 3:36 AM:Close examination shows that it is due to a huge influx of cheap young sobs who do not want to be forced to buy health insurance. If Obama wins, it will be the result of the most cynical political ploy in the history of ploys. Hillary should have kept running that ad explaining how what you need to do is call her, and she will call a hospital or doctor for you. She was apparently afraid it might lead to more hostage taking----I say if she can't stand the heat, turn on an air conditioner.
Jenna L wrote on January 1, 2008 3:46 AM:W00t!!!!!!!!!
I already see some spinning of this poll, which is understandable since it is a bit wonky. There is a huge number of Independents and young people expected, much more so that in 2004. But that makes sense to me. There has never, ever been this much attention placed on the Iowa Caucus. The turnout is bound to be extraordinary.
But the POINT is that this poll is not just a poll. The DMR actually has the power to make its prediction come true. When Iowans wake up tomorrow, it will be to this headline: Obama widens lead over Clinton. That is huge. I think people will align themselves with the winner, particularly if they are against Hillary. Also, this poll is widely considered "the gold standard" of Iowa polls among media organizations--that changes the dynamic of news coverage away from "Obama loses momentum." I think it is clear now that he didn't lose momentum at all -- his voters just weren't around for Christmas! Anyway, this is GREAT for Obama and I say again that he is going to win.
Abe wrote on January 1, 2008 4:10 AM:I'm wary about the high independent count. That said, Obama has closed the second-choice gap and his lowest voter model only puts him -2 to Clinton. That and having the largest(and second-most ingrained to Edwards) organization puts Obama in good position to come away with a win.
CalD wrote on January 1, 2008 4:19 AM:As I noted on the previous thread, Marc Armbinder sez Obama's internal polling does not show this high a proportion of independents choosing to caucus. I imagine there might even be some worries there about expectations setting.
M Miller wrote on January 1, 2008 4:51 AM:Some people will never be able to afford health insurance. Some people that do pay for it don't have money for other stuff they need (like money to get their car fixed so they don't die on the road or so it don't break down and they can't make it to their job).
There needs to be a major check on the insurance companies anyhow. At least the government wouldn't have someone buying 20 cars and living the high life because people decided to get their insurance. The money could be thrown back into the system and pay for better healthcare.
Do these rich people and people that have more money than other hardworking people realize that not everyone makes the same as them even though they may work just as hard OR HARDER?!
amber wrote on January 1, 2008 5:33 AM:Where is the man behind the curtain? I'm shocked that Team Hillary hasn't jumped on this page stomping around posting like Rumpelstiltskin, throwing dung like angry monkeys and the slime that Hillary-the woman who will never be president- and her surrogates are known for.
It's a good news to start off 2008. Edwards and Obama have helped Bill Clinton with his self-fulfilling prophecy that Hillary would not win Iowa.
Anonymous wrote on January 1, 2008 6:00 AM:Where's Greg Sargent on this one?
l wrote on January 1, 2008 7:42 AM:It looks like Richardson and Biden folks moved over to Obama in anticipation of their not being viable.
I have to say, the Obama camp must be deliriously happy today. This poll, like someone said above, is really self-actualizing so Obama is poised to win Iowa. My question is if he can do the same thing in the rest of the country before Feb 6th. Hillary is ahead just about everywhere else and there won't be that intense campaigning that happened in Iowa. That's very little time for him to get the name recognition that she has. I sure hope so. But I guess I still think Hillary will be the nominee.
Anonymous wrote on January 1, 2008 8:18 AM:" l " wrote regarding upcoming primaries: Hillary is ahead just about everywhere else and there won't be that intense campaigning that happened in Iowa."
There is a clear pattern: Mrs. Bill Clinton leads nationally and in most state polls based on name recognition. When individual primary elections draw near, state by state, voters pay attention, realize Mrs. Bill has not substance, and move to other candidates.
Campaigning will not be as concentrated after IA, NH and SC... that's why early wins there are important: mommentum and constributions to help the expanding campaign succeed.
Very few voters are undecided about Hillary once they look at her closely. The trends show, MANY AND MOST voters don't trust her, don't like her and don't want her for president.
Kefa wrote on January 1, 2008 8:22 AM:What folks have to understand is this....
THE POLL WAS TAKEN IN LATE NOV. that is all.
zogby's....which I have no great love for...is at least current.
Kefa wrote on January 1, 2008 8:30 AM:correction.....my mistake the poll I was looking at is correct my bad.....DRM poll is correct..MY BAD....MY BAD. SORRY.
Jan in Iowa wrote on January 1, 2008 8:38 AM:I just checked and this poll was taken December 27-30 so, yes it's the most current poll.
The poll predicting victory for Obama may help him not only win, but win big. I'm hoping some Edwards supporters may switch over, based on the facts in the memo that the Obama campaign presented yesterday about spending limits preventing Edwards from going the distance.
Jenna L wrote on January 1, 2008 9:10 AM:Something else that I think is good news for Obama is that his lead appeared to be increasing. He had 29% of the vote on the 27th, which increased to 33% by the 29th and 34% by the 30th. This really matches the theory that his voters were just out of town earlier. On the last day of the poll the numbers were 34% for Obama, 26% for Edwards and 23% for Hillary....Could Hillary come in third place??
Here is the link for the poll. The data is very interesting. http://www.desmoinesregister.com/apps/pbcs.dll/article?AID=/20071231/NEWS09/71231044#gslPageReturn
dcshungu wrote on January 1, 2008 9:48 AM:Jena L: Also, this poll is widely considered "the gold standard" of Iowa polls among media organizations--that changes the dynamic of news coverage away from "Obama loses momentum." I think it is clear now that he didn't lose momentum at all -- his voters just weren't around for Christmas!
It is interesting that the Presidential Markets do not seem to share your enthusiasm...
Why in the world would investors believe more than ever that Clinton would be the nominee just a day after a revered poll shows Obama to have increased his lead? Check out the INTRADE Presidential Markets' numbers on the right. The % of those who think that Clinton would be the nominee has jumped while that of those who think Obama would be has gone down. Edwards appears to have gained a point:
Candidate Today Yest. Diff.
Clinton 69.7% 67.1% +2.6
Obama 20.7% 24.0% -3.3%
Are the markets, like me, comfortable with Hillary coming out second in IA? Are they picking some "noise" that we are not yet aware of? Have the experts there analyzed the DMR poll numbers and found the same "irregularities" in the sample composition, etc, that I found, and decided that this poll was an outlier?
We'll know in just a couple of days!
bm wrote on January 1, 2008 9:53 AM:All that money
All those endorsements
Bye Bye Hillary
Very simple
The American people
view you unfavorably
And you are done as a candidate
never been happier
Derek wrote on January 1, 2008 9:53 AM:Happy new year indeed!
green heron wrote on January 1, 2008 10:01 AM:INTRADE prediction for Iowa:
Obama 46%
Clinton 35%
amber wrote on January 1, 2008 5:33 AM:Where is the man behind the curtain? I'm shocked that Team Hillary hasn't jumped on this page stomping around posting like Rumpelstiltskin, throwing dung like angry monkeys and the slime that Hillary-the woman who will never be president- and her surrogates are known for.
Sad commentary...
Madam, look in the mirror... The only person throwing any dung here is you and those showing great poise are the much maligned Hillary supporters who have repeatedly and unanimously pledged on this forum that they would support who ever emerges as our party's nominee. If you had any shame or decency, you'd immediately stop throwing bombs and join us in ensuring that we beat the GOP in November with whoever our candidate turns out to be.
It's a good news to start off 2008. Edwards and Obama have helped Bill Clinton with his self-fulfilling prophecy that Hillary would not win Iowa.The last time I checked, the Iowa caucuses were still about a day and half away, and no one had won or lost yet...
Do they serve booze this early where you live, ey, Madam?
colonpowwow wrote on January 1, 2008 10:10 AM:bm:
Projection vomiting, bm. Won't YOU be surprised when Senator Hillary Clinton secures the nomination by late February and wins the Presidency by a huge margin setting a new record for number of Democratic voters (over the nosebleedingly exciting Kerry/Edwards 2004 ticket)?
I won't be. I'm planning on signing up and driving at least 15 new voters (especially women) at Indiana University (my goal right now). We may not win Indiana for her, but we all want to be a part of this shining chapter in our country's progressive history.
By the way, I hear Nader's thinking of taking some time off from managing his stock portfolio online and may take another run. I'm sure there are a number of jobs open with his campaign right now so get your resume in early!
dcshungu wrote on January 1, 2008 10:12 AM:green heron wrote on January 1, 2008 10:01 AM:INTRADE prediction for Iowa:
Obama 46%
Clinton 35%
I would be happy with this prediction for IA if INTRADE's prediction for the final outcome holds (which I believe it will).
MIchael A wrote on January 1, 2008 10:12 AM:Happy new year dc and everyone else. I will stick with who knows. The polls have been all over the place. We'll find out on thursday. It is and has been a horse race. May the best person win.
Liam wrote on January 1, 2008 10:13 AM:Edwards could not handle Dick Cheney in the 2004 debate, but now we are supposed to believe, after he has done nothing for the last four years, that he is going to dictate to the entire ruling class, and change Washington for ever.
Do you believe in fairy tales!
brad wrote on January 1, 2008 10:14 AM:Good news.
Note to all Iowa caucus goers from a neighbor to the East, for what its worth:
I will forgive every humiliation Hayden Frye's teams visited upon my Badgers if Obama wins the Iowa Caucus.
Oh, and thanks again for Barry Alvarez.
green heron wrote on January 1, 2008 10:19 AM:Over the holidays I talked politics with my rather large family of inlaws. All the men--even the Republicans--prefered Obama or Ron Paul. All the women prefered Clinton, but none of them were rock solid in their support. I think this illustrates the dynamic at work in this campaign. It is undeniably a change election. None of the men liked the hard core Republicans. The women, for obvious reasons, were intrigued by the prospect of a female president, but this being an election following eight years of George Bush, recognized Hillary did not represent a change, and were wavering in their support. I think it looks good for Obama.
green heron wrote on January 1, 2008 10:26 AM:By the way, if Obama wins Iowa the INTRADE national prediction will change within the hour. It's just like what happens at the horse track: the announcer informs the crowd that a young unknown has won recently in a smaller venue, and immediately the odds change.
bm wrote on January 1, 2008 10:27 AM:colonpowwow
You are borderline crazy
Didn't you know that Nader has already endorsed Edwards.
Probably will help him beat Hillary for second in Iowa.
I just want to see the end of the Bush-Clinton era.
Go Bama
Beat the racist slime machine run by Hillary
I cant stand her and she could never win the Presidency
I agree with Liam, Edwards is a phony and his supporters are a bit naive to canonize him as a populist saint when his record (and his rhetoric only a few years ago) doesn't match the talk at all. I dig what he is saying, but it is obviously contrived. His tough guy talk isn't going to gain Democrats any victories either, he would just piss people off and lead the government to a standstill. Hillary is too divisive without even opening her mouth. Out of the three Obama is by far the most talented statesperson, he can bring people together and have them eating up his policies and they won't even notice they are trending left the whole time.
Obama has great experience in my opinion, but I think what is most important is his character and personality, both are what we NEED in this country right now (not in 4-8 years).
Kucinich for prez wrote on January 1, 2008 10:34 AM:One would think there are more Republican candidates than Democrats. Not on TPM .
cms wrote on January 1, 2008 10:37 AM:dcshungu -- the markets are reactionary, not predictive. If you want to make some money, buy Obama.
rg wrote on January 1, 2008 10:42 AM:I agree with Liam above. I like Edwards - but I've never been able to get that Cheney debate image out of my head. I think this poll reflects what is true - that Obama needs some non-traditional support to do this - and I think he has it. I also think that's why he's the best choice. We cannot afford a candidate that only a democrat can love. Go Obama!
awrbb wrote on January 1, 2008 10:44 AM:The INTRADE numbers dcshungu quoted are from before the DMR poll. So far today, the nomination contract for Clinton is down 3.2, while Obama is up 6.3. The changes in Iowa are bigger: Clinton is down 6.9 and Obama is up 15.5. If Obama goes on to Iowa (as INTRADE currently predicts, for whatever that's worth), look for the NH and national numbers to shift massively in his favor.
FFA wrote on January 1, 2008 10:45 AM:Are the markets, like me, comfortable with Hillary coming out second in IA? Are they picking some "noise" that we are not yet aware of? Have the experts there analyzed the DMR poll numbers and found the same "irregularities" in the sample composition, etc, that I found, and decided that this poll was an outlier?
These were the average prices at the Iowa electronic markets two days before the 2004 Iowa Caucus.
Dean - .521
Clark - .205
Kerry - .115
Draw your own conclusions.
dcshungu wrote on January 1, 2008 10:48 AM:A new CNN poll for IA has just been released:
Clinton 33%
Obama 31%
Edwards 22%
DES MOINES, Iowa (CNN) -- With two days to go until the Iowa caucuses, a new CNN/Opinion Research Corp. poll out Tuesday shows both the Democratic and Republican presidential nomination races tied at the top. But with a quarter of all Democratic voters and nearly half of all Republican voters still making up their minds at this late stage, almost anything can happen Thursday night in the first contest for the White House. Among Democrats, Sen. Hillary Clinton of New York wins the most support, with 33 percent of likely Democratic caucus-goers backing Clinton and 31 percent supporting Sen. Barack Obama of Illinois. But taking into account the survey's sampling error of 4.5 percentage points in the Democratic race, the race is virtually tied.dcshungu wrote on January 1, 2008 10:50 AM:Former Sen. John Edwards of North Carolina is in third place in the poll at 22 percent.
