Wharton Professor Philip E. Tetlock who works with the Office of the Director of National Intelligence on the Good Judgment Project.
The creative uses to find lost submarines by the Navy, such as the Scorpion after it sank in the North Atlantic in 1968, using the intelligence community, economic prognosticators, and the armed services, are now coming into play as the world seeks the whereabouts of Malaysia Airlines Flight 370 after its mysterious disappearance during its flight from Kuala Lumpur to Beijing, Friday, March 14, 2014.
In the case of the Scorpion, the Navy used a kind of crowdsourcing, but in that case using forecasters from diverse and relevant areas–submariners, ocean salvagers, physicists, and engineers. Each made an educated guess as to where the submarine lay based on different scenarios. Back then, they bet bottles of Chivas Regal for the winner. One Navy scientist, John P. Craven, zeroed in on an area about 400 miles from the Azores. The sub was found about 200 yards from the spot Dr. Craven estimated.
Wharton’s Philip E. Tetlock works with the Office of the director of National Intelligence on the Good Judgment Project, which tests forecasts of geopolitical events.
It is not clear if Dr. Tetlock has been called on in the hunt for Flight 370, although he and a colleague in Penn’s computer and information science department might. Such an elite crowd might include climate and ocean scientists, Boeing 777 engineers and commercial pilots who have flown the 777 and decision-making and computer scientists. They would deliberate and judge based on the myriad scenarios currenty being played out in the media and among search crews: terrorism, pilot error, sudden depressurization, and engine failure.
The experts would then be grouped into teams and, based on their current level of information, their prognostications would be weighted, evaluated and tested, according to another Penn faculty member, Lyle H. Ungar, a professor in the department of computer and information science who works on the forecasts.
Penn Professor Lyle H. Ungar
The search method is called “pooled forecasting.” It’s based on the idea that each person offers a small bit of insight and that those “weak signals from diverse experts accumulate quickly,” said Prof. Tetlock.
It’s a very sophisticaled version of Gladwell’s “bucket brigade” to douse a burning house or more precisely, a village turning out to find a lost child even though they don’t know the child is there.
Certainly the families, friends, and loved ones of the 239 people aboard the flight along with the 25 countries now searching for the plane have every reason to find it. And Penn may be playing a part of that effort.
Here’s Benedict Carey’s piece about Penn’s piece in the Good Judgment Project: