Polling the New Hampshire Primaries: What Happened?

There was no shortage of polls going into the New Hampshire primary in 2008 and it looks like we all missed the mark on the Democratic side. This will require a lot of scrutiny in the coming days and weeks, but here are some initial thoughts on what has been happening:

1. According to the exit polls, 18% of the voters said that they made up their minds on primary day. That is just an unprecedented number. I have polled many races, especially close ones, where 4% to 8% have said they finally decided on their vote the day of the election and that can wreak havoc on those of us who are in the business of capturing pre-election movements and trends. But nearly one in five this time?

2. It looks like the always feisty voters in both Iowa and New Hampshire have rejected pre-election coronations. In the case of Iowa, Democratic voters said that Mrs. Clinton is not inevitable, while in New Hampshire they were not ready to endorse the Obama train without checking the engine.

3. The compressed schedule of the two events may have had an impact. Normally the winning candidate gets an initial big bounce out of Iowa, and then plateaus. Then the next primary race begins. With less than five full days, Obama got his bounce in New Hampshire, then the settling down period began on the last day -- under the radar screen.

4. My polling showed Clinton doing well on the late Sunday night and all day Monday -- she was in a 2-point race in that portion of the polling. But since our methods call for a three-day rolling average, we had to legitimately factor the huge Obama numbers on Friday and Saturday -- thus his 12 point average lead. Unfortunately, one day or a day-and-a-half does not make a trend and we ran out of time.

5. Going into the New Hampshire primary, we certainly did see Clinton holding on to a significant lead among women and older voters. But we were focusing on Obama's massive lead among younger and independent voters. We seem to have missed the huge turnout of older women that apparently put Clinton over the top.

6. We expected that Obama would receive the lion's share of independents and drain the Republican primary of these voters. It now appears that, perhaps with a sense that Obama had a lock on the Democratic side, independents felt free to vote on the Republican side and reward their hero, John McCain.

We will pour through the data and try to come up with something more definitive, but those are my early observations. There is much speculation that Senator Clinton's crying incident may have offered voters -- especially women -- a peek at the human side of someone who is often seen as scripted. I think she also scored points during the ABC debate Saturday night when she declared, amid a discussion about the country's desire for a change in direction, that electing a woman would represent a big change in itself. Her numbers did go up in that last 24-hour period.

On the other side, most of us did a whole lot better coming close to the numbers on the Republican side of the aisle. But this is one of those cases that remind us that pre-election polls are guides to voter attitudes and shifts. All things considered in this and other cases, we pollsters still do a creditable job.

Comments

[Pine]

The simplest explanation for this discrepancy in the statistics is this: the math is consistent and correct, the vote recording process was wrong. Fraud.

It will always be true that whoever controls the count, controls the election.

Without open source hardware, software and a mechanically persisted (paper) record, these results will be the rule rather than the exception.

[JBCinSD]

Maybe the polls were right on. Consider these facts:

Premier Election Solutions is the new name for Diebold - same company, same people, different name.

· Diebold is a radically pro-Republican company. Its CEO was on Bush's election committee.

· The Republicans want to run against Hillary, not Obama.

· Diebold made the optical scanners that counted the votes in the cities (votes were counted by hand in rural areas).

· Hillary was stronger in the cities, Obama in rural areas.

· Obama's vote count (not percentage) was almost exactly as expected and reported in exit polls.

· Hillary's vote count was about 10% (I think that's the number) higher than expected by everyone, including her campaign.

· All Diebold had to do was set the software to count every vote for Hillary 1.1 times and she would win. So easy.

Bingo

[Misanthrope0]

I find it laughable that all the pollsters and the Mainstream media is so quick to dismiss their own polls.
Never do you hear of election fraud.
This is not even in the mix.
Sounds a little funny if you ask me.
Never have the polls been this far off.
What a bunch of shills.

They cant spin this anymore.
Voter fraud in IA is surfacing now, also.

[priscianusjr]

You didn't mention the "Diebold Effect."

[avergejoe]

interesting and informative,
but you fail to explain how EVERY poll made the SAME mistake in the SAME direction.
and NONE mentioned an almost 20% error margin.

[Purple Girl]

Zogby
Find a new reason for your company- we're tired of hearing you tell US what WE think!

You are ususally wrong anyway. You might as well be taking bets in Vegas on the simulcast horse races.