Clinton and Obama both gained 3 points since the last CNN/Opinion Research Corp. poll in mid December, with Edwards dropping 4 points.
"The survey suggests that for the Democrats, a three-way race may have effectively become a two-way race," said CNN Polling Director Keating Holland.
The poll indicates that Iowa Democrats believe Clinton has the best chance of winning in November and is the most experienced. Obama is seen as the most likable and the most honest.
"Edwards doesn't stand out on any of the qualities, according to poll," said CNN Senior Political Analyst Bill Schneider.
cms wrote on January 1, 2008 10:37 AM:dcshungu -- the markets are reactionary, not predictive. If you want to make some money, buy Obama.
I agree with the first part and will have to pass on the second, at least for now... :-)
john mccutchen wrote on January 1, 2008 10:59 AM:Blumenthal: DMR Polls Sets the Standard for All Iowa Polls
http://www.pollster.com/blogs/poll_of_pollsters_rating_the_i.php
What else is there to talk about?
3:23 am. Heh. Methinks I wasn't the only person engaged in non-political activities last night.
Just a point about INTRADE that seems to have escaped everyone. It is apparently much more difficult for Americans to make a deposit and participate in INTRADE since Bill Frist, of much accursed memory, snuck his pet Illegal Internet Gambling Act bill onto a defense appropriation bill at the last minute back in '06. Just a byproduct of Bill's noble goal of protecting U.S. casinos and tracks from having to compete for the money of compulsive gamblers in the interest of somehow protecting us from from the t'rr'sts who hate our freedoms.
My point here is that I think INTRADE's always overrated predictive power has been reduced by its having a larger percentage of Brits than it used to on the U.S. election markets. I think Brits tend to over-estimate the power of institutional favoritism and racism in making their bets about us. No data on the latter, just a perception.
Ooohhhh, my head . . . going back to bed now.
Kucinich for prez wrote on January 1, 2008 11:10 AM:deshengu, did CNN filter out the non-'frontrunners', or did you? Why?
dcshungu wrote on January 1, 2008 11:11 AM:john mccutchen wrote on January 1, 2008 10:59 AM:Blumenthal: DMR Polls Sets the Standard for All Iowa Polls
With nearly every other recent poll, including the latest from CNN/Opinion Research, painting a different picture, the DMR/Selzer poll would either be further elevated or it will crash and burn on Thursday night, especially if Edwards wins(a definite possibility due to his six-years-in-the-making organization) and Obama is third (a definite possibility considering his "constituency")...
Dan wrote on January 1, 2008 11:13 AM:This analysis of the turnout model of the DMR lays the ground for why this is such a strange turnout projection -- though not impossible.
Jenna L wrote on January 1, 2008 11:14 AM:"A new CNN poll for IA has just been released:
Clinton 33%
Obama 31%
Edwards 22%"
Wow...so it looks like EDWARDS is the one who peaked.
That being said, Edwards has a ton of support from rural voters. Their votes count MORE than urban counties. I think he will still come in second, while Hillary comes in at a humiliating THIRD.
I think that karma is alive and well.
I visited the site Hillaryis44 late last night, after the 10pm news of the Des Moines Register poll results.
One Hillary site commenter wondered why the Hillary camp-connected whispering campaign about 'drugs' and 'Muslim heritage' wasn't getting better results [that comment has been expunged as of this morning]. Another commenter dubbed HillaryLandRocks, at 10:39pm said, "Oh, yes, here in Western IA, The BO supporters I've talked to are Repubs. FWIW, I've talked some of them out of supporting BO. But the region is much larger than my reach...."
I suspect that the Hillary folks who were relying on trashing Barack Obama through vicious rumors or through directly 'talking some out of supporting BO' have no inkling of the difference between folks in the cut-throat eastern seaboard and the fair and square folks of Iowa, whatever the Iowans' party affiliation. Because they chose to denigrate Obama and because they chose to treat Iowans as less than astute, these Hillary folks have shot down their own preferred candidate's good will.
I myself spent days in western Iowa in October and November, and never did I disparage or even mention another candidate. Furthermore, western Iowa is a place of friendly, kind and wonderful folks [yes, mostly Republican] who are wise to and rejecting of such underhanded politicking.
DTM wrote on January 1, 2008 11:16 AM:The betting markets sometimes seem to reflect poll results a few hours before they are widely released. Other than that, they have no special predictive powers.
Which makes sense: the markets can do no more than reflect the information available to the participants in the market. And the exact problem with predicting the future is that we do not have enough information about the future to be able to predict it reliably.
jeanba wrote on January 1, 2008 11:17 AM:Blumenthal: DMR Polls Sets the Standard for All Iowa Polls
http://www.pollster.com/blogs/poll_of_pollsters_rating_the_i.php
DCShungu,
many people do believe the dmr polls are more predictable than cnn that`s why so much hype about this poll. who cares about cnn poll, none.
Kucinich for prez wrote on January 1, 2008 11:10 AM:deshengu, did CNN filter out the non-'frontrunners', or did you? Why?
For crying out loud, click the god-damned link that I had provided! I cut out anything that I was not interested in, including the Republicans, that is why...
Kucinich for prez wrote on January 1, 2008 11:17 AM:Yeah the DMR has power. They hurt Kucinich [and the debate] by capriciously barring him from their debate.
DTM wrote on January 1, 2008 11:18 AM:By the way, Jenna L is of course correct: the tracking data provided along with the overall poll results strongly supports the holiday effect hypothesis proposed by Blumenthal at pollster.
And as expected given that hypothesis, it appears that the holiday effect is still unwinding.
DTM wrote on January 1, 2008 11:25 AM:On the CNN poll: I personally wouldn't ignore it, but it is probably worth noting that it was conducted from 12/26-12/30, and the DMR poll was conducted from 12/27-12/30. As we have been discussing, the DMR tracking data strongly supports the holiday effect hypothesis proposed by Blumenthal (as have other poll trends). So, that extra day of polling on the front end, right after Christmas, was likely to significantly increase the magnitude of the holiday effect in the CNN poll.
dcshungu wrote on January 1, 2008 11:32 AM:
DTM wrote on January 1, 2008 11:18 AM:
By the way, Jenna L is of course correct: the tracking data provided along with the overall poll results strongly supports the holiday effect hypothesis proposed by Blumenthal at pollster.
And as expected given that hypothesis, it appears that the holiday effect is still unwinding
LOL. If there were a "holiday effect" Obama would not have been shown to be ahead because this poll was conducted Dec 27-30, the span during which the other polls, including the latest from CNN (Dec 26-30), that showed him "trailing" were taken. If you believe in a "holiday effect" or in "Santa", then you must also believe that this DMR poll is an outlier, which I do.
Kefa wrote on January 1, 2008 11:32 AM:the ladies are at it now....
http://politicalticker.blogs.cnn.com/
With the trillions at stake in the US presidential race, I find the posters in this blog dangerously naive.
Do you really think that the defense oil and media industries would allow the polls to be honest? They stole the last 2 elections. Do you think they would balk at stealing polls? When they [or their friends] own them?
Please, talk about something that could have an impact. Like repairing the vote-counting system. Or the electoral college.
Kefa wrote on January 1, 2008 11:37 AM:Jenna L....the Paper endorses HRC and it means nothing. It polls say BO and it helps
him. Please explain? It seems to me an endorsement helps a person more then a poll. Hey, but that's just me.
how was the cnn poll sample different from dmr, anyone knows
dcshungu wrote on January 1, 2008 11:43 AM:The "holiday effect" continues:
Clinton holds lead as Romney slips in Iowa
By John Whitesides, Political CorrespondentDTM wrote on January 1, 2008 11:50 AM:DES MOINES, Iowa (Reuters) - Democrat Hillary Clinton and Republican Mike Huckabee hold narrow leads on their top rivals two days before the state opens the presidential nominating race, according to a Reuters/C-SPAN/Zogby poll released on Tuesday.
Clinton, a New York senator, maintained a stable four-point edge over Illinois Sen. Barack Obama, 30 percent to 26 percent, in the Democratic race. Former North Carolina Sen. John Edwards was in third at 25 percent, down one point overnight.
Huckabee, a former Arkansas governor, widened his lead over Romney among Republicans to 29 percent to 25 percent. Romney, a former Massachusetts governor who has been on the attack against Huckabee, slipped two points overnight.
Arizona Sen. John McCain remained a distant third in Iowa's Republican race at 12 percent, with former Tennessee Sen. Fred Thompson gaining two points to climb to fourth at 10 percent.
The poll of 925 likely Democratic caucus-goers and 928 likely Republican caucus-goers was taken Friday through Monday and has a margin of error of 3.3 percentage points for each party.
Iowa is the first test in the state-by-state fight to choose candidates for the November presidential election, and about 7 percent of Democrats and 6 percent of Republicans remain undecided after months of heavy campaigning[...]
dcshungu,
I am not sure you understand Blumenthal's hypothesis. The holiday effect he hypothesized would skew the results of polls in favor of Clinton and against Obama the closer they were taken to Christmas. But of course all sorts of other factors would continue to create differences between various polls, such as different likely voter models.
That is why the tracking data in the DMR poll supports the hypothesis. It does in fact show Obama ahead in all three of the two-day periods. But it also shows his numbers going up and Clinton's numbers going down the farther away from Christmas we get.
Which, incidentally, is why the holiday effect hypothesis is currently something a Clinton supporter should want to be true. The alternative hypothesis to explain these tracking results is that there is real momentum in favor of Obama and away from Clinton, and obviously you would not want that to be true.
Of course you have offered the last defense when the polls aren't going your way: you declared this poll an "outlier". But if that is the best you can do, your candidate is not in good shape as far as polling trends is concerned.
bnb wrote on January 1, 2008 11:50 AM:It's too bad the DNC and RNC insists on Iowa and New Hampshire picking our candidates. They have no cities.
Jenna L wrote on January 1, 2008 11:54 AM:Dcshungu,
I think the DMR poll is far superior to the Zogby and CNN polls for two reasons
1) If you analyze the data of the DMR poll, Obama has TREMENDOUS support among Independents. This is the support that puts him in the lead. A poll that is only talking to Democrats will produce the above results that ties him with Hillary. DMR discusses this (I recommend you take a look) and gives exactly those numbers for likely democrats. So, the Zogby and CNN polls only serve to echo the finding of the DMR poll.
2) When you compare track records, DMR was the closest by far to the actual results of the 2004 caucus. Zogby and etc were way off.
And a bonus reason: Now that DMR has declared Obama the winner, he has won! Iowa voters pay attention to this poll more than any other. People will move away from Edwards and towards Obama in a attempt to beat Hillary. So, yes, I think all signs point to him winning.
DTM wrote on January 1, 2008 11:54 AM:dcshungu,
Actually, so far the Zogby tracking polls are not inconsistent with the holiday effect hypothesis, although it is true they do not provide the same strong confirmation as the ARG and DMR polls. But it will certainly be interesting to see what happens with the Zogby tracking polls as more come in.
dcshungu wrote on January 1, 2008 12:02 PM:DTM wrote on January 1, 2008 11:50 AM:dcshungu,
I am not sure you understand Blumenthal's hypothesis.
I do. The question is: do you?. It was just that, a hypothesis, with lots of caveats that you completely ignore. There is no evidence so far of a "holiday effect." If you believe that there is such an effect, then you must agree with me and with that the DMR a poll is might be outlier. You have been mindlessly hanging onto the theory of a "holiday effects" through Obama's recent turbulent days and won't let go, even suggesting that polls taken after Xmas had begun to show Edwards and HRC declining ad Obama going up, although only one of 6 polls at the time had shown Obama "leading". Now, it is only 2 of now 9 recent polls that show Obama "leading"! For consistency, if you believe that the purported "holiday effect" is for real, you must conclude that the DMR poll is an "outlier." It is as simple as that.
Anonymous wrote on January 1, 2008 12:05 PM:After taking my time this day to check the method behind the DMR poll it seems IMHO it all come from it giving a huge influx of first-time caucus goers the benefit of the doubt that they are gonna come out in record numbers to make this all come out.
Chancy, very optimistic. We will see. It is bucking the odds. Big time.
and what I forgot also forgot to say was it also calls for the record turnout to be onesided toward BO. Giving JE/HRC none of the turnout.
colonpowwow wrote on January 1, 2008 12:16 PM:I don't know if I'll be around Thursday or Friday, so I'll post my "spin" early.
Clinton-Obama-Edwards finish = Hello, Mme. President, Goodbye JRE.
Obama-Clinton-Edwards = Big momentum for Obama going into NH. Hillary better nip this there or she'll 100% have to win South Carolina and Nevada. Goodbye, JRE.
Clinton-Edwards-Obama = Hello, Mme. President. Goodbye Obama since Edwards and his supporters will hang on for awhile longer and that will limit his available pool of not-Hillarys.
