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

[DRaymond]

You can to a lot of post mortem about why the pollsters were all wrong but that does not change the fact that you were wrong. The margin of error in your polls is underemphasized, underreported, and most likely underestimated.

The margin of errors reported in polls are statistical abstracts based on sample size and presume that your really are getting a good sample. It presumes that the people who responded to your poll are exactly like those who did not respond when there is clearly at least one difference, the second group does not like talking to pollsters!

[byrdland49]

"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."

So...the act of releasing the polls influenced the outcome of the election. Remind me again, Mr. Zogby, what useful purpose these polls have? Other than allowing talking heads in the media to claim they have some degree of magical ability in predicting the future...

[LaBohemian]

Or, perhaps, Mr. Zogby, this will be the election that relegates your disservice to democracy to some quaint dustbin. We will decide, not pollsters or pundits. In the era of cell phones and caller ID, what scientifically credible segment of the population answers the phone, especially during dinner hour?

[buckbuck11]

I believe that it wasn't so much that the "tering up" incident made Hillary more human, rather that the reaction to it in the MSM (particularly nasty on Fox) and in the blogosphere (on both sides)caused a strong reaction, particulary among women, against the blatant unfairness shown to Mrs. Clinton. Even though I'm voting for Edwards in the NY primary, I thought the press was working way to hard to vilify Sen. Clinton.

[souris]

.......no big mystery, NH is still using the hackable DIEBOLD system, and the "horse race" that fuels the money behind our politics [and the corporate media who rake in mega $$$$'s] needs for there to be the excitement of the back-and-f­orth......­..also, the Repukes want Hillary to run against...­..........­move along, nothing new here to see...............

[MAX1]

.

So, only Obama supporters dodged to Clinton camp? Please explain how polling for the Repub's and the rest of the Dem's were on target.

ANYONE?

.

[BWonka]

About 6-8-10 months ago, everyone said it would be Clinton and McCain. It was. It is.

What's the surprise?

[Tremonius]

Another problem besides counting in polling is the assigning of purpose and plan to the results without any establishing data. We are told the voters "want the candidate to fight for the nomination" or "don't want a coronation." I doubt seriously the electorate has any such abstract general plan. They wake up to a new day and are blown by such monumental winds as, did Hillary cry? Did she cackle? Are those cable creeps piling on?

When you have the numbers right, then maybe we'll read your background summary.

[Robert11010]

Get real ... polls are never wrong to this extent - have we learned nothing from Ohio in 2004? The neocons have their grubbly little fingers firmly entrenched in the voting machines that were used in New Hampshire, and they want Hillary to win the democratic nomination, so that when they steal the election in November, they can be believed when they bring out the "most people hate her" mantra ...

[lovethesinner]

"All things considered in this and other cases, we pollsters still do a creditable job."

Perhaps "creditable," but credible?

Polling is dead. It's no longer science.

Everyone knows they're being manipulated, and cell phones have changed the math. You can't factor data you don't have, or keep up with a dynamic opinion that is turbocharged by the internet. The fact that you were all wrong in the same way proves this.

But, now is not the time for polls, anyway.

We are leaving the era of politics and moving to the era of music.

As the buzzzzzz of statistics grows softer, we hear the tune inside our hearts, singing the familiar verse and chorus...

Can we do this?

Yes, we can.

[JonBFertippton]

Was this primary vote really 81% a Diebold affair? Maybe the advance polling was not so wrong. I would like to see what the exit polls had to say - before they were adjusted to correspond with the final (Diebold?) result. Hillary is obviously the GOPs, and therefore Diebold's, desired candidate. The only thing left to wonder is which Republican Diebold favors for the final formalities.

[Chavez08]

What are "Super Delegates"???

[CyKhaan]

Please grow up and say "Bradley Effect." You can look it up. Go on, say it. And stop being so white. You might be able to see something for a change other than your own image.

[Mormondude]

If your premise about many independent voters leaving Obama to flock to McCain is correct, then why are you touting your accurate predictions on the Republican side?

It may be true, but without that flow of independent voters, you would have messed up the Republican predictions.

Personally, I think that when pollsters have enough power to actually CHANGE THE OUTCOME of elections in this way, they need to dial it back a whole lot. This is getting totally out of control.

Without all the polling hoopla, this Clinton win would have been nothing special. McCain's win would have been nothing special. Now suddenly they're the most spectacular political feats in the history of democracy.

I think given the situation, the onus is on the pollsters to explain exactly what they ADD to the election, and how they justify their newly assumed role in this process.

[docbradd]

The excuse that you "had" to lag the data and include earlier polls, while ignoring a trend is lame. Why not report both, the 3 day lagged data and the trend (both appropriately labeled)? In something as fluid as voting, I would think that trends are as important as single data points taken from a three day average.
By the way, why don't you label all your polls as being aggregates over three (or whatever) days? That seems like important information.