[neworleanslady68]

1. She didn't cry. Her voice sort of caught for a moment.

2. I am amazed over and over again that men comment on the "voice catching" incident and the debate performance without mentioning the "iron my shirt" thing. Do you have any idea how sick most of us are of the sexism we women live with every day of our lives? It's ALMOST enough to make this female Obama supporter vote for Hillary. Why don't the men commenting remark on THIS incident? All the women I know are talking about it and were infuriated!

[The Hedonistic Pleasureseeker]

What happened?

What?

Like we don't know? Large discrepancies are prima facie evidence of vote tampering.

Smell the coffee yet?

[LeftLeaner]

You people in the MEDIA try to CREATE the OUTCOME, rather than follow it.

If every candidate had an equal voice, John Edwards would be running away with the delegates. As he represents the largest (ignored) interest group - The MIDDLE CLASS.

The only problem is that, they don't know it, because of your empy hero-worship for a rhetorical speaker, whose favorite word is "CHANGE".

No one asks WHAT it is he is going to CHANGE.

Wouldn't that be helpful before we choose our next so-called LEADER.

Remember, people voted for Bush because they'd like to have a beer with him.

I have greater demands from my candidate than that - that's why we need JOHN EDWARDS in '08.

[LeftLeaner]

You all talk up front like pompous experts, then you justify your errors, when things turn out differently.

Why can't you all just let the people vote, and NOT try to INFLUENCE and muddle their decisions by offering your so-called OBJECTICE, professional analyses.

I think we'd all be better off, if you just let the voters make up their own minds.

It certainly influences some people's phyches and decision-making process.

[MrMurder]

Nice of you to assume McCain is deemed independent.

You were just as wrong about that.

Others note you try to project identical undecideds to decided vote polling.

That's being obtuse.

Most of America is not in the shape New Hampshire is in terms of infrastructure, appropriated federal dollars brought in, per capita income, satisfaction with the way things are going, urban density, or employment trends.

Keep being wrong Zogby. It's why you joined CSpam to try and slant things rightwards...

[Mayme]

I think it would be a great exercise, to see just how well the optical scan machines work vs hand counted ballots. What a great place to look at this. Take the town of Manchester NH and count the ballots by hand. See if the scan machines really counted correctly.

It appears that the hand counted ballots in NH were more aligned with the polls, while the machine counted ballots did not follow the polls.

[marysandra]

"We seemed to have missed ......the older women"No surprise to this "older women" and there are quite a lot of us,actually. Since Hillary is an "older woman",may I ask how old are your pollster employees? Are they all under 30? I am an Obama supporter,mainly because he seems less like Bush than the others,but please don't discount us older women, most of us vote, lots of us have disposable income, and we even read newspapers and books,imagine that,even if we have to do it after working all day at some pretty good jobs. I admire Sen. Clinton, but I am not willing to see all the skeletons in the closets come out again, and am sick to death of the Bush-Clinton-Bush trio, now going for a quartet.Obama is thoughtful, gracious, and smart enough to pick good people to work with him. That's enough for me. Hillary needs to stay in the Senate, New York State needs her.

[samothrellim]

It would be helpful if Mr. Zogby could comment on the figures which show a wide disparity between hand count and Diebold count results in the Obama/Clinton race resulting in a 5+% bump for Hillary and a 3*% drop for Barack, as well as early exit polls showing Barack the winner. It's too early to cry fraud, but certainly these anomaly's need to be investigated.

[jnelsonleith]

The real problem with the "polls were wrong" assumption is that the polls weren't wrong. They were within a few points for every candidate in Iowa and within a few points for almost every candidate in NH.

When your clock is right 30 days in a row then is wildly wrong on Day 31, a rational person doesn't conclude that "clocks don't work!" The scientific approach is to look for what differed when measurements contradicted each other.

Obama was black in Iowa, so let's drop the race nonsense. Clinton was a woman in Iowa, that's not it either. And, I'm not prepared to blame the secret NH ballot, because that means that Clinton supporters are far more likely to lie (and, for racist reasons) than supporters of any other candidate. While that might be a fun idea to entertain, it's absurd at face considering the percentage points involved.

The polls were done the same for all the candidates, but we know from what happened in Sutton that votes were being counted differently for different candidates in different places. The difference, the the cause of the discrepancy, is in the vote count. It's time to stop all of these silly conflict-avoidant apologetics and get serious about our election system.