Edwards-Clinton-Obama = Worst case scenario for Obama. Edwards still doesn't challenge Clinton in national polls but this may help him break through. Obama is toast.
Obama-Edwards-Clinton = If Clinton doesn't poll within a few points of the 1st or 2nd spot, huge setback for her. Makes it 100% necessary for her to win South Carolina and Nevada - never mind New Hampshire.
Edwards-Obama-Clinton = Less of a setback for Clinton than Obama win (again, if she's close), but she'd still best win 2 of the next three primaries.
dcshungu wrote on January 1, 2008 12:22 PM:The Dissembler: Which, incidentally, is why the holiday effect hypothesis is currently something a Clinton supporter should want to be true. The alternative hypothesis to explain these tracking results is that there is real momentum in favor of Obama and away from Clinton, and obviously you would not want that to be true.Of course you have offered the last defense when the polls aren't going your way: you declared this poll an "outlier". But if that is the best you can do, your candidate is not in good shape as far as polling trends is concerned.
That is a non sequitur. I would say that polls are going my way....all of them except 2 since Xmas! Wait until pollster.com or RCP update their charts and you will see. I am not saying that Clinton would win, but there is no indication right now, other the noise about the predictive power of the DMR poll, that Obama has caught fire...au contraire! The only good thing that has happened for Obama is his apparent "lead" in the DMR poll, which can sway some in IA and alter the "narrative", although most major news outlets seem a bit cautious and low key about it. CNN, I am sure, would be touting their poll, which might offset DMR's influence a bit.
Iowa is still a toss up and Edwards might even win or Obama could be third (the nature of his "constituency" is cause for concern)
Richard L. Adlof wrote on January 1, 2008 12:28 PM:Over the last six years the Register . . . Like most of corporate owed media has under gone a hard shift to the recessivist. Fascist plutocracy is wonder thing. It taints facts with its stench. Its readership has shifted accordingly . . . Therefore the pool of folk to poll has shifted.
Huckleberry and Obama are sufficently right wing toolie without being totally wackadoo. That is what this poll says. Only Obamaites and Clintonians will read into it more than is there.
Happy feasting boys and girls. Tomorrow is the only poll that counts.
dcshungu wrote on January 1, 2008 12:36 PM:Oops. My attempt to link to Campaign Diaries using HTML did not work above in the sentence :
"If you believe that there is such an effect, then you must agree with me and with
http://www.campaigndiaries.com/2008/01/final-dmr-poll-smashes-all-conventional.html
that the DMR a poll is might be outlier."
Check out Dan's analysis at the link above about the weird turn out model used by DMR. It was also my concern with this poll and others around the web have also expressed "surprise", but Ann Selzer is a reputable pollster so that the DMR projection is possible!
Grant wrote on January 1, 2008 12:37 PM:Richard L. Adlof wrote: Tomorrow is the only poll that counts.Really? And which one is that? awrbb wrote on January 1, 2008 12:38 PM:
You're confused, dcshungu. If there's a holiday effect, the proof lies in the trends. It doesn't matter who's leading, or in how many polls.
Dan wrote on January 1, 2008 12:44 PM:And we've had THREE new Iowa poll since the DMR, and all three have Clinton ahead (though the third has a huge Edwards lead once second-choices are accounted for).
moondancer wrote on January 1, 2008 12:45 PM:deschungu
You've been a loyal HRC supporter from way back. We'll find out in a couple of days whether your spin is just that or not.
DTM wrote on January 1, 2008 12:48 PM:dcshungu,
I have carefully considered your arguments, and they indicate that you simply do not understand the hypothesis in question (or you are at least pretending not to understand).
I'll state this just one more time: the proposed hypothesis only implies that the closer we are to Christmas (before or after), the more the numbers will be skewed in Clinton's favor and away from Obama. And in fact the evidence available so far supports that hypothesis.
Your reply is based on looking at whether or not Obama is leading in certain polls. But since the proposed hypothesis is only about relative effects, it implies nothing in particular about who would actually be in the lead. So, your focus on who is leading in various polls simply has nothing in particular to do with testing the hypothesis in question.
DemAC wrote on January 1, 2008 12:48 PM:A very good 7News/Suffolk University poll for Hillary Clinton from New Hampshire. Clinton is peeling away from the undecideds and the poll is also indicative of how McCain’s surge is hurting Obama. In NH (as well as in Iowa) Obama is trailing among Democrats and so he relies heavily upon Republicans and Independents, the very same group that McCain considers his base.
Keith wrote on January 1, 2008 12:50 PM:From Politico:
"I asked DMR pollster Ann Selzer whether the percentage of independents and Republicans saying they'd caucus had risen since the last survey."
"Same method. More independents saying they will definitely/probably caucus," she said in an e-mail.
http://www.politico.com/blogs/bensmith/0108/Polling_and_turnout.html
Ni Daye wrote on January 1, 2008 12:59 PM:Look at Obama, you should see the image of GW Bush. Both are packaged goods that have no substance but the consultants think they can successfully sell to the people. Remember Bush's "Uniter", "Restoring honor to the WH'. These two people are the same. They talk good but have no real experience to deliver. if Americans cannot learn from their experience of 2000, they deserve the last eight years and another eight years of same stupidity.
DemAC wrote on January 1, 2008 1:07 PM:Ni Daye,
Oh but Democrats have learned. The problem is that any win of Obama in the primaries is based on record turnout of Independents and Republicans; the very same folks that elected GWB in 2000 and 2004.
Left to their own devices Democrats have certainly learned the lesson.
dcshungu wrote on January 1, 2008 1:09 PM:DTM wrote on January 1, 2008 12:48 PM:dcshungu,
I have carefully considered your arguments, and they indicate that you simply do not understand the hypothesis in question (or you are at least pretending not to understand).
With all due respect, please go away if you do not understand this simple fact: You have no evidence for a "holiday effect." The only reason you have been pushing this canard has been to spin away Obama's poor showing most recent polls (which still continues!). For there to be a "holiday effect" as you postulate, the DMR poll that was conducted during the same time period and shows Obama ahead has got be an "outlier." That is the conclusion that one must draw if you believe in a "holiday effect." You simply must be consistent or you lose whatever credibility you still have...
I am, in fact, done with this.
Good day.
DTM wrote on January 1, 2008 1:13 PM:Of course Obama is far more experienced in most relevant areas than Bush was in 2000, and has a completely different path to this point (e.g., his father was not President of the United States).
DTM wrote on January 1, 2008 1:16 PM:dcshungu,
As I noted before, your comments indicate that you don't understand the hypothesis (or you are pretending not to).
Ni Daye wrote on January 1, 2008 1:19 PM:DTM wrote on January 1, 2008 1:13 PM:
--- Are you kidding? GW was a govenor of the second largest state of the nation for six years and his father was the former president of the US. He did have a record of working with Democrats or Dixcrats in Texas. You really think Obama is more experienced than GWB?
Your support of Obama is based on your emotion and hunch. That's very romantic but life is tough! You have to be more practical if you don't want to be hurt!
Anonymous wrote on January 1, 2008 1:32 PM:The "Holiday Effect" continues with this InsiderAdvantage Iowa poll:
Reallocated Numbers Based on Second Preferences Gives Edwards A Solid Lead
InsiderAdvantage/Majority Opinion Poll: Clinton and Edwards in Virtual Tie in Iowa
Compiled from InsiderAdvantage and Southern Political Report staff reports
December 31, 2007 — Using the same polling methodology that successfully predicted the outcome of the 2004 Democratic Caucus in Iowa, InsiderAdvantage/Majority Opinion Research has been conducting a daily tracking poll among likely voters in the Jan. 3 Iowa Democratic Caucus, and it shows a statistical tie between Hillary Clinton and John Edwards, with Barack Obama starting to lag.
Clinton has 30%, Edwards 29%, Obama 22%, with 14% committed to other candidates and 5% undecided.
The survey was conducted Jan. 28-29 among 788 likely Democratic voters in Iowa. The poll has been weighted for gender and age. It has a margin of error of plus or minus 3.4%.
Critically, Edwards was the second choice of 62% of those who supported other candidates that did not receive the required 15% of the vote. Clinton was the second choice of 21% and Obama of 17%.
Using the reallocation methodology InsiderAdvantage used in 2004 – which correctly indicated a fairly comfortable win for John Kerry – our new poll reveals that, if the caucuses were held today, the reallocated final outcome would be:
Edwards: 41%
Clinton: 34%
Obama: 25%
[...]
First off the cnn/opinion research poll is a joke an old clinton campaign donor is the owner of the said polling company. http://mydd.com/story/2007/5/29/175615/870
As for the DMR poll, it is highly unlikely that there would be that big of an influx of first time caucus goers, if you look at the internals the DMR, is predicting only 50% of the turn out to be Dems, and 40% to be independants. That don't jive with the usual 80% dem 20% indies. That said obama is relying on very high turnout which rarely if ever happens, but if he does my hats off to him he will have done something that i have never seen before, and will be deserving of this win. But i just don't see it happening.
The Other Steve wrote on January 1, 2008 1:39 PM:As far as the Republican results go. The Republican caucus works a bit differently than the Democratic one. Each candidates supporters get up and give a speech, and during this time period one get's a fair sense of how many supporters a candidate has.
Then there is a poll taken.
What is likely to happen, is if a large percentage of Republicans are very much against Huckabee, they may choose to align with whoever they think is in position to make sure he doesn't gain 1st place. So mcCain supporters could go to Romney, or something like that.
The Democratic side is a bit different as you get to reshuffle the deck.
DTM wrote on January 1, 2008 2:00 PM:Ni Daye,
The Governor of Texas actually has a more limited role than most governor, and of course being a governor does not include something like foreign policy experience. Moreover, Bush had no real public policy experience before being elected governor. So, no, I was not kidding.
dajafi wrote on January 1, 2008 2:22 PM:Why Obama leads in the DMR poll:
Thirty percent of the poll's respondents said a candidate's ability to bring about change is the most important, followed by 27 percent who said their priority is choosing a candidate who will be the most successful in unifying the country.Asked which candidate would do the best on these themes, caucusgoers most commonly name Obama.
...
Clinton receives more support from women 55 years old and older than her rivals, and she and Obama draw evenly from the pool of female caucusgoers between 35 and 54 years old.However, she trails Obama badly among women under 35, with just 15 percent to his 57 percent.
Obama's advantage among younger women reflects his decided advantage among younger voters in general. A majority of caucusgoers under 35 support Obama, more than three times the support Edwards receives from them and five times Clinton's.
...
Clinton remains the favorite of the party faithful, with support from a third of self-described Democrats. However, Obama is the clear choice of caucusgoers who affiliate with neither the Democrat or Republican parties, with roughly 40 percent of them backing him in the survey.The support from non-Democrats is significant because a whopping 40 percent of those planning to attend described themselves as independent and another 5 percent as Republican.
So the dynamic is this: older, bitter-end hyper-partisan Democrats for Clinton versus younger, less partisan Democrats and independents for Obama.
Which coalition do you think is more likely to secure a victory in November? Or to provide a political base for progressive policy change once in office?
My favorite rhetorical twitch from the Hillbots on this site is when they accuse Obama of being some kind of closet Republican. This is risible on its face, but it also constitutes a tacit admission of the fact that Obama is vastly more appealing to non-hardcore Democrats than the woman who is the living symbol of the zero-sum, scat-throwing politics of the last 15 years.
Kucinich for prez wrote on January 1, 2008 2:28 PM:If Clinton takes Iowa, it either means Iowans want us to stay in Iraq, and be forced to buy health insurance, or the corporations stole another one.
Marcia S wrote on January 1, 2008 2:34 PM:For the last week we have been subject to total bullshit and posturing for Hillary on this site by Greg Sargent, dcshungu, Colin Powpow, Tara and others. Whether they were discounting real polls and pointing to Intrade, putting value in clearly pro-Hillary outlier polls, or just plain discounting what any of the Obama supporters had to say (just like their leadership team), they took the cake for dishonesty, arrogance, condesension and nastiness.
Well...the most reliable and last poll of the contest now has Obama firmly in the lead. The Intrade polls now have Obama in the lead. And Mr. Mo is on his side. Moreover, Hillary is about to finish in third and lose NH big. Good riddance you nasty little Clinton people.
AJM wrote on January 1, 2008 2:34 PM:Some 125,000 participate in the Iowa caucuses. Some percentage of that are Republican. So whoever wins 50 percent on the Democratic side has probably at max some 40 thousand or so voters.
Michigan votes on January 15th. Some million and a half people are expected to vote. Obama didn't bother to run because as he explained to a proponent of the ealier Michigan primary he did not expect to win there.
k wrote on January 1, 2008 2:44 PM:The Gannett owned Des Moines Register front page has 3 Democratic candidates on the front page. Where's the objectivity?
roo_P wrote on January 1, 2008 2:53 PM:AJM:
Michigan votes on January 15th. Some million and a half people are expected to vote. Obama didn't bother to run because as he explained to a proponent of the ealier Michigan primary he did not expect to win there.
Er, pretty much ALL of the Democratic candidates "did not bother to run" because Michigan and Florida broke DNC rules.
DemAC wrote on January 1, 2008 3:00 PM:As, according to the DMR poll, Democrats are not choosing Obama.
