New Hampshire’s Polling Fiasco

There will be a serious, critical look at the final pre-election polls in the Democratic presidential primary in New Hampshire; that is essential. It is simply unprecedented for so many polls to have been so wrong. We need to know why.

But we need to know it through careful, empirically based analysis. There will be a lot of claims about what happened - about respondents who reputedly lied, about alleged difficulties polling in biracial contests. That may be so. It also may be a smokescreen - a convenient foil for pollsters who’d rather fault their respondents than own up to other possibilities - such as their own failings in sampling and likely voter modeling.

There have been previous races that misstated support for black candidates in biracial races. But most of those were long ago, and there have been plenty of polls in biracial races that were accurate. (For more on past problems with polls in biracial races, see this blog I wrote for Freakonomics last May.) And there was no overstatement of Obama in Iowa polls.

On the other hand, the pre-election polls in the New Hampshire Republican race were accurate. The problem was isolated to the Democratic side - where, it should be noted, we have not just one groundbreaking candidate in Barack Obama, but also another, in Hillary Clinton.

A starting point for this analysis will be to look at every significant Democratic subgroup in the New Hampshire pre-election polls, and see how those polls did in estimating the size of those groups and their vote choices. The polls’ estimates of turnout overall will be relevant as well.

In the end there may be no smoking gun. Those polls may have been accurate, but done in by a superior get-out-the-vote effort, or by very late deciders whose motivations may or may not ever be known. They may have been inaccurate because of bad modeling, compromised sampling, or simply an overabundance of enthusiasm for Obama on the heels of his Iowa victory that led his would-be supporters to overstate their propensity to turn out. (A function, perhaps, of youth.)

Prof. Jon Krosnick of Stanford University has another argument: That the order of names on the New Hampshire ballot - in which, by random draw, Clinton was toward the top, Obama at the bottom - netted her about 3 percentage points more than she’d have gotten otherwise. That’s not enough to explain the gap in some of the polls, which presumably randomized candidate names, but it might hold part of the answer.

The data may tell us; it may not. What’s beyond question is that it is incumbent on us - and particularly on the producers of the New Hampshire pre-election polls - to look at the data, and to look closely, and to do it without prejudging.

– By Gary Langer

Popularity: 1% [?]

The main discrepancy was in the Obama Clinton contest.
Those Diebold op-scan machines are the exact same ones that were hacked in the HBO documentary, Hacking Democracy.

Diebold favors Hillary - hand count for Obama.
2008 New Hampshire State Primary Results
A Closer Look At The Count
TOTAL VOTES FOR DEMOCRATS: 286,139
All averages and counts are calculated when the page loads, using vote data sourced from: www.politico.com
Table Comparing Machine vs Hand Counts
Candidate Total Votes Avg. Overall Votes
by Machine Avg. Overall
by Machine Votes
by Hand Avg. Overall
by Hand Machine VS Hand Votes
by Unknown** Avg. Overall
by Unknown
Clinton 111,827 39.081% 82,108 39.618% 20,190 34.908% 4.709% (13,475 votes*) 9,529 45.266%
Edwards 48,386 16.910% 34,927 16.853% 10,170 17.584% -0.731% (-2,093 votes*) 3,289 15.624%
Gravel 397 0.139% 273 0.132% 80 0.138% -0.007% (-19 votes*) 44 0.209%
Kucinich 3,858 1.348% 2,609 1.259% 1,055 1.824% -0.565% (-1,617 votes*) 194 0.922%
Obama 104,030 36.356% 75,251 36.309% 22,335 38.617% -2.308% (-6,604 votes*) 6,444 30.611%
Richardson 13,154 4.597% 8,975 4.330% 3,209 5.548% -1.218% (-3,485 votes*) 970 4.608%
Other 4,487 1.568% 3,108 1.500% 798 1.380% 0.120% (343 votes*) 581 2.760%
TOTALS: 286,139 207,251 57,837 21,051

*Votes are tentatively won or lost with the assumption that the machines are conferring advantages or disadvantages.
**Unknown towns (where the data doesn’t specify counting method) include: Harts Location, Manchester, Waterville, Wentworth’s Location

