I don't suppose you read the Proceedings of the National Academy of Sciences very often. I know that I don't, despite such papers as "Possible relation of water structural relaxation to water anomalies" and "Differentiation of integrals in higher dimensions."
But thanks to Robert Lee Hotz, writing in the Wall Street Journal, I did dig up a recently released paper from a group of British researchers at the Psychometrics Centre of the University of Cambridge entitled "Private traits and attributes are predictable from digital records of human behavior." Notwithstanding the title, every Facebook user should pay close attention to what it details, as it sounds an alarm to all Facebook users, and that likely includes you and your kids.
Using advanced computer models, the researchers took the data provided in demographic profiles and compared this data to the Facebook "likes" provided by the 58,000 volunteer American Facebook users. The researchers found that "Facebook Likes can be used to automatically and accurately predict a range of highly sensitive personal attributes including: sexual orientation, ethnicity, religious and political views, personality traits, intelligence, happiness, use of addictive substances, parental separation, age, and gender."
In 88 percent of the cases, the pattern of "likes" accurately showed who was gay and who was heterosexual. It could accurately differentiate between African Americans and Caucasian Americans in 95 percent of cases. It distinguished between Democrats and Republicans in 85% of cases.
The purpose of the study was to determine whether Facebook likes could provide "easily accessible digital records of behavior" which could be highly useful to social scientists. The answer is clearly yes. But such data — down to the level of an individual Facebook user — is potentially even more valuable to a marketer.
Facebook Inc. public-policy manager Fred Wolens told reporter Holtz that "the ability to use public 'Likes' to profile its users is 'hardly surprising', marketers and social scientists have long used online information, from ZIP Codes to musical tastes, to predict personal characteristics."
Why is this ability to distinguish specifics about Facebook users by studying their "likes" potentially dangerous? The researchers provide an example.
"People may choose not to reveal certain pieces of information about their lives, such as their sexual orientation or age, and yet this information might be predicted in a statistical sense from other aspects of their lives that they do reveal. For example, a major U.S. retail network used customer shopping records to predict pregnancies of its female customers and send them well-timed and well-targeted offers. In some contexts, an unexpected flood of vouchers for prenatal vitamins and maternity clothing may be welcome, but it could also lead to a tragic outcome, e.g., by revealing (or incorrectly suggesting) a pregnancy of an unmarried woman to her family in a culture where this is unacceptable."
The bottom line here is to realize that every time you "like" something on Facebook, you are telling the world a little something about yourself. Those "little somethings" can eventually add up to a very detailed portrait of you, revealing things you might not like the world to know.