Obama as the anti-Democratic candidate?
Its about the future stupid wrote on January 1, 2008 3:01 PM:AJM wrote on January 1, 2008 2:34 PM:
Michigan votes on January 15th. Some million and a half people are expected to vote. Obama didn't bother to run because as he explained to a proponent of the ealier Michigan primary he did not expect to win there.
Obama, edwards, biden, richardson all decided take their names off the ballot in michigan, DK also wanted to but got the paper work wrong. There will be clinton, dodd, DK, gravel and non-commited. The reason they droped is because the DNC issued them to be in violation of DNC rules and had their delegates pulled from the process as well as florida.
That's right roo, the DNC and RNC rules clearly state that no state with a major city in it shall have too much say in picking presidents.
hello_world wrote on January 1, 2008 3:28 PM:roo_P, you missed AJM's larger narrative. At one point, it was believed that if Hillary stumbled in Iowa, NH was set up to be the perfect firewall state for her. Even as Obama started eating into Hillary's numbers in New Hampshire this didn't bother Clinton supporters so much as long as the picture coming out of Iowa was still reasonably murky.
Now that there is a reputable pole (and basically, the most reputable one out there as far as Iowa caucuses go) that takes into account the ways that Obama has expanded the boundaries of support for the Democratic party, the new firewall state apparently now will be one that no other Democrats contesting do to party rules.
What a glorious victory that will be for Senator Clinton, indeed.
hello_world wrote on January 1, 2008 3:34 PM:The above comment should have read:
...the new firewall state apparently now will be one that no other Democrat will be contesting do due to party rules.
Happy New Year folks.
Anonymous wrote on January 1, 2008 3:35 PM:colonpowwow wrote on January 1, 2008 10:10 AM:
bm:
I won't be. I'm planning on signing up and driving at least 15 new voters (especially women) at Indiana University (my goal right now). We may not win Indiana for her, but we all want to be a part of this shining chapter in our country's progressive history.
By the way, I hear Nader's thinking of taking some time off from managing his stock portfolio online and may take another run. I'm sure there are a number of jobs open with his campaign right now so get your resume in early!
Got an e-mail from Robert Novak bragging about your ditching the Indiana thing to head for Oklahoma and get in on all the Bloomberg fun if neither party appears likely to nominate a neocon who can win more votes than he can. Man, he has a big mouth. In spite of the probability that you are not as much of a cutie as Valerie Plame---and not nearly so useful to our country---I will write them a letter of reference on your behalf if you need one.
meh, they don't let you use strike tags here, but I trust you got the picture. ^_^
CalD wrote on January 1, 2008 3:41 PM:Marc Armbinder seems to think the Register poll's turn-out model projects 200,000 will attend the democratic caucus and points out that 2004 caucuses saw 15% independents and 1% Republicans in the mix (not 40% and 5%).
I guess 200,000 would be about right unless a lot fewer Democrats turn out than last time -- like 2/3 as many, maybe -- which hardly seems likely. And David Yepsen has suggested that 2004 may have actually set the previous turnout record (122k - 124k, depending on who you believe) since 1980 estimates were probably inflated. I guess anything is possible, but some things do seem more likely than others.
Keith wrote on January 1, 2008 4:20 PM:The problem with looking at 2004 is that none of the Democratic candidates appealed to ANY of the independents or Republicans. And the Republicans didn't have a primary (Bush was the incumbent).
2004 was 2004. This year we have a much more engaged electorate, candidates that appeal beyond the traditional Democratic base and some of the most sophisticated GOTV efforts the state has ever seen. Clinton is counting on first-time caucus goers. Obama is counting on first-time caucus goers. With stakes this high, I'd be surprised if there wasn't record turnout.
Its about the future stupid wrote on January 1, 2008 4:21 PM:CalD i have no idea where DMR thinks there will be a 60+% increase in caucus goers from 2004 to 2008, the number may go to 150k but 200k i wouldn't hang my hat on that any day. The DMR poll is suggesting that as many indies and republicans will vote in the dem caucus as dems, not likely.
dcshungu wrote on January 1, 2008 4:34 PM:Marcia S wrote on January 1, 2008 2:34 PM:For the last week we have been subject to total bullshit and posturing for Hillary on this site by Greg Sargent, dcshungu, Colin Powpow, Tara and others. Whether they were discounting real polls and pointing to Intrade, putting value in clearly pro-Hillary outlier polls, or just plain discounting what any of the Obama supporters had to say (just like their leadership team), they took the cake for dishonesty, arrogance, condesension and nastiness.
Well...the most reliable and last poll of the contest now has Obama firmly in the lead. The Intrade polls now have Obama in the lead. And Mr. Mo is on his side. Moreover, Hillary is about to finish in third and lose NH big. Good riddance you nasty little Clinton people.
To Say stuff like that, when only two polls in recent nine have had Obama "leading", shows near total lack of critical thinking. Surf the web and you will realize how cautious and muted nearly every major news outlet out there seems to be about the DMR poll. The reason is simple: a model in which only 50% of Dems turn out to vote in their own party's caucuses seems improbable, although it might well happen. Had Edwards come out ahead, I am sure that the whole place would be abuzz with talks of his "resurrection"; or, we would be hearing about "inevitability" once more if Hillary had come out on top in this poll. Since the DMR poll came out, three new polls have been published, and none of them shows Obama "leading." No matter how much you wish for your candidate to win, that only two polls out of nine support the notion that he is ahead ought to give you pause. And, BTW, all of the indicators that we had pointed to before are still with Clinton, the DMR poll notwithstanding. Just look around you and you'll see it, including the latest poll that shows her opening a two-digit lead in NH. Moreover, there is virtually no negative "narrative" or noise about HRC's second place finish in this poll among the chattering classes or the MSM, despite the fact that they had considered her "inevitable" not too long ago. That is a positive for her, and the slew of IA polls showing her tied or ahead that has just come out should further mute the positive "narrative" that Obama would have gotten out of the DMR poll results. Instead, there is a "but" thrown in there every time someone tries to parse the DMR poll...
Iowa is a toss up, including a possible win by JRE, who has a tremendous organization and a solid advantage in rural areas...
So, Marsha, please show us some critical thinking. When we tout our candidate's advantage, at least we try to base it on various sources of presently available data.
Finally, regarding which poll is an "outlier", the definition of the term as used in statistics and in "hard" sciences leaves little doubt: an "outlier" is a data point that does not fit with the rest. Such a data point is usually considered suspect, as the DMR poll should be considered.
moondancer wrote on January 1, 2008 5:11 PM: Keep spinning, dcshungu. I have a favorite too. I'll vote for the emergent dem, just not enjoy it as much if its not Edwards.
I don't like spin doctors hovering for months posting crap for their candidates here. If I wanted that, I'd read some MSM whore. If you want my opinion you guys are starting to act a little panicky. The USS Clinton seems to be listing.
Hows that for science?
Dschungu must work for team Hillary on her blogging team. You gotta give her credit for chutzpah albeit no substance.
hello_world wrote on January 1, 2008 5:21 PM:@dcshungu
That's an awful lot of words to say that you don't understand what an outlier is.
The DMR poll is in no way an outlier, as it reflects the same thing every other legitimate poll is showing. That heavy turnout significantly helps Barack Obama. The DMR poll just shows that it believes that Obama's organization in Iowa is in place to such a degree that they'll be able to take advantage of this shifting political climate. Argue with that leap, but in no way does that make the pole an outlier.
Anonymous wrote on January 1, 2008 5:24 PM:A week ago dcshungu was quoting pollster.com and praising ARG. Pollster .com now states that ARG is the worst poll of them all and dcschungu has some convoluted analysis as to why we shouldn't trust DMR even though the consensus of pollsters has said that it is the most reliable, it was conducted from 12/27-12/30 and it shows Obama with the biggest mo he's had in a few weeks. Yeah, good luck dcshungu. Your gal is headed south (And I don't mean South Carolina).
Btw, I find it also funny how Michigan and NV have alll of a sudden made it onto these threads. Can you say lowering of expectations? The only lowering is Hillary's chances of winning.
Anonymous wrote on January 1, 2008 5:28 PM:As the following table shows, the Des Moines Register "Iowa Poll" conducted by Selzer and Company easily earns the highest marks, with virtually all rating it either very (36%) or somewhat (50%) reliable. The other pollsters with the highest scores are nationally known media surveys: ABC/Washington Post, the Pew Research Center and CBS/New York Times.
See: http://www.pollster.com/blogs/poll_of_pollsters_rating_the_i.php
The pollsters receiving the lowest scores are Zogby International, the American Research Group and Rasmussen Reports. In the case of Zogby, four out of five pollsters rated their surveys as not very (28%) or not at all reliable (52%).
Anonymous wrote on January 1, 2008 5:38 PM:Hey Eric - where's Greg? Waiting for a pro-Hillary poll before he gets back online?
Anonymous wrote on January 1, 2008 6:18 PM:moondancer wrote on January 1, 2008 5:11 PM:Keep spinning, dcshungu. I have a favorite too. I'll vote for the emergent dem, just not enjoy it as much if its not Edwards.
I don't like spin doctors hovering for months posting crap for their candidates here. If I wanted that, I'd read some MSM whore. If you want my opinion you guys are starting to act a little panicky. The USS Clinton seems to be listing.
Hows that for science?
So who the fuck asked you come here? Spinning and pontificating is what we do here; if that does not suit, just get the hell out, and don't let the door hit you on your way out...
You wanna be nasty? Well, two can play that stupid game.
Bye bye...
dcshungu wrote on January 1, 2008 6:21 PM:Anonymous wrote on January 1, 2008 5:15 PM:Dschungu must work for team Hillary on her blogging team. You gotta give her credit for chutzpah albeit no substance.
That great incisive analysis and well-researched rebuttal leave me speechless...
dcshungu wrote on January 1, 2008 6:33 PM:hello_world wrote on January 1, 2008 5:21 PM:@dcshungu
That's an awful lot of words to say that you don't understand what an outlier is.
Don't invite me to post again my list of my 2007 "hard" medical science publications in journals such as IEEE Engineering and Medicine and Biology or Biological Psychiatry. I do know what outliers are because I have been searching for them almost every day for more than two decades.
Now, go read the charts or look at the actual numbers at pollster.com, and then pray for your messiah...
When Magic Backed Hillary, Did Money Buy Love?
January 1, 2008 03:45 PM
In late-March 2007, Earvin "Magic" Johnson made his first ever contribution to a presidential campaign -- $2,300 to Barack Obama.
No formal endorsement accompanied the cash, but most political operatives thought that the famous Lakers point guard would be firmly in the corner of the first African-American to have a real shot at winning the presidency.
On August 28, Johnson made the second contribution of his life to a presidential candidate. This time, however, the recipient was Hillary Clinton. Johnson gave her $2,300 for the primary election and another $2,300 to use in the general election, if she won the Democratic nomination.
In addition, the contribution -- chickenfeed to someone of Magic's wealth -- was backed up with an all-star endorsement.
After Oprah Winfrey, the richest African-American in the country, announced that she would hold a September 6 fundraiser for Obama at her Santa Barbara home, Magic countered with an offer to host a Clinton fundraiser at his Los Angeles house on September 14. As the two candidates competed for the support of black luminaries, Johnson announced that co-chairs of his Hillary event included Motown Records founder Berry Gordy, musician Quincy Jones, and music executive Clarence Avant.
"Senator Hillary Clinton understands the domestic and international issues better than anyone," Johnson said in an August 12 press release announcing his fundraiser. "[Hillary] has the experience and knowledge to help lead our country and get us to a better place. We need a winner as our next president."
On December 17 and 18, Magic joined Bill and Hillary for a tour of Iowa. Speaking in Davenport, Johnson infuriated Obama supporters when he told the crowd, "You don't want somebody in there that is young or a rookie at politics" -- a direct dig at Obama.
What happened between Johnson's $2,300 contribution to Obama in March and his $4,600 donation to Clinton in August?
Over the summer, a seemingly unrelated series of events intervened.
In June, Johnson, who is an extraordinarily successful entrepreneur, began talks with billionaire Ron Burkle about forming a joint investment fund of $750 million or more to take advantage of the growing and lucrative markets in inner city and minority-owned businesses, according to Eric Holloman, president of Johnson's Canyon-Johnson Urban Fund.
Burkle has been one of Bill Clinton's closest friends and supporters. Burkle and Bill Clinton were often seen together in and around Los Angeles, and Clinton frequently stayed at Burkle's mansion, which was the scene of much-reported partying. Burkle gave Bill Clinton a lucrative consulting job as a senior advisor to his investment firm, Yucaipa Companies.
When Hillary announced, Burkle became a strong backer and encouraged Johnson to support her.
Now, spokesmen for both Burkle and Johnson flatly dispute that there is any connection whatsoever between the Burkle-Magic investment partnership and Magic's Hillary endorsement.
"There is no correlation at all between the [Johnson-Burkle] private equity fund and him backing Hillary Clinton," said Eric Holloman. "No one has told Earvin what to do since Pat Riley [former Lakers coach]. Earvin needs no more fame, no more money."
