Pollster.com: Looking for New Hampshire Exit Polls?

January 08, 2008

Looking for leaked exit poll results from New Hampshire? Sorry to disappoint, but whatever their merits, we are unlikely to see any such leaked results until moments before the polls close.

In past years, the network consortium that conducts the exit polls distributed mid-day estimates and tabulations to hundreds of journalists that would inevitably leak. In 2006, however, the networks adopted a new policy that restricted access to a small number of analysts in a "quarantine room" for most of the day and did not release the results to the networks and subscriber news organizations until just before the polls closed (information that did ultimately leak to blogs). As far as I know, that process will remain in place today.

Here are a few tips for making sense of the exit poll data that you do see tonight:

1) An exit poll is just a survey. Like other surveys, it is subject to random sampling error and, as those who follow exit polls now understand, occasional problems with non-response bias. In New Hampshire (in 1992) and Arizona (in 1996)* primary election exit polls overstated support for Patrick Buchanan, probably because his more enthusiastic supporters were more willing to be interviewed (and for those tempted to hit he comment button, yes, I know that some believe those past errors suggest massive vote fraud -- I have written about that subject at great length).

2) The networks rarely "call" an election on exit poll results alone. The decision desk analysts require a very high degree of statistical confidence (at least 99.5%) before they will consider calling a winner (the ordinary "margin of error" on pre-election polls typically uses a 95% confidence level). They will also wait for actual results if the exit poll is very different from pre-election poll trends. So a single-digit margin on an exit poll is almost never sufficient to say that a particular candidate will win.

3) Watch out for "The Prior." At least two networks are likely to post exit poll tabulations shortly after the polls close that will update as the election night wears on (try these links for MSNBC and CNN). Those data are weighted to whatever estimate of the outcome the analysts have greatest confidence in at any moment. By the end of the night, the tabulations will be weighted to the official count. Typically, the exit poll tabulations are weighted to something called the "Composite Estimate," a combination of the exit poll data alone and a "Prior Estimate" that is based largely on pre-election poll results. So if you look to extrapolate from the initial tabulations posted on MSNBC or CNN (as we did here on Election Night 2006), just keep in mind that in the estimate of each candidate's standing in the initial reports will likely mix exit poll and the pre-election poll estimates (not unlike the kind we report here).

Finally, if you would like more information on how exit polls are conducted, you may want to revisit a Mystery Pollster classic: Exit Polls - What You Should Know. Happy New Hampshire Primary Day!

*Clarification: The original version of this post implied that the 1996 overstatement occurred in New Hampshire.

Note: An apology to those who continue to receive the "too many comments" error message when attempting to post a comment here. Suffice it to say, you have not posted too many comments to Pollster.com. We have been trying to squash this bug for months now (without success) and, unfortunately, the very heavy traffic we are experiencing today aggravates the bug.

-- Mark Blumenthal

January 08, 2008 in Exit Polls, New Hampshire 2008, The 2008 Race

Comments

FlyOnTheWall:

Thanks, Mark, for a well-written post.
I was struck by something this morning, looking at the final two days of tracking polls, and was hoping that you could illuminate the issue.

There have been 16 numbers released over the past two days, all conducted entirely since Iowa. Of these polls, 14 have pegged Hillary's support in a very narrow range of 28-31 percent. (The other two are from Suffolk, which has been a consistent outlier throughout primary season, but even Suffolk is only at 34.)

The Obama polls, by contrast, are all over the map. They put him anywhere from 32 to 42 percent, and are fairly evenly distributed over that range. In other words, everyone seems to agree on Hillary's level of support - but what determines the margin is the level of support for Obama.

What gives here? My instinct is to assume that we're looking at some sort of house or modeling effect. That is, pollsters may be using a wide range of estimates for the numbers of independents who will show up, the numbers of first time voters, or the numbers of young voters - all groups that heavily favor Obama and have become tremendously unpredictable in these past few days.

Your thoughts?

Posted on January 8, 2008 10:23 AM

____________________

Elvo Roveli:

I belive what you maybe experiencing also is the amount of people who do not want to be polled or contact. I made a few calls last night for my candiate of choice and several callers would answer, "I know who I am voting for and I am not telling you" or "I voting for John F Kennedy".
Everyone wants to know. But I think the residents of early voting are driven crazy with all the calls and polls.

Posted on January 8, 2008 11:11 AM

____________________

Mark Lindeman:

By the way, as long as we're talking about New Hampshire exits, we might mention that the general election exit polls had double-digit discrepancies (using average Within Precinct Error) in 1992, 1996, and 2004 -- pretty big also in 1988. Of the five elections for which E/M provided data, only 2000 looks sort-of-maybe OK.

