Your Facebook Likes Predict if You’re Gay (But Can They Predict What You’ll Buy?) [Download]

• Your Facebook Likes can predict your sexual orientation with 88% accuracy .
• If you’ve Liked Desperate Housewives or Britney Spears, you’re more likely to be gay. On the other hand, Liking Shaq is predictive of heterosexuality.
• Liking Harley Davidson or the retailer Sephora is predictive of low intelligence, whereas Liking Curly Fries is predictive of high intelligence.
• If you’ve Liked Hello Kitty, you’re more likely to be creative, but not very conscientious.
• Oh, and if you Like Camping, you’re more likely to be neurotic.

Just some of the headline findings of ground-breaking research from the University of Cambridge published yesterday in the Proceedings of the National Academy of Sciences.

The study, Private Traits and Attributes are Predictable from Digital Records of Human Behavior (full download), authored by researchers Michal Kosinskia, David Stillwell and Thore Graepel is based on sample of 58,466 volunteers from the United States, obtained through the myPersonality Facebook application (www.mypersonality. org/wiki), which included their Facebook profile information, a list of their Likes (n = 170 Likes per person on average), psychometric test scores, and survey information.  Patterns between Likes and psychometric data were then correlated, patterns found, and predictive power was measured.

The University of Cambridge study, funded by Microsoft and Boeing, has spawned a free new online personality test, youarewhatyoulike.com, based on your Likes (see example profile below).

Welcome to the Brave New World of “Big Data” - massive, fast changing and diverse datasets characterized by the 3 V’s of Volume, Velocity and Variety.  If the privacy minefield can be negotiated, the opportunity for brands and retailers is clear – harness these datasets to deliver better customer experiences by delivering the right information, service or products to the right people at the right time.

For example, this study – whilst not focused on consumer behaviour – might tell the management of Sephora to target the intellectually-challenged, and avoid marketing, signage and store design that is too mentally taxing.  On the other hand  there are advertising opportunities for Curly Fries – buy up science spots and science site ad placements. Own a camping store or site? Then plan the layout and offers for neurotics.  Of course, this is reductio ad absurdum, but it is the future.

The immediate opportunity for tech companies, brands and retailers is clear – replicate this study with a shopping and media focus and find patterns in Like data predictive of shopper behaviour, values, lifestyles and personality.

Facebook, you just got useful.

 

 

Kosinski, M., Stillwell, D.J & Graepel, T. (2013) Private traits and attributes are predictable from digital records of human behavior. Proc. of the National Academy of Sciences

We show that easily accessible digital records of behavior, 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 sepa- ration, age, and gender. The analysis presented is based on a dataset of over 58,000 volunteers who provided their Facebook Likes, detailed demographic profiles, and the results of several psychomet- ric tests. The proposed model uses dimensionality reduction for preprocessing the Likes data, which are then entered into logistic/ linear regression to predict individual psychodemographic profiles from Likes. The model correctly discriminates between homosexual and heterosexual men in 88% of cases, African Americans and Caucasian Americans in 95% of cases, and between Democrat and Republican in 85% of cases. For the personality trait “Openness,” prediction accuracy is close to the test–retest accuracy of a standard personality test. We give examples of associations between attri- butes and Likes and discuss implications for online personalization and privacy.