Frank J. Quintero, spokesman for Burkle's Yucaipa Companies, described as "preposterous" any linkage between planning the fund and the endorsement. "Ron and Magic have been friends for a dozen years and have supported the same candidates," Quintero said.
Johnson's support for Hillary has been apparently unaffected by a separate development involving Bill Clinton and Burkle. In late September, well after Magic Johnson's Hillary endorsement and fundraiser, the relationship between Burkle and Bill Clinton was reportedly severely strained by a September 26 Wall Street Journal article detailing some of controversial dealings by Burkle's company, Yucaipa.
The WSJ story reported that a young Italian businessman had convinced Burkle and a Clinton aide to invest millions of dollars in a controversial and poorly-run venture buying up urban properties owned by the Catholic Church. Clinton, allegedly furious at Burkle, began the process of withdrawing from Yucaipa.
"So who the fuck asked you come here? Spinning and pontificating is what we do here; if that does not suit, just get the hell out, and don't let the door hit you on your way out...
You wanna be nasty? Well, two can play that stupid game."
Sure sounds like bitterness in Hillaryland to me..
Anonymous wrote on January 1, 2008 6:42 PM:Anonymous wrote on January 1, 2008 5:24 PM:A week ago dcshungu was quoting pollster.com and praising ARG. Pollster .com now states that ARG is the worst poll of them all and dcschungu has some convoluted analysis as to why we shouldn't trust DMR even though the consensus of pollsters has said that it is the most reliable, it was conducted from 12/27-12/30 and it shows Obama with the biggest mo he's had in a few weeks.
Show me where I had praised ARG? Any poll is only as good as its methodology, so that my only basis for accepting or disparaging a survey is the quality of the underlying methodology.
If you believe that only 50% of Dems would show up to caucus in IA, as the DMR model assumed, then that is your privilege. I find that hard to believe, so question the methodology; let's leave it at that.
One questionable poll a MO does not make...
dcshungu wrote on January 1, 2008 6:50 PM:Sure sounds like bitterness in Hillaryland to me.. blockquote>Anonymous wrote on January 1, 2008 6:50 PM:LOL. I always know I have hit a raw nerve and the nail right on the head when the insults start...
I just came from surfing the net. Instead of being positive as a result of his showing in the DMR poll, the narrative has gone South (and that is not South Carolina) on Obama, Big Time.
Wow....Marc Penn must be really getting desperate if he's now posting on the TPM message boards.
Isn't it ironic that in his CYA memo of the day he praises the Zogby poll, a poll that he has slammed in the past when it didn't go his way.
What a friggin loser. Get ready for a Shrumian legacy, Marc.
Anonymous wrote on January 1, 2008 6:55 PM:Hey dcshungi-
Check out HuffPost, the front page of the NYT, and the front page of the DM Register. You must have missed them in your travels in dreamland.
Anonymous wrote on January 1, 2008 6:59 PM:Front page story at NYT right now:
January 1, 2008
Campaigns Feeling Effects of Iowa Poll
By KATHARINE Q. SEELYE
DES MOINES — Dennis Kucinich today urged his supporters in Iowa to back Senator Barack Obama as their second choice at the caucuses Thursday if his support is not strong enough to be viable in the 1,781 precincts across the state.
Mr. Kucinich said in a statement that he and Mr. Obama had something in common: “Change.” They both also have long opposed the war in Iraq.
The depth of Mr. Kucinich’s support is minimal; just one percent of caucus-goers chose him in today’s Des Moines Register poll, which showed Mr. Obama with a widened lead over his Democratic rivals.
But the timing of the Kucinich announcement -- coinciding with the poll results -- suggests a possible “poll effect,” in which marginal candidates and undecided voters can start to move toward a perceived winner.
The results of the poll continued to reverberate across the campaign trail on Tuesday as the clock ticked toward the caucuses Thursday night.
Mr. Obama seemed to have a fresh bounce in his step as he set off on his first fly-around tour of Iowa. At the first of four events on the day, he addressed an audience of more than 1,000 people in Des Moines, saying: “I think 2008 is going to be a good year. That’s what I think. I think some big things might happen in 2008.”
While Senator Hillary Rodham Clinton, President Bill Clinton and their campaign advisers publicly projected confidence and optimism, hitting the campaign trail with brio Tuesday morning, there was substantial surprise and consternation behind the scenes. Her first event, Ames, drew about 750 people.
Her advisers said that no one had predicted that the poll would show Mr. Obama as the preference of 32 percent of caucus-goers compared with 25 percent for Mrs. Clinton and 24 percent for John Edwards.
Soon after the Register poll was published Monday night, the Clinton camp sent out a memo questioning the methodology and the sampling, including the fact that 40 percent of likely Democratic caucus-goers described themselves as independents.
“A lot of people read about this poll and were totally taken aback,” said one prominent Clinton donor in Manhattan, who spoke on condition of anonymity because he did not want to be seen criticizing Mrs. Clinton’s operation.
On the Republican side, Mike Huckabee led in the Register poll with 32 percent of those polled preferring the former Arkansas governor, compared with 26 percent for Mitt Romney. That good news for Mr. Huckabee came a day after facing guffaws from a room full of reporters for showing them an ad criticizing Mr. Romney that he had pulled from the air. Today, Mr. Huckabee tried a new tack.
“I need your vote because I can’t do it, I can’t spend enough money here” to fight back with his own ads, he told a crowd of about 200 people in Sergeant Bluff, saying his victory would be a historic statement “in terms of the influence of money and the tone of presidential politics.”
Mitt Romney was in central Iowa, where he went after Mr. Huckabee for making critical comments about President Bush’s foreign policy on Monday.
“I think we should come together and recognize the great work our president is doing and not take our rhetoric or our plays from Democratic playbook,” Mr. Romney said. “This is the kind of stuff you expect of the Democrats, but it’s certainly not something you expect of a presidential contender on the Republican side.”
Mr. Obama did not dwell on the results of the poll. Privately, his aides questioned the numbers themselves, but said the trajectory matched the momentum that they were seeing in data they are collecting hourly through telephone calls and door-to-door canvassing.
“The polls look good, but understand this,” Mr. Obama told supporters jammed inside the gymnasium at Roosevelt High School. “The polls are not enough. The only thing that counts is whether or not you show up to caucus. The only thing that counts is whether over the next 72 hours you are willing to work for this.”
A fresh batch of Obama volunteers from across Iowa and the country descended on neighborhoods, passing out literature about the caucuses to remind voters of their specific precinct location for Thursday evening. As the campaign hurtles to a close here, aides believe the momentum from the poll could be a booster rocket in pushing supporters to their caucuses.
“After 10 months, it looks like it just might work,” Mr. Obama said. “It looks like it might be paying off. Our bet on the American people might just work, but we’ve got more work to do.”
In a brief interview with reporters before his plane flew from Des Moines to Sioux City, Mr. Obama said he put more stock in what he was seeing on the ground, rather than the findings of a particular poll.
“I think it’s very hard to read what’s going on except for the fact that we’ve got these great crowds with unbelievable energy,” he said. “We’re still seeing these undecideds coming to our events, which is really exciting because it gives me a chance to persuade them one last time.
“And now, I think it’s going to come down to who gets their supporters out. I’m putting my money on my organization. It’s as good and as dedicated and as intense as I’ve ever seen.”
When Mr. Kucinich ran for president four years ago, and was sagging in the polls, he urged his supporters to make a similar second-choice partnership with Mr. Edwards. At the time, aides to Mr. Edwards said that move helped contribute to his second-place finish in the caucuses.
Mr. Edwards began a 36-hour marathon campaign swing on Tuesday, starting with an event with than 500 people in Ames.
“I don’t need a poll to tell me that we’re moving,” Mr. Edwards said.
“We’re moving every single day.”
Joe Trippi, an advisor to the campaign, said he was skeptical of the poll’s findings. “It doesn’t make sense at all,” he said. “You’d have to have 220,000 people voting for that poll to be right. If that’s what’s going on, there’s no historic model for it.”
Mr. Trippi, who was Howard Dean’s campaign manager in 2004, invoked his experience in that year’s Iowa caucus, when the Dean campaign predicted unprecedented turnouts, which never materialized. “I was the guy here last time who thought it would be 200,000 people,” he said. “It didn’t happen.”
The three Democrats leading in the polls followed one another across the state, all stopping in Council Bluffs. The Clinton campaign sought to match the bad news from the poll with some good news from the donors, announcing Monday night that Mrs. Clinton had raised more than $100 million in 2007.
Still, some Clinton donors in New York who spoke on condition of anonymity said Tuesday morning that they were unnerved by the Register poll. They said that they had always been told by the Clinton high command that Iowa would be a challenge but they said they had been led to assume that Mr. Edwards would most likely come in first; they never expected an Obama blowout.
Some Clinton advisers said that if Mr. Obama has actually managed to attract a whole new crop of independent voters to the caucus process, he would be very tough to beat at the caucuses. But they expressed skepticism that the poll results would hold up, noting that the caucuses are a highly unpredictable process that would come down to whether people would actually show up at their precincts and whom they would favor for their second choice.
Mrs. Clinton’s day on the trail did not reflect the internal concerns. She came out swinging, starting off the final 48 hours of her Iowa campaign by assuring her audience in a hotel ballroom in Ames that her administration would repair the missteps of the Bush presidency.
“After seven long years of George Bush and Dick Cheney,” she said as the crowd groaned, “starting Thursday night, we are taking our country back and you all are going to lead the way.”
She was joined on the trail with her daughter, Chelsea, former Iowa Gov. Tom Vilsack, and her mother, Dorothy Rodham, who was last seen on the trail two weeks ago.
Mrs. Clinton appeared feisty yet relaxed, lacing her stump speech with jokes about drug companies and a strange technical difficulty that fed her speech back through the speakers. Hearing words that she had just spoken reverberate back through the packed room, she paused, looked up at the ceiling, and laughed.
“I was in Muscatine yesterday in the middle of going on about the Bush administration, and the microphone completely died,” she said. “I said, ‘I know they’re a little obsessed with me, but this is getting absurd.’ “
Julie Bosman, Cate Doty, Patrick Healy, Jeff Zeleny contributed reporting.
Michael A wrote on January 1, 2008 7:56 PM:Good post on the NY times. It is kind of interesting in how the obama camp reads polls vs. stalag clinton II. If this poll was the reverse and stalag clinton II was in the lead, they would be crowing about inevitability and that clinton II was the next president and all that other garbage that we have been hearing for months. Camp obama plays it down. A totally different kind of attitude and a good attitude. The clinton II people are still crowing about madam president in the face of this poll and the trends. Too funny.
By the way, in response to colonpowwow's prognostication. I bet an iowa close loss for obama and a new hampshire close loss is not devastating if it happens. It shows his viability and then a win in s. carolina. Then february 5 is up in the air. It would be better if he won both by a large margin and then after february 5 it would be all over, but he would still be viable as long as the results are close, which they look like they will be regardless, unless people see through the clinton II garbage. Edwards is toast and sucking up votes from obama. He has no cash and his phoney schtick is tiresome. He will be gone regardless of the iowa outcome very soon.
By the way colonpowwow, we might have a dem in the presidency now if all you clinton II lovers had rounded up 15 of your friends regardless of gender in the last election. Pathetic.
dcshungu wrote on January 1, 2008 8:25 PM:Anonymous wrote on January 1, 2008 6:55 PM:Hey dcshungi-
Check out HuffPost, the front page of the NYT, and the front page of the DM Register. You must have missed them in your travels in dreamland.
HuffPo is Obamaland, so no surprises there. The NYT piece illustrates what I had said. The narrative is not as glowing as it could have been. The caveat about the methodology and challenges to it are mentioned, giving the impression of "he won BUT..." sort of thing. Had Edwards or HRC won, the punditry would be hyperventilating about the possibilities for either and drawing up all kinds of outcome scenarios. The DMR poll simply isn't being talked about as one would have thought, considering the anticipation that preceded its release...It is almost a let down really.
Its about the future stupid wrote on January 1, 2008 8:35 PM:Michael A wrote about polls. Camp obama plays it down.
Of coarse he is playing it down, i do not believe their is any way possible to increase turnout from the last election by 60%, its my belief that when he is seen taking second or maybe even third, it will deflat his entire camp.
Later he also wrote; Edwards is toast and sucking up votes from obama.
This is the most rediculous statment i have kept hearing from the obama supporters. I guess in their eyes all votes not cast for clinton is somehow obamas votes, i am no fan of the clintons (bill or hill), but at this point if i had to vote for one or the other it would not be obama.
Anonymous wrote on January 1, 2008 8:37 PM:Again, nice spin...but time to face reality, dcshungu. Hillary is losing momentum everywhere and will finish in 3rd in Iowa and everyone on the ground knows it.
I suspect that Obama is going to win with more of a margin than the polls are showing. His supporters are disproportionately young, and likely do not have and/or answer land lines via which most polls are still conducted.
Hillary's supporters are disproportionately older and are the only people who have and answer land lines even when the caller ID does not recognize the number.
blackstar wrote on January 1, 2008 8:40 PM:The DMR poll simply isn't being talked about as one would have thought, considering the anticipation that preceded its release...It is almost a let down really.
