|Do you feel smarter?|
You: In college I thought that what people liked—especially their favorite music, movies, and books—told you everything you needed to know about them. It was one of the surprising discoveries of adulthood that the complexities of personality can't be defined quite so easily or simply.
Or can they? Researchers at the University of Cambridge have found that Facebook "likes" can be used to guess with a high degree of accuracy your gender, sexual orientation, political leanings, religious affiliation, whether you're a smoker, and even whether your parents split up before you were 21. As Geoffrey Mohan writes, perhaps it's not unexpected that someone who likes "Proud to be a mom" is most likely female and a parent. But more surprising is that David Stillwell and his Cambridge colleagues found that likes also correlate with traits such as intelligence, openness to new experiences, and conscientiousness—at least, as far as these traits can be accurately measured by personality tests. ("Liking" curly fries correlates with high IQ, while liking Hello Kitty indicates that you are open to new experiences but not highly conscientious.)
One or two of these correlations could be dismissed as statistical happenstance; taken together, they can create a highly detailed picture of who you are. And it doesn't matter whether, in any individual case, the correlations are accurate, as long as employers, insurance companies and marketers think that they are. What you like can say a great deal more about you than you may realize. (Geoffrey Mohan, "On Facebook, you are what you 'like,' study finds" Los Angeles Times, March 11, 2013)
Behavioral economist Alessandro Acquisiti has explored what motivates us to act against our own interests and surrender important personal information. And he's made some disturbing findings:
- We are inconsistent: If shoppers are offered a $10 coupon and told that it will be increased to $12 if they share information about their purchases, only about 50% agree to share. But if they are offered a $12 coupon and told that it will be decreased to $10 if they keep their data private, 90% agree to share.
- We are irrational: Participants in a study of lying, stealing and drug use were far more willing to share data that could personally identify them if they were given a greater degree of choice about what information to share.
- And we're distractable: Students surveyed about cheating were far more likely to admit to it if an unrelated offer popped up while they were answering the survey.