FOTW (someone probably already said this), seems like the numbers point to "undecideds" breaking to Obama or, perhaps better, choosing between Obama and undecided depending on how hard they are pushed. The proportions of independents could also enter into that, for sure (although it isn't obvious why an indy influx alone would affect Obama's numbers and not Clinton's). But I haven't stared hard at all the numbers together.

Posted on January 8, 2008 11:12 AM

____________________

FlyOnTheWall:

Elvo,

That may be, but it only explains the discrepancy if Obama voters turn out to be less likely to talk than Hillary voters - and if they're responding at different rates to different pollsters. I still think we're looking at sampling or modeling problems.

Posted on January 8, 2008 11:20 AM

____________________

s.b.:

It is true, not only is Hillary's support rock solid it's been climing in the past two days. Obama's is subject to greater swings as his support has less consistency and history in NH. This makes sense, as his bounce was from Iowa, fair enough. However will NH voters allow Iowa to dictate how they vote? Since Saturday, Obama's numbers have gone down consistently, except the Suffolk poll. Why any polsters are still including Saturday is a mystery. Rasmussen's numbers are very interesting. If you look at their rolling averages and do the math, Obama was only up by 1% last night in their polling, rolling average of 7% up but that still includes Sat when he was up by 14%. Given this, and the fact that 25% of voters, who say they are likely to vote, are still undecided means it could break either way. Realisitically, it looks like it will break Obama, but only if his youth and independents and Repubs come out to support him. Many may vote for McCain thinking Obama is winning anyways. I think it will be much closer than predicted Obama 38%, Clinton 35%, but there is still a slim possibility Clinton can win. As Cokie Roberts said, he has to sustain record numbers of new voters, independents, and republicans in every primary or Caucus, not just Iowa. Whether or not that is possible is yet to be seen.

Posted on January 8, 2008 11:37 AM

____________________

Michael McDonald:

Also be on the lookout for another feature of New Hampshire's election law that can slightly skew results. The law permits independents to select which party's primary to vote in. When an independent does so, they change the party affilation on their voter registration to that party. If they do not later change back to independent, they lose the ability to choose which primary to vote it.

Past New Hampshire primary results, report thousands of write-in votes in a party's primary for candidates of the other party. Presumably, these are cast by people who showed up to the polls and discovered that they could not choose which party's primary to vote in, so they did the next best thing and cast a write-in vote for their favorite candidate.

The results of these write-in votes are often reported days after the election, long after media attenion moves on to the next state. The exit polls cannot take these write-in votes into account (at least until they are reported). Fortuntately, there are relatively few of these write-in votes relative to the other votes and the percentages are somewhat similar to the within-party primary results, so the bias this induces is small. Still, if the election is particularly close, the rank ordering of candidates could be affected by this quirk in New Hampshire election law.

Posted on January 8, 2008 11:39 AM

____________________

Joe Helfrich:

I suspect that the polls are over--particularly in terms of Obama's support--because pollsters are making adaptations to their turnout model in response to the unprecedented increase in Iowa.

The trouble is, I don't know if anyone knows if what happened in Iowa is likely to happen in New Hampshire. Obama spent a lot of time and money making that happen for the caucus. Anyone know if he's done the same thing for today's event?

Posted on January 8, 2008 12:49 PM

____________________

GH:

Answer to Joe Helfrich... I'm not sure how much Obama invested in increasing turnout in New Hampshire, but as you've probably seen by now on numerous websites, the turnout appears to be explosively high. He may have excited the independent/youth vote so much by his Iowa victory that he doesn't even need to actively have his campaign do it.

Posted on January 8, 2008 1:16 PM

____________________

jakbqwik:

It's beginning to look like Ron Paul might steal New Hampshire from McCain by a very small margin. Too funny!

Posted on January 8, 2008 1:37 PM

____________________

brattleboro:

Not stated here: are substantial exit polls even being conducted? Not that I've heard/

Posted on January 8, 2008 1:55 PM

____________________

larry :

Alot of comments coming at polls indicate that voters were swayed by Romney's performance in Fox's Sunday debate. Last minute Romney surge?

Posted on January 8, 2008 2:07 PM

____________________

Darren:

It seems like Ron Paul has no chance at the nomination, ever. Shocking!

Posted on January 8, 2008 2:16 PM

____________________

ProblemWithCaring:

Thank God for the death of rampant exit polling and the Voter News Service.

They and Murdoch had even NPR calling the race for George Bush, a full 2 score before the Supreme Court gave it to him....

Posted on January 8, 2008 2:18 PM

____________________

Mark Lindeman:

ProblemWithCaring, not so much.

Dunno what "rampant exit polling" you refer to, but the exit polls are on.

