Thursday, September 6, 2007

Do We Overvalue Expert Advice?

We are just a year away from the US presidential elections, and a question that keeps coming up during broadcast debates with the candidates is "are you experienced?"

Obviously, it's a valid question to ask. After all, we expect candidates to have a long history of political involvement and the expertise required to make the right decisions in sticky situations.But a new study appearing in the journal Interfaces suggests that experience may be overrated anyway. It shows that people without any expertise in foreign policy can make snap predictions about the outcome of major conflicts almost as well as experts.

Researchers from the University in Australia and the University of Pennsylvania asked policy experts and undergraduates to assess slightly disguised versions of real conflict situations. These included a 1970s border dispute between Iraq and Syria, a nurses' strike, and an unfolding dispute between football players and their management.The researchers received 106 responses from policy experts and 169 from undergraduates. The experts accurately predicted the outcome of conflicts in 32% of the cases, only slightly better than the 29% score achieved by the students. Chance guesses would have given a score of 28% on the test. Experts with less than five years of experience also achieved a 36% accuracy rate – outdoing more senior people in their profession who were 29% accurate.

In conclusion, experts do use their judgement to predict what will happen. However, their forecasts are of little value in terms of accuracy. Sometimes, they lead people into false confidence.

from New Scientist Short Sharp Science blog by Roxanne Khamsi

No comments: