THURSDAY, March 19, 2009 (HealthDay News) -- Imagine you're going to a movie. What would better predict whether you'll like the film: the opinion of a friend who's seen it or the details you know about the cast and plot?
Researchers say you'd be best to rely on your pal.
"If you are trying to figure out whether to become a lawyer or vacation in Bermuda, our research suggests you would do much better to ask people who are lawyers or have been to Bermuda about their experience than to gather factual information about each of these decisions," said Matthew Killingsworth, a Harvard graduate student and an author of the research, published in the March 20 issue of Science.
In essence, "if you want to know how happy something will make you, ask someone who's been there or done that," he said.
What does it matter? The study suggests that people are prone to make poor predictions when relying on what they know about something instead of on the opinions of others. Making matters more complicated, "people do not believe this," the study authors wrote.
In the study, researchers set up "speed dates" between heterosexual female and male undergraduates. The women met men for five minutes and then told researchers what they thought of them. The experiment involved 33 women and eight men.
Women did a better job of predicting whether they'd enjoy a date with a man if they'd previously read another woman's opinion about him based on a speed date. They did worse at predicting whether they'd like a man if they simply had a photo and basic information about him before the date.
In fact, the women were 50 percent less likely to wrongly predict whether they would enjoy the date if they had heard about the man from another woman, Killingsworth said.
Still, he said, "women think they will make better predictions based on the guy's profile."
Killingsworth said people simply overvalue the worth of "factual information."
For instance, someone thinking about becoming a lawyer or traveling to Bermuda "might become overly focused on the lawyer's salary or the crime level in Bermuda and run the risk of vastly undervaluing the information provided by people who've actually had the experience you're contemplating," he said.
Dr. Peter A. Ubel, professor of medicine and psychology at the University of Michigan, said the study is well done but would be of limited use in the everyday world. "I'd like to see [similar research] in more realistic domains of life," he said.
Still, Ubel said, the findings suggest that "the things we think should influence our decisions -- like, 'Tell me more about the situation' -- don't always help us."
And in general, he said, "we often do a rotten job of making predictions."
The University of Waterloo, in Ontario, Canada, has more on making decisions.