2006 November 16 Thursday
Silent Video Clip Viewing Best Way To Choose Winning Political Candidates

People who spend lots of time parsing sentences from political candidates are far less able to choose winners than those who pursue a shallower and quicker approach.

People watching short clips of silent debate footage are able to predict political election winners more accurately than predictions based on reports of economic conditions, finds a study supported by Dartmouth, the University of Chicago, Harvard's Kennedy School of Government, and the University of Michigan's Institute for Social Research (ISR).

"We found that snap decisions based on charisma are a good predictor of election outcomes," says Daniel J. Benjamin, an assistant professor of economics at Dartmouth and a fellow at ISR. "But you need to measure charisma with silent video clips rather than sound-on clips because knowing about candidate policy positions disrupts people's ability to judge the non-verbal cues that really matter." Benjamin was a co-author of this study with Jesse M. Shapiro of the University of Chicago.

After watching ten-second silent video clips of competing gubernatorial candidates, participants in the study were able to pick the winning candidate at a rate significantly better than chance. When the sound was turned on and participants could hear what the candidates were saying, they were no better than chance at predicting the winner. For the study, Benjamin and Shapiro showed 264 participants, virtually all Harvard undergraduates, ten-second video clips of the major party candidates in 58 gubernatorial elections from 1988 to 2002.

Researchers found that the accuracy of predictions based solely on silent video clips was about the same as or greater than the accuracy of predictions based on knowledge of which candidate was the incumbent and information about the prevailing economic conditions at the time of the election, including the unemployment rate and any changes in personal income for the year prior to the election.

You can imagine how political parties could use this information. The key is to choose candidates that look good when speaking. Hire some people to view silent clips of potential candidates from your party. Then persuade the highest scorer to run and put money behind that person.

What I'd like to know: Would still pictures be enough to predict winners? Just do it based on looks? I've read that this year the Democrats tried to field better looking candidates. If they did then I bet that helped.

Share |      By Randall Parker at 2006 November 16 05:51 AM  Politics Human Nature

Ned said at November 16, 2006 10:23 AM:

Interesting but somewhat dismaying finding. The candidate who LOOKS the best usually wins, not the most effective candidate, the most competent, the most honest or the one with the best ideas. Nevertheless, this is what modern politics is reduced to - a beauty contest. I wonder how Abraham Lincoln or Thomas Jefferson would have looked on the silent screen.

Fred said at November 16, 2006 10:36 AM:

This explains a lot. A friend of mine teaches a course titled "Democracy and Other Myths." I think he may find this article useful.

Someone (maybe Dwight McDonald)said all systems of govenment are really just oligarchies in disguise, or words to that effect. I agree, and I rarely even bother to vote anymore. Increasingly, it's like being given a choice between a ham and cheese sandwich, or a cheese and ham sandwich. "Have it your way."

Hank G said at November 16, 2006 10:58 AM:

Iraq is a diversion. As the army attacks Iraq, the US gov't erodes rights at home by suspending habeas corpus, stealing private lands, banning books like "America Deceived" from Amazon, rigging elections, conducting warrantless wiretaps and starting 2 illegal wars based on lies. Soon, another US false-flag operation will occur (sinking of an Aircraft Carrier) and the US will invade Iran, (on behalf of Israel).
Final link (before Google Books bends to gov't demands and censors the title):
America Deceived (book)

Half Sigma said at November 16, 2006 12:49 PM:

Warren Harding was chosen to be the Republican nominee because he looked good. It was the first national election when women were allowed to vote.

John S Bolton said at November 16, 2006 4:46 PM:

I wouldn't be so sure about all that.
These were Harvard undergrads, whose liberal views would cause them to consider the welfare and compassion-peddling candidate
to be the winner.
Or, if not so, then let's see the study repeated with evaluators more like the electorate.
politicians generally are neither handsome nor charismatic.
Their average age alone would tell you that.
If the people were so susuceptible to looks and aggressive, dominant body language, their representatives would be 25, athletic and not lawyers.

beowulf said at November 16, 2006 10:36 PM:

John S. Bolton,

That's a good point, Harvard undergrads aren't the typical voter, in political preferences or IQ. To a high IQ person, every politician sounds slow-witted, if they didn't, they'd be talking completely over the heads of many voters. I'd like to see this replicated by a random sample of voters.

As to your other point, the reason that no one votes for 25 year olds is you do need to bring at least a little gravitas to the table. That was one of the things that hurt John Edwards last time, the man was 50 but looked 35. The rap was he was too "inexperienced" but I think it was his youthful looks that cost him the nomination. If he looked like, say, Lloyd Bentsen, his one term of the Senate would have been enough experience to appear "presidential".

Bob Badour said at November 17, 2006 9:02 AM:

What does it matter whether the test subjects are the typical voter? The result shows they predict the winners (at least better than by chance) regardless. I don't even see what relevance their political leanings are unless one things the typical undergraduate has any clue who the past gubernatorial candidates were in foreign states.

It's not like the study looked at the current election: The test subjects picked the winners in elections from 1988 to 2002.

John S Bolton said at November 17, 2006 9:43 PM:

No, the point is that when they were allowed to consider the verbal presentation, the students got it wrong.
The general public might have shown much less difference in accuracy between their estimations from the visual-only, and the verbal-included.
The high-IQ may be more sensitive to very brief flashes of involuntary gestural information.
The electorate is not primed, nor yet selected-for, a tendency to see the further-right position as weaker.

Bob Badour said at November 18, 2006 8:11 AM:

Okay, I see what you are saying. How does that make elections any less beauty pageants?

Bob Badour said at November 19, 2006 9:57 AM:

It strikes me that this method might only measure the candidates' own assessment of the election. Humans naturally give off body language signals related to confidence. It would be difficult to separate those signals as cause and effect.

Certainly, a confident person would more likely attract votes, but someone who believes he will win the election would naturally feel more confident.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Web parapundit.com
Go Read More Posts On ParaPundit
Site Traffic Info
The contents of this site are copyright