Researchers at the University of Cambridge say that, using publicly-available Facebook Likes only, it's possible to get surprisingly accurate estimates of users’ race, age, IQ, sexuality, personality, substance use and political views.
"I am a great fan and active user of new amazing technologies, including Facebook. I appreciate automated book recommendations, or Facebook selecting the most relevant stories for my newsfeed," says said Michal Kosinski, Operations Director at Cambridge's Psychometric Centre.
"However, I can imagine situations in which the same data and technology is used to predict political views or sexual orientation, posing threats to freedom or even life."
The team analysed a dataset of over 58,000 US Facebook users, who volunteered their Likes, demographic profiles and psychometric testing results through the myPersonality application.
Facebook Likes were fed into algorithms and cross-checked with information from profiles and personality tests. The researchers then created statistical models able to predict personal details using Facebook Likes alone.
Their models, they say, proved 88 percent accurate for determining male sexuality, 95 percent accurate in distinguishing African-American from Caucasian American and 85 percent accurate differentiating Republican from Democrat. Christians and Muslims were correctly classified 82 percent of the time, and even relationship status and substance abuse could be accurately predicted between 65 and 73 percent of the time.
This information, though, didn't come from obvious Likes such as 'gay marriage', but relied on inference - aggregating huge amounts of less informative but more popular Likes such as music and TV shows.
Even seemingly obscure personal details such as whether users’ parents separated before the user reached the age of 21 were accurate to 60 percent - enough to make the information 'worthwhile for advertisers', suggest the researchers.
The researchers also tested for personality traits including intelligence, emotional stability, openness and extraversion, and found equally accurate results. When it comes to the openness trait, for example – the spectrum of those who dislike change to those who welcome it – observation of Likes alone turned out to be just as informative as an individual’s actual personality test score.
Some Likes had a strong but seemingly incongruous or random link with a personal attribute, such as Curly Fries with high IQ, or That Spider is More Scared Than U Are with non-smokers.
Another team member, Thore Graepel from Microsoft Research, says he hopes the research would prompt Facebook users to think more about privacy.
"Consumers rightly expect strong privacy protection to be built into the products and services they use and this research may well serve as a reminder for consumers to take a careful approach to sharing information online, utilising privacy controls and never sharing content with unfamiliar parties," he says.