Wednesday, January 09, 2008

Simple as ABC

Gary Langer at ABC blogs struggles with trying to explain Hillary Clinton's victory -- in defiance of the polling results -- in New Hampshire. Langer's approach is to examine every possible technical reason for the election results without resorting to conspiracy theories.

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.)

One academic has an interesting hypothesis, the slight bias created by the alphabetical listing on the ballot ("Clinton" comes before "Obama") may have been enough to nudge Hillary upward in a close contest.

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.

Without advancing any conspiracy theories of my own, the question that springs to mind is why we should eliminate such theories from the reckoning. What kind of evidence would cause an investigator looking into statistical shocker to discount the possibility of fraud? Mere improbability would not be enough. Improbable events happen without the intervention of conspiracies. For example, an Czechoslovakian airline stewardess survived a free-fall from 33,000 feet after her aircraft was blown up by a terrorist bomb in mid-air. No one suspects that someone conspired to make survive.

Conspiracies, it seems to me, should only be considered when a sequence of suspicious events is established to have occurred. Since Barack Obama has conceded New Hampshire, it is unlikely that anyone with standing will come forward to allege fraud based on the existence of suspicious events. Thus conspiracy theories are eliminated from the reckoning at the outset. And we have left is wondering whether the order of the names on the ballot mattered.


Blogger geoffb said...

two things that come to mind that may have influenced the results.

I heard the weather was exceptionally warm for January much as this week has been for me in Michigan.

I also heard several weeks ago that someone from out-of-state could come to New Hampshire and by stating their intention to move there be able to vote in the primary. If true, and it seemed to be in the report I heard, then what about all those people bussed in for the Clinton rallies? Could they have voted?

1/09/2008 08:11:00 AM  
Blogger David M said...

The Thunder Run has linked to this post in the - Web Reconnaissance for 01/09/2008 A short recon of what’s out there that might draw your attention, updated throughout the check back often.

1/09/2008 08:14:00 AM  
Blogger RonF said...

Me, I like to lie to pollsters. I bet other people do, too.

1/09/2008 09:36:00 AM  
Blogger Peter said...

You mention Clinton and poll in the same paragraph and my stomach turns, bile rises to my throat, and an overwhelming sense of dread fills my tightening brain.

1/09/2008 09:51:00 AM  
Blogger Nomenklatura said...

A much more entertaining hypothesis is the reverse Bradley effect described by Mickey Kaus.

The 'Bradley effect' describes how white voters may tell pollsters they intend to vote for a black candidate in order to put their lack of racial prejudice on display, but then vote differently in the privacy of the voting booth, leading to polling upsets exactly like this one.

Kaus noted the idea that "Iowa's public caucuses led Dem voters to demonstrate their lack of prejudice by caucusing for Obama". Thus in Iowa you had a rare opportunity to show off your lack of prejudice in front of your neighbors, and even look down on the ones who did not join in supporting Obama. I can certainly believe that many left-leaning Iowan Democrats may have found this irresistible.

In New Hampshire by contrast the primary is a secret ballot, eliminating the opportunity for a public preening display and producing the observed rebound towards Hillary.

1/09/2008 10:06:00 AM  
Blogger Peter said...

This comment has been removed by the author.

1/09/2008 10:17:00 AM  
Blogger RonF said...

It was Mike Royko, Chicago Daily News/Sun Times/Tribune columnist who advised people to lie to pollsters, but I while I find plenty of people who reference this, I can't find the actual column itself. I do remember reading it, though, when it came out. Reading Royko's column every day was worth the price of the paper all by itself.

1/09/2008 11:09:00 AM  
Blogger El Jefe Maximo said...

It was clearly the Right Wing Republican Attack Machine. . .in there dimpling chads, butterflying ballots and putting Clinton's name first. Then they went and bussed in (flew, actually, on black helicopters) the Florida Supreme Court to declare their intention to move there permanently. . .

1/09/2008 11:21:00 AM  
Blogger LarryD said...

The problem with the "Bradley Effect" theory is that Obama got the votes the polls predicted, it was Hillarys votes that weren't in line with the polls.

1/09/2008 11:46:00 AM  
Blogger whiskey_199 said...

Hillary won among lower income voters. Hillary won among women. Hillary won among state, local, and federal government workers who turned out in droves to support her. Hillary won among unionized employees (which turns out to be mostly government workers).

Hillary also won among older voters. Many of Obama's crowds were college and high school kids -- people working jobs could not take off from work but did vote in record numbers.

Hillary also benefited from late deciders who were turned off by lack of specifics by Obama in his speeches and like specifics by Hillary.

1/09/2008 12:43:00 PM  
Blogger Shropshirelad said...

I am with ronf, I always lie to pollsters, too, as a matter of policy.

Even allowing for a few saboteurs like us, it is hard to pinpoint why the predictions were so wrong. There are too many factors involved. Even something as simple as the placement of names in alphabetical order might plausibly affect outcomes.

Still, I think there might be some truth to nomenklatura's idea that the difference between Iowa's results and New Hampshire's could have something to do with the curtains the voters closed to cast their ballots.

I don't mean to suggest anything unsavory about anyone, but:

A poll's a fine and private place/
And none, I think, do there embrace.

1/09/2008 01:14:00 PM  
Blogger The Fitness Diva said...

People are really just sick and tired, and want something better and different, dudes.
And they're ready to elect a woman or a black man to get it.
That's it in a nutshell.

1/09/2008 01:51:00 PM  
Blogger Kate said...

I'm sure there are several reasons why the polls were wrong. In the end I believe it was Hillary's passion that gave her those critical votes. Some people criticized her for tearing up, but really that helped her gain some last minute support. I gave her a high rating for passion on even though I don't want her as the candidate.

1/09/2008 01:53:00 PM  
Blogger always right said...

Re: Whiskey_199's analysis

Look at the breakdown demography of Hillary supporters, that's the solid block traditionally voted dem anyway.

She can NOT claim that she's a CHANGE candidate. She's the "same old, same old" 40-year-old model (not her age) apparently still selling strong!

The hopeful sign (maybe?) is that younger generations gotten wiser to the old shticks, and wanted real change in Obama. Even though we have no idea what change Obama is talking about.

1/09/2008 02:56:00 PM  
Blogger musemonk said...

former prime minister of canada, john diefenbaker: "dogs know what to do with polls!"

1/10/2008 12:45:00 AM  
Blogger Doug said...

What geoffb said...
Parking attendents reported buses and out of state cars, but of course Bill and Hill would never resort to skulldugery, no would they?
Such a high standard:
"I intend to move to NH."

Guess it beats letting illegals vote, but the outcome is the same.

1/10/2008 10:16:00 PM  

Post a Comment

Links to this post:

Create a Link

<< Home

Powered by Blogger