-----------
you truly are in psychological denial. do i, as an Obama supporter, believe the numbers from the DMR poll mean Obama is going to win the primary in Iowa? no. neither does the Obama campaign.
but to say its not being "talked about" as though it weren't a big deal on the eve of the primary is so patently false it must either be a willful fabrication or a reflection of your total idiocy. the first page of every political blog and online paper is commenting on the results this DMR poll, and the majority of these are saying it is a significant boon for Obama.
basically, when you as a Hillary supporter come out and say things like this that are totally dishonest and/or inane, it makes your candidate look bad. and it would completely lose you credibility, assuming you had any credibility to lose. (you don't seem to)
CalD wrote on January 1, 2008 8:42 PM:Michael,
Are you kidding me? The Obama campaign started trying to spin this poll 12 hours before it came out. Whatever their reasons, they sure were acting like people expecting it to go the other way. Now they seem to be acting very much like people with a bad case of overhyped expectations on their hands.
Grant wrote on January 1, 2008 8:43 PM:Michael A: “Stalag Clinton”??? Was there not a discussion just a few threads ago about your blatant sexism; now you’re upgrading to Nazism? National Socialism!
What the heck is wrong with you? I know you have a pathological problem and hates Hillary Clinton beyond reason. But must you offend every intelligent person on this site just because you’re not right in your head? Can’t you just sit somewhere and hate for yourself???
My great grandfather was kept in a Stalag when he fought the Nazis. I can tell you that there were no caucuses, no primaries and no elections in the Stalag.
Nazism my ass. With all due respect for the eventuality that you suffer from some kind of psychological dysfunction: FUCK YOU.
Anonymous wrote on January 1, 2008 8:43 PM:She is so toast. It is only a matter of time. More from the NYT on the DMR poll:
More on the Register Poll
By KATHARINE Q. SEELYE
The Des Moines Register poll, which showed Senator Barack Obama pulling away from Senator Hillary Rodham Clinton and former Senator John Edwards, had some interesting numbers below the top line. (To refresh your memory: Mr. Obama had 32 percent, Mrs. Clinton 25 percent and Mr. Edwards 24 percent.)
For one thing, it showed women supporting Mr. Obama and Mrs. Clinton equally. If this is the case, that’s obviously bad news for Mrs. Clinton, who has based her campaign on drawing a disproportionate share of women.
Both the Clinton and Edwards campaigns say the poll is flawed. It is out of line with other recent polls and the caucuses are notoriously difficult to forecast. But it showed 32 percent of women going to both Mr. Obama and Mrs. Clinton. Only 20 percent favor Mr. Edwards.
At the same time, men favor Mr. Obama and Mr. Edwards equally, at 31 percent and 30 percent respectively. The glaring deficit is men for Mrs. Clinton; only 16 percent support her.
Other polls have also shown that men are less comfortable with Mrs. Clinton than with her male rivals. That is all the more reason her campaign needs to shore up any lagging support among women.
But Ann Lewis, a senior adviser to Mrs. Clinton who oversees the women’s vote, says
Mrs. Clinton’s support among women continues to be strong and that because the poll oversampled independents, its results are “out of whack.”
“This poll has more to do with a change in sampling technique than any change of opinion among Democratic women,” she said. “We’ve had six or eight in the last three days and all of those except one had the race very close.”
The Register poll did fall in line with some broad trends that other surveys have indicated for some time:
By big margins, Mr. Obama draws young people (56 percent of 18- to 34-year-olds), those with college degrees (39 percent), those making more than $70,000 (37 percent), and those living in urban areas (37 percent).
Mrs. Clinton bests her two rivals only among those with no more than a high school diploma (32 percent) and those who identify themselves as Democrats (33 percent). She ties Mr. Edwards for support among people 55 or older (30 percent) and essentially ties Mr. Obama among those who make less than $50,000 (28 percent, with Mr. Obama at 27 percent).
Hey Cal-
Spin this poll? You guys were spinning the ARG poll and Zogby within minutes on every blog you could post to. Isn't ironic that fat man Penn's memo today trumpeted the same Zogby methodology he ripped another asshole for last month when it didn't agree with him. Face it. No one likes your candidate. Once Edwards is eliminated his camp goes to Obama. Kucinich...to Obama. Biden...to Obama. Hillary is a goner.
Ava wrote on January 1, 2008 8:50 PM:People are so gullible arguing Clinton’s and Edwards’ talking points with respect to criticizing the DMR poll. The DMR has not changed their polling methods since 2004 or ALL YEAR. Ann Selzer, the pollster (she conducted the poll in 2004 as well), has said so 50 times. When Edwards and Clinton were ahead earlier this year they weren’t questioning the poll’s “methods” now they’re behind and they are trying to start a media war over the results of the poll. It’s hilarious.
Roberta Berlinger wrote on January 1, 2008 8:53 PM:What is most interesting about this poll is that it shows that independents and Republicans are actually likely to make the effort to come out and caucus for Barack Obama. Think about the insights this presents about the mind of the electorate right now. It is precisely why Obama leads all of the other Dem candidated in every head to head poll against every Republican opponent.
My personal view is that McCain will probably win if the GOP is smarter enough to nominate him unless Bloomberg jumps in (again, probable) and sucks some of the life out of the moderate GOP bloc. Regardless, Obama is the true change candidate in a change election and offers the Dems the best chance at retaking the WH since, well, Bill Clinton in 1992.
Anonymous wrote on January 1, 2008 9:03 PM:From Mark Blumenthal at Pollster.com:
So is the Register/Selzer poll right or wrong about a the potential for "a dramatic influx of first-time caucus goers" and independents on the Democratic side? I know our comments section, and the political blogosphere, will be alive with speculation, but we really will not know for certain until Thursday night.
What I can say is that when we polled campaign and media pollsters last week, the Des Moines Register and Selzer were the runaway winners as the most trusted Iowa poll. My hunch is that their reputation results partly from an awareness of their past success and methods but mostly to an appreciation of what is at stake for the pollsters. Selzer is a Des Moines based researcher, and this survey is easily the most important her company has conducted since their last pre-caucus poll in 2004. Under those circumstances, other pollsters trust her to sweat the details
One intriguing footnote: That final 2004 Register poll showed Kerry with 26%, Edwards with 23%, Dean with 20%, Gephardt with 18%. On the same Sunday, Zogby/Reuters also released a three-day rolling average tracker with fresher interviews (conducted Thursday to Saturday, 1/15-17) but with Edwards running five points lower (Kerry 24%, Dean 23%, Gephardt 19%, Edwards 18%). Edwards had 17-18% over the nights of interviewing that coincided with the Register pol
Oh brother, grant, your welcome. I won't say stalag clinton again. Incidentally, I was not implying nazism, but I guess it could be interpreted that way. Where were you when posters were comparing clinton II to a fascist? I never did that and I didn't see you going nuts at the time. You clinton II lovers are soooo sensative. Give me a break.
Sorry, I can't get on your bandwagon. I'm not that blind. The sexist bs is just way over the top. Sorry again, I may be alot of things but sexist isn't one of them. And, just because I am not on your bandwagon doesn't mean I hate your candidate or that I'm sexist. I just think that she is phoney and full of bs. Does that make me a sexist or mean that I hate her? Of course not, that's absurd.
Cald, I think all this crowing about polls is meaningless, especially when the poll that counts is 48 hours away. It's a dead heat right now and who knows what's going to happen. We'll see on Thursday. See, I'm not crowing about this poll and saying wow obama's won it already and hail to president obama. That's what clinton II people do.
Before, I get blasted by a bunch of clinton II lovers, I apologize for using the term stalag, which is german for camp, to identify camp clinton II. Man, the first amendment just does not exist as far as clinton II people are concerned.
Anonymous wrote on January 1, 2008 9:11 PM:Dick Polman at the Philly Inquirer today:
For the Obama candidacy, size matters
If the newly-released Des Moines Register poll is correct - and this survey is widely respected for its readings of the Iowa mood on the eve of the Iowa caucuses - we could soon be witnessing an historic moment in presidential politics.
The poll, posted last night, reports that an African-American candidate will top the Democratic field in this overwhelmingly white state, buoyed by an unprecedented wave of first-time caucus voters - many of them independents (and even some Republicans) who appear determined to re-register on the spot as Democrats, just for the opportunity to support Barack Obama. Which is why the final Register poll projects a seven-point Obama victory over Hillary Clinton (35 percent to 27 percent) in the Democratic caucuses Thursday night.
Other new surveys dispute this point spread, but the Register poll, with its home-field advantage, arguably has a better feel for the dynamics of the likely Democratic turnout. The numbers, if true, are provocative: Forty percent of the folks planning to attend the Democratic caucuses describe themselves as independents, and five percent say they are Republicans. And Obama is the clear favorite of these potential newcomers.
Could this actually happen, that 45 percent of the Democratic caucus-goers might be outsiders lured to the event by Obama? It remains to be seen, of course, whether these people will actually follow through on their vows to participate, but, at minimum, their strong interest in Obama is further evidence that he has potential crossover appeal - the kind that is crucial to winning a November election. So keep an eye on the turnout figures Thursday night; the larger (and more politically diverse) the turnout, the better it is for Obama. Size matters.
In 2004, when John Kerry finished on top, roughly 80 percent of the Democratic caucus participants described themselves as party regulars. If the final Des Moines Register poll is correct, and Obama tops the candidate field thanks to an expanded turnout, of which only 55 percent are party regulars, he can make the case to the people of New Hampshire (who vote in their primary next Tuesday) that he's the most electable Democratic candidate...thereby keeping the heat on Hillary.
No wonder the Hillary spinners are trying so hard to dismiss the Des Moines poll. Right on schedule this morning, my email box contained a missive from Mark Penn, Hillary's pollster/strategist, who naturally takes issue with the poll's turnout projections. Penn points out that only 15 percent of the 2000 caucus-goers were independents, and that only 19 percent were independents in 2004. He complains that the Register pollsters "are depicting an unprecedented departure from historically established turnout patterns," and that if we look at the '08 Iowa horserace through the prism of the established patterns, then Hillary is in darn good shape.
Yet here's what he is essentially arguing: the smaller the turnout, and the more it is dominated by Democratic regulars, the better it is for Hillary. Penn is not disputing the poll data showing Obama's popularity among independents; he's only arguing that the independents won't show up en masse to participate. (Or so he hopes, since he's basically signaling that she's at risk if the turnout is large and politically diverse.)
In other words, he implicitly acknowledged Obama's crossover appeal, which is hardly a ringing advertisement for his own candidate's electability.
Ava wrote on January 1, 2008 9:21 PM:Even if we assume 2004 turnout patterns, Hillary is still saying that under that scenario she only beats Obama by 2 points. Obama has done an amazing job bringing in first-timers and independents in Iowa, so even if the DMR's projections are only partially right, it still looks good for an Obama win, particularly with Kucinich supporters helping him out in many precincts. This is an historic election in many ways. I think there will be a nice turnout.
CalD wrote on January 1, 2008 9:25 PM:Hey Bobby,
I don't know what you mean by "you guys." I'm not in the habit of spinning polls myself. As someone who spends more time than is probably healthy poring over political polls, I do enjoy discussing them. I'm not above pointing it out when someone says something stupid about one, but "spinning" would kind of be the opposite of that.
I'll also guess you must have missed all the Obama kidz jumping all over the latest CNN poll today with charges of intentional bias. That's something you can pretty much set your watch by when a poll comes out that's unfavorable for Barack Obama, but also it's something I rarely hear any of the Clinton or Edwards fans doing it.
DTM wrote on January 1, 2008 9:29 PM:A quick note on "outliers":
A basic problem with comparing polls from different pollsters in Iowa is that there is no consensus likely voter model for the Iowa caucuses. As a result, different Iowa polls are not necessarily trying to sample the same population. And if they are not trying to sample the same population, then they can't be easily used to identify "outliers". That is because any differences in the results between contemporaneous Iowa polls may be the result of different likely voter models and thus different target populations, as opposed to random sampling error.
Accordingly, it is fair to wonder whether the likely voter model used by the DMR poll will prove accurate this year (and as Blumenthal notes, we won't really know until Thursday night). But that is not grounds for calling this poll an "outlier".
Bobby wrote on January 1, 2008 9:31 PM:Hey Cal-
See Marc Penn's memo today and the many posts by dschungu, colinpowpow, you, Kefa and Greg Sargent, among others, on this blog.
CalD wrote on January 1, 2008 9:43 PM:Hey Anonymous,
You should really stop posting entire newspaper articles in the comments section of other people's blogs. Ever heard of copyright laws? It also costs our gracious hosts more money for BW. It loads down servers unnecessarily. It slows down page loads. It interrupts the flow of conversation for other users and makes them do a lot of extra scrolling. And of course one can never rule out a little creative editing on the part of the person posting. In short, it's extremely poor etiquette.
We get that you know how to copy and paste. It's great that you've mastered that skill so early in life. Now try demonstrating that you know how to read and analyze. Instead of posting an entire article, summarize what you think is important about it, excerpt a relevant quote or two maybe, and include a link to the original source so that people can easily read it in its original form if they're so incline.
What kind of name is Anonymous BTW? Sounds Greek.
CalD wrote on January 1, 2008 9:48 PM:Bobby,
I'm not responsible for anything any of the other people you mentioned and as I said before I am not in the habit of spinning polls. They are what they are, an inexact but useful science -- or perhaps, more of an art in the case of polling the Iowa caucuses.