The exit poll data contributed to the first bad call for Gore, although even that was buttressed by vote count data. (Oops.) The later call had nothing to do with exit polls -- all that data would have been supplanted in the projection model long since. It was influenced by the bad vote count numbers from Volusia County (and, I think, another county).

Posted on January 8, 2008 2:33 PM

____________________

Mark Blumenthal:

Note: I posted thoughts on the comments from "FlyOnTheWall" here.

Posted on January 8, 2008 2:47 PM

____________________

alisamleo:

Collective Intellect has some interesting reporting with intelligence from the blogosphere:

http://www.collectiveintellect.com/2008/01/running_up_to_the_iowa.php

Posted on January 8, 2008 2:58 PM

____________________

markg8:

The Manchester Union Leader editor? publisher? was just on MSNBC and looked glum about the call for extra ballots being all Dem. That bodes well for Romney getting more votes than McCain though he'll still get half of what Dems get.

Posted on January 8, 2008 3:32 PM

____________________

JT:

s.b. wrote:

"Since Saturday, Obama's numbers have gone down consistently, except the Suffolk poll."

This is incorrect. The Zogby poll:

1/3 - 1/5 Clinton +1
1/4 - 1/6 Obama +10
1/5 - 1/7 Obama +13

Posted on January 8, 2008 3:52 PM

____________________

DCR:

I think his point was the fifth (Saturday) is still being included in tracking polls released on the 7th.

And that the numbers for the individual days show drops, not the rolling average.

Don't know if this is true on Zogby. But Zogby is still including Saturday numbers in the latest poll.

Posted on January 8, 2008 5:37 PM

____________________

Aakash:

It is very likely that voter irregularities and vote fraud caused Pat Buchanan to lose the Arizona Primary (and thusly, the GOP nomination) in 1996... I am glad that this issue is still being addressed today, however one feels about it.

I am looking for the New Hampshire exit polls online... Is because of new regulations, that networks and other sources do not report exit poll data, on the day of the primary?

Thank you very much for this entry!

Posted on January 8, 2008 7:14 PM

____________________

s.b.:

YEs JT and Zogby was dead wrong. I couldn't tell with zogby because of the way the days were rolled. Was he lower on the 4th or the 7th? You can't tell from how the poll stands even in your note. Each days tellies were on their site. Who they were polling is anyone's guess, because for a reputable firm they sure do look bad now. The trend was down from Sat.

Exit polls now show Clinton won by 40% to 36%. So the downwar dtrend shown by rasmussen was more correct eventhough chosing to include Saturday and not just go with their Sunday poll of Obama 1% up was wrong in this bounce situation, with a serious downward trend.

They made an error. They wanted obama to win and did not go with the better data from Sunday but included the bounce day.

Their loss as a polling firm in reputation, but their daily trend up 14, up 6, up 1, leads you exactly to down by 4% today for Obama.

Posted on January 8, 2008 9:56 PM

____________________

s.b.:

Wow that was a really incoherent post.

I didn't find the Zogby individual daily numbers until later on their site. Yes they showed an upward trend for Obama when everyone else showed a downward trend. That's bad for a polster. Even if your numbers are wrong, you want to show you had the right trend.

I suspect they adjusted their sampling to reflect voter demographics in Iowa and a larger youth and indie turnout for Obama, again a poor choice for a polster.

Posted on January 8, 2008 9:59 PM

____________________

abiggerdog:

Dr. Steven Freeman explains why the 2004 exit poll indicated massive vote fraud in favor of Bush during that election:
http://www.appliedresearch.us/sf/Documents/ExitPoll.pdf

"Pollster" above wants us to believe it's due to sampling error or flaws in the poll taking process. Just keep saying that if you want to keep your job -- and you want to stay alive. The zionists who control this country have ways of getting rid of people who speak up.

Posted on January 9, 2008 6:52 AM

____________________

john monto:

Exit poles where used to check accuracy of elections. They have less than one percent margin of error. The fact that there was a twenty point difference from exit poles to actual vote count is more than a little funny. Of course, if no one makes a stink about it when Hillary wins, no one will be able to make a stink about it when Macane
wins in Novembrer. We are asleep at the wheel and moving fast.

Posted on January 9, 2008 8:05 PM

____________________

starchild:

Now that the New Hampshire primary is over, there are lingering questions about the integrity of the vote. Dennis Kucinich challenged the results. It would be nice to see the media doing its part by taking this seriously, and for a start, publishing the full exit polling results, and if they don't match the official outcome, conducting thorough investigations into why that might be. There are rumors that Obama actually won New Hampshire on the Democrat side, and that Ron Paul did significantly better than reported on the Republican side.

Posted on January 25, 2008 5:19 PM

____________________

starchild:

Posted on January 25, 2008 5:19 PM

____________________