[SydGris]

I sadly share the suspicion of Diebold hackable voting machines. I'm frustrated no major media outlets are even discussing it. Have no journalists done their home work on this issue? Even seen Hacking Democracy? 40% of New Hampshire's precincts are hand-counted, which equals about 25% of the votes. All the rest are counted on hackable Diebold op-scan systems, with completely hackable memory cards, all programmed and managed by LHS Associates. As Bev Harris of BlackBoxVoting.org who seems to share my concern, says, LHS is the "chain of custody" in New Hampshire elections.

[rjcrane]

What hasn't been explained yet to me is why the candidates' own polling in both camps showed Obama with a double digit lead on election day and why the exit polling results given to Chris Matthews at 5:00 pm showed Obama with a 5 point lead. Why and when did this lead switch or was it wrong from the beginning because of poor exit polling? I think the biggest factor in Obama's loss was that too many people took his victory as a given and either didn't vote or voted in the GOP primary where they at least felt they might make a difference either to help Ron Paul or McCain.

I also feel a lot of woman who might even have not liked Hillary voted for her anyway just because they felt the media and other candidates were beating up on her too much. In fact this explanation if true becomes the worst possible reason for anyone to vote for someone to elect the next President of the U.S. In other words, more women voted for Hillary not because they felt she would make the best choice but because they felt sorry for her.

RJ Crane
topplebush.com

[Zhiroc]

I think I'll post a number of my points here, rather than keep replying, which is harder to see.

1) The exit poll results match the vote results within 1%. This does not support the notion of machine fraud, unless you also postulate that the exit polls were altered.

2) Obama received, within the margin of error, the level of support that the tracking polls predicted (35-41% in the ones that Olbermann cited last night on Countdown). This tends to refute the argument for a Bradley effect (that people lied about who they voted for)

Thus, my conclusion is that for whatever reason (the debate, Monday's emotional display, Monday's Iran incident, race, the phase of the moon, etc...) the undecideds broke about completely for Clinton and/or the voting demographics did not match the data model.

[wrabbitt]

i don't need polls to tell me who to vote for, HRC is a Washington insider,always has been, always will be. the only change she will lead us into is in name only,she will be a "W" in sheeps clothing we can't elect someone because of her sex, or lack of, my vote is against Hillary, so poll that!

[runnyknose]

The polls were correct!!!

After analyzing the NH polling details at

http://ronrox.com/paulstats.php?party=DEMOCRATS

I am convinced that Diebold got hacked again.

The Republicans don't want to face Obama in November so they did the same thing they did in 2004 - fudge the actual results to get Clinton nominated.

[FrankN]

Saw you on Jon Stewart's Daily Show. As he said it would have been better interview if you'd put up more of a defense, by citing this article for one thing.

[Wilburrr]

I don't see the polls as being wrong. Zogby and others showed the race had closed to within 2 points the day before the primary. The polls had a 5% margin for error and there was a 4% shift by the end of the voting. What was "wrong" with that? Data is never wrong; it is our interpretation of data may be confused.

[buddyd]

The only the polls missed was the 40% undecided. The were so wrap up in defeating Hillary. It gave me great pleasuse in watching the media squrim. Chris Matthews and Tim Russert both looked like they were having a serious BOWEL Movement when the results started going Hillary's way..Even the Huffington Post was enjoying Hillarys (DEFEAT)Shame on all of you. As and African American Male I have been, and always will be a Clinton supporter..HA! HA! HA!

[rhettab]

will anyone dig into the deibold machines vs. paper ballots. have heard that the paper ballots were in line with the polls and the machines were not hmmmm.....do the clintons have clout in NH?

[live by the golden rule]

Dear John Zogby-

You say:
I think she also scored points during the ABC debate Saturday night when she declared, amid a discussion about the country's desire for a change in direction, that electing a woman would represent a big change in itself.

How about asking in your future polls whether people like candidates who can give detailed fact-filled, thoughtful responses to issues.
Such detailed answers unfortunately completely bore the reporters who cover the campaigns, so we never read the substance of these fact-filled responses in news coverage. However, voters do get to see detailed answers, from those candidates who can give them, at Town Meetings, and, fortunately, we all got to see it in the Saturday debate, where the format allowed more than short sound bites.
Under this particular debate's rules, where detailed analysis was allowed without buzzers, Senator Clinton's intelligence and hard work came through. So perhaps that's when she scored points, especially among hard thoughtful working women.