Carol wrote on January 1, 2008 9:57 PM:CalD, Isn't that the GOP tactic? Attack the messenger when you can't stomach the message? Anon gave full attribution and the articles provide much insight for some of us who don't have the time to surf for them.
Cheers!
dcshungu wrote on January 1, 2008 10:02 PM:Regarding those who hyperventilate about how Seltzer surveys are the most accurate in predicting the outcomes of IA caucuses, the following should give make you think and at least conceded that there is the likelihood that this poll may be, yes, an "outlier." Over at pollster.com someone had made the following astute observation/comment:
I checked the link you provided for the "accuracy" of DMR's 2004 poll. And what I saw was that their "likely voters" poll was just as flawed as everyone else -- only their "definite voter" poll proved better than other polls. And since the numbers for the poll you are citing are of "likely voters", not "definite voters", the idea that this poll is somehow more "reliable" than others doesn't make a whole lot of sense to me.
...a case of comparing oranges and apples and calling concluding that it is all the same...
That, coupled with the "unorthodox" model in the DMR poll, makes this poll even more suspect. Though my objections are based on these glaring "abnormalities", I am sure it will seen as "spin." So be it...
These articles are very interesting. Dc and the other clinton II people have been crowing that her iraq war vote and iran war vote were positioning by her to be "centrist." Well, if she was so "centrist" wouldn't independents and republicans be flocking to the "centrist" candidate? Seems to me the centrist and uniter for the country is obama, not clinton II.
Eleanor M wrote on January 1, 2008 10:10 PM:dcshungu,
The issue is that this year in Iowa is a lot different than others, particularly given that the turnout is expected to be over the top and many more disenchanted Independents and Republicans, based on surveys and observance from folks on the ground, will show up. And they're not going to show up for Hillary. They are discontent and want change.
The only folks who seem to be critiquing the Selzer polls are those from the Clinton and Edwards camps or supportive of them. All of the folks who have covered the caucuses for years put much more value in their methodologies than any others. The poll on Pollster that shows they having the unanimous support from other pollsters (with Zogby, ARG, etc. in dead last) is evidence of this. You might not agree with it because you are a Hillary staff member but it is pretty obvious to most that Obama is going to win and may win very big.
Anonymous wrote on January 1, 2008 10:12 PM:She doesn't know what she is and Marc Penn's polls seem to tell her a different thing each week. At least the Clinton 1 campaigns knew how to stay on message.
dcshungu wrote on January 1, 2008 10:22 PM:DTM wrote on January 1, 2008 9:29 PM:A quick note on "outliers":
A basic problem with comparing polls from different pollsters in Iowa is that there is no consensus likely voter model for the Iowa caucuses. As a result, different Iowa polls are not necessarily trying to sample the same population. And if they are not trying to sample the same population, then they can't be easily used to identify "outliers". That is because any differences in the results between contemporaneous Iowa polls may be the result of different likely voter models and thus different target populations, as opposed to random sampling error.
Accordingly, it is fair to wonder whether the likely voter model used by the DMR poll will prove accurate this year (and as Blumenthal notes, we won't really know until Thursday night). But that is not grounds for calling this poll an "outlier".
Nonsense. Despite differences in methodologies, which might account for the so-called "house effect", every poll is trying to "measure" the same "quantity": public opinion. So, if you have a bunch of polls and all of them show one TREND (e.g., UP),and then you have just one poll that shows a different TREND (e.g., down) there should be a strong suspicion that the latter poll is "measuring" something else altogether. Its results do not fit with the rest and therefore it would be considered an "outlier." [note that this says nothing about "accuracy", which we really know little about at this point].
By the way, I see where you tried to peddle your "holiday effect" theory at the pollster's discussion on this DMR poll but got ignored. Nevertheless, I think that Blumenthal did debunk your purported "evidence" for the existence of a "holiday effect" in the DMR poll two-day trends:
I have not tried to find that Yepsen quote. However, the numbers in question appear in a chart at the very bottom right of the Register story on the Democratic sample.
The caption reads:
"During four days of polling, Hillary Clinton's two-day rolling average declined slightly while support for Democratic rivals Barack Obama and John Edwards rose somewhat."
The changes on the chart from the first two days to the last two days of calling are:
*Obama +5 (29% to 34%)
*Edwards +3 (23% to 26%)
*Clinton -4 (27% to 23%)
However, if I assume 400 interviews in each two-day subgroup, none of these changes achieve statistical significance (the p values I get are .13, .32 and .19 respectively, and that assumes simple random sampling).
Posted by: Mark Blumenthal | January 1, 2008 9:43 AM
The prosecution simply must now rest, your honor...
DTM wrote on January 1, 2008 10:26 PM:dcshungu,
You should have fact-checked that comment. Here is the post in question:
http://www.pollster.com/blogs/how_accurate_were_the_iowa_pol.php
On the one hand, it is true that the "definite voter" result was closer to getting Kerry's percentage right, and in general had a lower mean error. On the other hand, it was actually Selzer's "likely voter" model that got the order of finish among the top four correct--Selzer's "definite voter" results had Dean beating Edwards. And it was correctly predicting the order of finish in 2004 that is usually cited as evidence of the DMR poll's accuracy (note, for example, that every other poll in the chart had Dean beating Edwards).
All that said, I agree there is no definitive evidence the DMR poll is going to be better this time than the other polls at predicting the order of finish.
Anonymous wrote on January 1, 2008 10:27 PM:Keep dreaming and drinking the Marc Penn Kool-aid dschungu. Must be tough facing reality after thinking it was inevitable all year long. Fact is that the more people engage and get to know her, the more they realize that she is not a leader or someone they desire to be their President.
roo_P wrote on January 1, 2008 10:33 PM:Carol,
I agree with CalD on this topic (and apparently pretty much nothing else): even attributed full articles are out of bounds both for being disruptive as well as being outside Fair Use.
Link to it -- and do not paraphrase it.
dcshungu wrote on January 1, 2008 10:34 PM:Bobby wrote on January 1, 2008 9:31 PM:Hey Cal-
See Marc Penn's memo today and the many posts by dschungu, colinpowpow, you, Kefa and Greg Sargent, among others, on this blog.
I try to discuss polls on the basis of their "scientific" merit but you wouldn't know it since you are too busy ascribing malicious intent to anyone who appears like he might be getting in the way of your "messiah's" pre-ordained anointment. I have no such power, as we live in what might still be considered a democracy...
I would probably be better off posting my comments on polls exclusively at pollster's where the level of discussion is more mature, erudite, and largely dignified... I had missed their discussion earlier on the DMR poll. It was quite interesting as well as enlightening.
DTM wrote on January 1, 2008 10:36 PM:dcshungu at 10:22,
You start with a false premise: that all the polls are trying to "'measure' the same 'quantity': public opinion."
In fact, the polls are trying to measure voter preference within a very limited subset of the public, namely people likely to attend the Iowa caucuses. And again, that is precisely the problem with trying to compare Iowa polls: different pollsters have different likely voter models for the Iowa caucuses.
As for the holiday effect hypothesis: again, you do not seem to understand the hypothesis. Blumenthal (notably in a post prior to mine) was addressing whether or not the DMR tracking data supported a hypothesis that there was a real trend in support for Clinton, Obama, and Edwards. But as I have tried to explain to you many times now, the holiday effect is not a real trend.
Now, it is true that taken alone, the tracking data from the DMR poll is probably equally consistent with a hypothsis that these results were just random sampling variations. But the DMR poll is only part of the evidence we have to support the holiday effect hypothesis.
All that said--I agree we still do not know for sure whether there was a holiday effect along the lines of what Blumenthal hypothesized. But the evidence is mounting.
Anonymous wrote on January 1, 2008 10:36 PM:The only "pre-ordained" messiah was Hillary by Hillaryland. Unfortunately, their pre-ordainment is not happening as planned.
roo_P wrote on January 1, 2008 10:41 PM:dcshungu,
I say this as an engineer: you are a typical engineer projecting your unrelated knowledge to areas you know f-all about.
Your "hard science" credentials are worth exactly nothing when it comes to polling, and you have not established yourself as someone who intimately understands the art -- not science -- of polling and its limitations.
First and foremost you would probably want to start actually reading the poll data before analysing it.
Not that I am a poll expert or anything. But neither are you.
Susan wrote on January 1, 2008 10:46 PM:If Hillary's strength is "experience" from having been around Bill and in the Senate, how could she not know that Musharraf wasn't running in the upcoming Pakistani elections? She made the comment today on the stump.
Excuse me but isn't this something that Mike Huckabee would say? She was so quick to trot out pictures of her and Chelsea with Bhutto last week yet doesn't even know that these are parlimentiary elections and that Musharrraf was reelected earlier in the year. Puh-leazze. This truly exemplifies the difference between Bill and Hillary.
Good for Biden. Today:
"We have a number of candidates who are well-intended but don't even understand Pakistan," Biden said.
Biden did not name his rivals, but made clear that his main focus was Sen. Hillary Clinton of New York.
"One of the leading candidates said — God love her," Biden began ...
"But to say Musharraf is up for election! Musharraf was elected — fairly or unfairly — president six months ago. It's a parliamentary election" that is coming up in Pakistan, Biden said.
Yes, if Obama had made a gaffe like this the HRC campaign would be screaming about his lack of "experience." Of course, HRC's experience goes unquestioned no matter what.
CalD wrote on January 1, 2008 11:01 PM:I just checked Marc Armbinder's math regarding the turnout model of the DM Register poll and I can see now where he's getting the 200,000 figure.
If you start with the estimated attendance for the 2004 Democratic caucuses of 124,000, and the 2004 percentages I've heard kicked around all day of 15% of Independents 1% Republicans and the rest Democrats, then the number of Democrats in 2004 was:
124,000 * 0.84 = 104,160
So now, according to Seltzer, we're going to have 40% Independents and 5% Republicans. That leaves 55% Democrats. If you assume that they'll get at least as many Democrats as last time (and I've yet to hear anyone argue we should expect any less) then:
Independents = 104,160 * (40/55) = 75,753 and
Republicans = 104,160 * (5/55) = 9,469
Add up the three groups and it comes to 189,382 -- or 186,327 if you use the more conservative 2004 attendance estimate of 122,000. So Armbinder might have been just a tad high if you assume the number of Democrats were constant, but not much. If Democratic attendance were up 5-7% over 2004, which hardly seems out of the question either, you'd be right there.
dcshungu wrote on January 1, 2008 11:07 PM:Michael A wrote on January 1, 2008 10:04 PM:These articles are very interesting. Dc and the other clinton II people have been crowing that her iraq war vote and iran war vote were positioning by her to be "centrist." Well, if she was so "centrist" wouldn't independents and republicans be flocking to the "centrist" candidate? Seems to me the centrist and uniter for the country is obama, not clinton II.
Michael A: Maybe you have not yet heard. There is a lefty revolt of some sort going on right now, as they've finally realized what I have been trying to say for some time now: You and Obama do not have anything in common; he has been running as a Repub and that is why they love him! Remember that he wants to "fix" social security, is scaring the voters by claiming that mandates would force people who can't afford insurance to get one anyway, that the UN is a useless organization? Well, Markos has had enough and just announced that he's had a change of heart about voting for Obama, which he'd announced about a week ago he was intending to do. He tells us why below and at the end of the link...Go on, read it and then follow the link...
The Obama close by kos Tue Jan 01, 2008 at 10:52:11 AM PST You know, I was going to vote for Obama and even announced that a week or so ago. But this is a great example of why it's best to wait and see how things shake out. Not being blinded by candidate worship, it's easier to sniff out the bullshit. And you have to have your head stuck deep in the sand to deny that Obama is trying to close the deal by running to the Right of his opponents. And call me crazy, but that's not a trait I generally appreciate in Democrats, no matter how much it might set the punditocracy's hearts a flutter. I don't get to vote for another month, so we'll see how the next four weeks go. Certainly none of these guys have earned my vote yet. Amazingly enough, none of them walk on water, no matter what their frenzied supporters might think.
Cheers!
DTM wrote on January 1, 2008 11:14 PM:CalD,
I suspect you are right on where the math is coming from.
But as an aside, I have seen it suggested that there may have been a trend in favor of independent registration since 2004, meaning that some current independents in Iowa are people who were registered as Democrats in 2004, or similarly for new registrants would likely have registered as Democrats in 2004 (according to the registration patterns at that time). If all that is true, it would actually support a hypothesis that the gross number of registered Democrats attending the caucuses could go down without turnout being lower, since the pool of registered Democrats could be smaller.
DTM wrote on January 1, 2008 11:16 PM:dcshungu,
And of course Dennis Kucinich, a noted Right Winger, just announced he was urging his supporters to make Obama their second choice in Iowa.
CalD wrote on January 1, 2008 11:19 PM:dcshungu:
I was thinking to myself earlier that Obama was starting to sound a lot more DLC than anyone I'd heard in a long time. If he keeps it up, maybe Joe Lieberman will endorse him.