Table Comparing Votes vs Town Size
Candidate Votes
Small Towns Avg. Overall
Small Towns Votes
Medium Towns Avg. Overall
Medium Towns Votes
Large Towns Avg. Overall
Large Towns
Clinton 7,103 34.259% 30,444 37.569% 74,280 40.288%
Edwards 3,664 17.672% 14,518 17.916% 30,204 16.382%
Gravel 29 0.140% 112 0.138% 256 0.139%
Kucinich 358 1.727% 1,184 1.461% 2,316 1.256%
Obama 8,055 38.851% 29,664 36.607% 66,311 35.966%
Richardson 1,206 5.817% 4,049 4.997% 7,899 4.284%
Other 318 1.534% 1,063 1.312% 3,106 1.685%
TOTALS: 20,733 81,034 184,372
Small Town has less than 500 votes.
Medium Town has between 500 and 1,500 votes
Large Town has more than 1,500 votes

BlackBoxVoting.org results were are astonishing:
Results tallied for 209 out of 236 of the municipalities.

By Percentage
Method Hillary Clinton Barack Obama
Diebold Machines 53.23% 46.77%
Hand Count 47.47% 52.53%

By Votes
Method Hillary Clinton Barack Obama
Diebold Machines 82860 72807
Hand Count 18898 20912

By Number of Municipalities Won
Method Hillary Clinton Barack Obama
Diebold Machines 54 33
Hand Count 43 77

About 81% of the votes will be “counted” by the Diebold machines.

Ann Catherine and anyone else “harping” that “People are not obligated to share their voting intentions with anyone” need to realize that many Americans believe that the Florida 2000 & Ohio NO ONE who voted for Bush but told a pollster they were voting for Gore. I have however, seen plenty of documentation that that election was stolen. Here’s part of the solution; elimination of the electoral college and elimination of Diebold-type voting machines.

Objective and independent-minded reporters have a Costitutional obligation. This is to cover “news” Their legal obligation is: not to brainwash victims into postmodernist alternative attitudes, opinions and beliefs, so they’ll think they’ve heard news but really haven’t. Only Reality matters in this context; ad anything that interferes with the apperception of what is real by the individual voter warps, poisons and destroys the process of informed and rational choice which is an “elction”. Charisma does not matter. Innuendo can’t matter. We’re not electing a false claim nor a hairdo, a smile nor a suit. Lies, fake-promise slogans and lying promises can’t count. False polls and horse-race- type speculations, endorsements, context dropping analyses and public news tsars’ media hype and misleading headlines can’t count. What has tyo matter is SCIENCE–what the candidates know, define, say their prioritized agenda will be, and their statement of how they expect to accomplish those ends and how honestly they present what they are and what they stand for to the citizens (who will be the future beneficiaries or victims of their course of real actions) count. Until we REGULATE the formin which non-fictoinal statements have to be presented, and separate opinions from attested fact, from standards-based evaluations and from categorizing definitions and full proofs–until that happens, purveyors of lies and myths and attitudes will continue to be given access to the scientific level that only truth tellers have the right to access. Until we do this hard job of policing our would-be tyrants and of making our elections something other than a poll-driven exercise in fakery (abetted and further warped by silly-minded by incompetent newstwisters), we cannot never have a real election. We’ll see New Hampshire’s three-day fiasco on a broader empire-wide scale, in this former nation of regulations and rights. This appalling exercise in horse-race, insider guesswork, misrepresentation, character assasssination, swiftboating lies and futility has to be cleaned up; I say it is an Augean Stables of crime, lies and nonsense. News is what candidates do, say, and premise, their history, their character and their campaign appearances. Newsmen are supposed to cover the election, not their own biases; polls interfere with that process. I say they have to be done away with entirely. if one person is influenced to vote differently because of any true or false “poll”, rather than the prior voting of a substantive sort ad the facts presented and evidenced, then something has been introduced into the campaign which destroys its purpose–one man, one vote. Reporters are to cover what candidates do and say and have done and have enacted and have promised and delivered–anything else they might do is fraud and totalitarian fakery carried out at the expense of truth as understood by one mind at a time–by each voter. One mind thinks, evaluates, votes and suffers good or bad results from such a choice. One suggestion—in any primary vote, let’s have the voter rank the candidates in order of acceptability. 5, 3 and 1 for the top three finishers. Vote for your party, yes. Only this way can we really have a meaningful election and not some brainwashed choice of our next public-interest totalitarian fraud anointed by newshacks, media tsars’ lies and the falsity of “leading candidate” centrality where there is no such thing, one individuals presenting themselves as candidates.