CalD wrote on January 1, 2008 11:22 PM:DTM: Got any hard numbers on party ID or Registration in Iowa or just blowing smoke out your ass to try and cloud the issue?
dcshungu wrote on January 1, 2008 11:22 PM:Anonymous wrote on January 1, 2008 10:27 PM:Keep dreaming and drinking the Marc Penn Kool-aid dschungu. Must be tough facing reality after thinking it was inevitable all year long. Fact is that the more people engage and get to know her, the more they realize that she is not a leader or someone they desire to be their President.
Why do people keep posting this garbage over and over again, and not try to use whatever gray matter they might have between their ears? Despite every metric that shows that Clinton remains the one to beat, they keep making these asinine statements without bothering to provide a single rationale for their belief. It seems to me that this is a striking reflection of their puerile fascination with Obama: No critical thought at all...just "gut feeling" or "intuitive" determination that he is the "messiah"...
Why are you "Anonymous" anyway, if you are so sure of yourself?
Get lost...
DTM wrote on January 1, 2008 11:24 PM:CalD,
As I previously noted, this is a point I have seen other people suggest. So, in fact I don't have any numbers on party registration in Iowa.
By the way, once again, your tactics won't work on me.
CalD wrote on January 1, 2008 11:28 PM:By the way, once again, your tactics won't work on me.
My tactic of pinning you down and not letting you squirm away or change the subject when you start blowing smoke out your ass? I disagree. I think it works just fine.
DTM wrote on January 1, 2008 11:42 PM:CalD,
No, of course I meant the flamebait (your profanity, attempted insults, and so on) just won't work on me. In contrast, I am always happy to have a substantive discussion.
dcshungu wrote on January 1, 2008 11:43 PM:DTM wrote on January 1, 2008 10:36 PM:dcshungu at 10:22,
You start with a false premise: that all the polls are trying to "'measure' the same 'quantity': public opinion."
In fact, the polls are trying to measure voter preference within a very limited subset of the public, namely people likely to attend the Iowa caucuses.
LOL. That is understood, DTM. Polls are "inferential statistics", where you sample a representative segment of the population and then draw conclusions that you hope would apply the entire population. Is there any other way of doing polls, short of doing the impossible task of asking every single person in the entire population what they think? Where exactly is the error in my premise, other than your limited understanding of how these things are done? There is not that much difference between polling public opinions and, for example, conducting a clinical trial to assess the efficacy of a drug, except for the fact that the latter often has objective markers of therapeutic response and efficacy and is, thus, more quantifiable. That is the sort of things that I have been doing for a long time, so, once more, please spare me the lectures...
I had already rested my case, and the judge tells me that the verdict is..."guilty of spinning polling data to suit a pre-determined narrative!"
Anonymous wrote on January 1, 2008 11:44 PM:dschungu, desperation is not becoming, for you, Hillary or Marc Penn. Don't worry, you'll get a vacation soon enough.
Anonymous wrote on January 1, 2008 11:45 PM:Big Obama endorsement coming in the next couple of days.
DTM wrote on January 1, 2008 11:51 PM:dcshungu,
Again, the point you seem to be missing is that "the entire population" being sampled is defined by the likely voter model. So if two different pollsters use two different likely voter models, they are actually sampling two different populations.
Maybe this will help. Consider an extreme case offered for illustrative purposes: Pollster A thinks the only likely voters in Iowa are men. Pollster B thinks the only likely voters in Iowa are women. So, Pollster A screens out all women and reports only the preferences among men. Pollster B screens out all men and reports only the preferences among women. Given this setup, Pollster A and Pollster B will be sampling two entirely separate populations.
Now in the real world, the populations defined by the various likely voter models are going to be overlapping, not entirely separate. But the basic point remains the same: the likely voter model defines the population being sampled, and so different likely voter models imply different populations.
dcshungu wrote on January 2, 2008 12:00 AM:DTM wrote on January 1, 2008 11:16 PM:dcshungu,
And of course Dennis Kucinich, a noted Right Winger, just announced he was urging his supporters to make Obama their second choice in Iowa.
A Non sequitur that says nothing about the real revolt out there among the "natives"... Take a pick as to why Kucinich did it:
As for Kucinich's Sloppy Seconds...CalD wrote on January 2, 2008 12:10 AM:It's interesting not just because the first-tier candidates are looking for all the second-choicers they can get at this point. But because Kucinich sent his non-viable supporters to John Edwards in 2004. Since then, of course, Edwards has only moved closer to Kucinich politically and ideologically--on the war, health care, corporate power.
On the other hand, Obama is the only first-tier candidate who opposed the war from the begininning, so maybe that's where the affinity comes from.
--Noam Scheiber, TNR
No, of course I meant the flamebait (your profanity, attempted insults, and so on) just won't work on me. In contrast, I am always happy to have a substantive discussion.
Poor thing. My profuse apologies if I offended your tender sensibilities DTM. From now on I'll refer to it as blowing smoke out your backside. Or better still, your posterior.
Anyway, I'm all for substantive discussion so by all means, produce your hard numbers or authoritative sources to back up your earlier assertions and let's have at it. Otherwise DTM, given your record on intellectual honesty (or lack thereof), I'm afraid we'll have to assume you're just blowing smoke out your posterior until proven otherwise.
CalD wrote on January 2, 2008 12:15 AM:200,000 turning out for the Iowa caucuses sure would be something to see though (if you happen to be some kind of political geek, I guess).
dcshungu wrote on January 2, 2008 12:22 AM:Anonymous wrote on January 1, 2008 11:45 PM:Big Obama endorsement coming in the next couple of days.
How big? LOL. Bill Clinton endorsing Obama? It'd better do tomorrow if it is designed to have any impact on Thursday...
DTM wrote on January 2, 2008 12:24 AM:CalD,
I understand that what you want to do is provoke some sort of emotionally-charged response from me--that is the nature of flamebait. But as I have noted, such tactics don't work on me. So, in fact I am not offended by your flamebait. I am just letting you know that these efforts are wasted on me.
Anyway, as I noted twice before, I was just relating to you an idea that I had seen someone else propose. As I noted once before, I personally have no numbers to share with you on this issue. So, I am not vouching for the accuracy of the claim, although I do think it is an interesting possibility. Perhaps, however, you do not find it as interesting as I do.
DTM wrote on January 2, 2008 12:44 AM:dcshungu,
By the way, our "outlier" conversation above has just been mooted a bit by the latest Strategic Vision poll, which shows Obama +5 over Clinton in Iowa. So even if we ignored the likely voter model problem, it appears the DMR poll is no longer an "outlier".
Interestingly, this is the second night in the row this sort of mooting has happened. Last night, your argument was that no recent Iowa poll showed Obama with a statistically significant lead, and then the DMR poll came out. Tonight, your argument was that the DMR poll was the only such poll and therefore it was an "outlier", and now the Strategic Vision poll has come out.
It is probably too late for us to unwind this conversation and start over, But I will just note again that in some sense the holiday effect hypothesis is now in Clinton's favor, insofar as it would suggest these most recent polling trends are not real, but rather just a gradual unwinding of the holiday effect. And for good or ill, I for one am not tempted to change my mind about that.
blackstar wrote on January 2, 2008 12:53 AM:As for Kucinich's Sloppy Seconds...
------------
yeah, Kucinich and Obama both opposed the biggest American foreign policy mistake in a century, perhaps in our entire history.
where did your candidate stand on that issue? she voted her support, right?
and i like the "criticism" of Obama that asserts he's really a right-wing conservative in disguise, using the same tired 2 or 3 examples over and over again, long after his rationale for taking those positions has been completely and adequately outlined. that's pretty compelling to thinking voters who are following the campaigns.
wait no, its not.
blackstar wrote on January 2, 2008 12:57 AM:200,000 turning out for the Iowa caucuses sure would be something to see though (if you happen to be some kind of political geek, I guess).
------------
yeah, "the political geeks" who actually follow the election process with something approaching critical thinking skills and a sense of of how important a moment in American history this election is for us to get right.
in other words, not Hillary supporters.
Anonymous wrote on January 2, 2008 12:58 AM:DTM wrote on January 2, 2008 12:44 AM:dcshungu,
By the way, our "outlier" conversation above has just been mooted a bit by the latest Strategic Vision poll, which shows Obama +5 over Clinton in Iowa. So even if we ignored the likely voter model problem, it appears the DMR poll is no longer an "outlier".
Big spin! I thought you were just found guilty of doing that! Make it moot because now Obama looks better! Three of ten polls show him "leading" and, therefore, the "holiday effect" is moot and so is the "outlier" issue... Double LOL. I am happy with these polls because they kept the narrative from switching to Obama's favor as a result of the DMR poll. If you're happy, I am happy.
Bonsoir et bonne nuit!
hello_world wrote on January 2, 2008 1:01 AM:My tactic of pinning you down and not letting you squirm away or change the subject when you start blowing smoke out your ass?I think we see an awful lot of that from all sides here. I actually find DTM to be one of the most factual and reasonable posters that you see over here. He does a good job with both representing his candidate of choice, and just engaging in a civil on-topic conversation without resorting to negativity or blatant baseless spin.
It's been fun reading the back and forth here as we head into the final hours.
DTM wrote on January 2, 2008 1:21 AM:dcshungu (assuming that was you),
Actually, the Strategic Vision poll helps confirm the holiday effect hypothesis. The previous SV poll taken 12/26-12/27 had Obama at 30 and Clinton at 29. This 12/28-12/30 version has Obama at 32 (+2) and Clinton at 27 (-2). Again, that is the pattern predicted by the holiday effect hypothesis: as Christmas recedes into the past, Clinton should be trending down, and Obama trending up, until the entire holiday effect is unwound.
Again, at some point I think you are going to realize this hypothesis is actually becoming a lifeline for your candidate. But it might be asking a bit much for you to admit that here.
DTM wrote on January 2, 2008 1:51 AM:CalD,
Again, perhaps you aren't interested, but I just happened across where I read the idea in question before I related it here. See this thread:
http://marcambinder.theatlantic.com/archives/2008/01/kucinich_urges_supporters_to_c.php#comments
It is Ava's comment at 9:43.
CalD wrote on January 2, 2008 2:07 AM:DTM wrote: I understand that what you want to do is provoke some sort of emotionally-charged response from me... [blah, blah, blah]
The only response I want from you, DTM, for the third or fourth time now, is for you to provide hard numbers or authoritative sources to back up the assertions that I called you on in the first place. I don't understand why that should be so hard, here in the information age (unless of course you were just blowing smoke).
CalD wrote on January 2, 2008 2:19 AM:DTM: Did you actually go and read Ava's source? I'm going to guess not, since it obviously doesn't actually back up your original assertions. There's nothing there to support the notion of a massive increase in independent voters or a massive decrease in Democratic party ID in Iowa in the past four years.
CalD wrote on January 2, 2008 2:55 AM:RE: Strategic Vision and the Holiday Effect. There's a term for a +/- 2% difference between two numbers in a poll with a +/- 4% margin of error. That's properly referred to as a statistically insignificant difference.
Those two Strategic Vision polls are back-to-back two-day polls taken over a contiguous 4-day period, BTW. They're also the same sample size so you can just average them together unweighted to get a single 4-day poll with an MoE of +/- 3%. But it still comes out as a 3-way statistical tie.
Kefa wrote on January 2, 2008 6:27 AM:the Rep.s plan is to re-reg as Dems on the spot to help BO's numbers. VRWC might work in Iowa but not in NH. We're in it for the long haul.
DTM wrote on January 2, 2008 8:22 AM:CalD,
As I have now noted multiple times, I have no numbers involving party registration in Iowa, and I don't vouch for the accuracy of Ava's suggestion. Frankly, I don't get why you keep asking me for something which I have repeatedly said I do not have.
On the holiday effect: again, taken alone, any of these trends in individual polls might be consistent with just random variation (although it seems to me the trend in the ARG polls was statistically significant). But they are also consistent with a holiday effect. And as we get more and more such polls, the cumulative evidence more and more supports the holiday effect hypothesis.
In fact, if you look closely at pollster's Iowa charts, you can see the black trendline (the average of the last five polls) fits the predicted pattern, and the red trendline (the more sensitive regression) appears to me to have just started to turn down for Clinton and up for Obama. And all this is without yet including the Strategic Vision poll.
Anyway, I agree that we still do not know for sure if all this was a holiday effect, or if we are seeing real trends, or if it was just random variation. And we probably won't ever know for sure, because the series is about to be cut off by the actual caucus. Still, I do think there is quite a bit of evidence at this point to support Blumenthal's hypothesis, and it would be quite a coincidence if the real or random trends just happened to fit the exact pattern predicted by that hypothesis.
Michael A wrote on January 2, 2008 8:39 AM:Ditto hello world, DTM is one of the best posters on the site. I have learned quite alot from him over time, which in my view is the purpose of these types of blogs. The vitriol and attacks are a waste of time and annoying. Information exchange is what is important, which obviously is a problem for clinton II people. No information, just attacks.
Michael A wrote on January 2, 2008 8:53 AM:Dc, funny post, but I don't personally believe that the daily kos is the be all end all of the left and in fact markos probably fits alot of the insults that you have strewn at me over the months more than I do. I am with 70% of the population against the war in Iraq, which you interpret at being leftist. I am for universal health care to help people and america's businesses compete. I believe in helping people, which on its face is not leftist. I would consider myself a moderate, which is what obama is.
