5 ways to prepare your advertising infrastructure for disaster

Liking Jennifer Lopez will get you more Facebook friends than Iron Maiden, straight men like professional wrestling more often than they like Glee, and Mormons are more agreeable than fans of Timmy from South Park.

That's just a grab-bag from research just published at the Proceedings of the National Academy of Sciences, conducted by the University of Cambidge's Psychometric Centre and led by David Stillwell of Microsoft Research.

Using an analysis of logs of nearly 60,000 volunteers, including users of an app developed by Stilwell called myPersonality (he notes in a conflict of interest statement in the study that he received revenue from the app), the research has a serious point: merely using the Facebook “Like” button leaves a digital footprint sufficient to predict your political leanings, sexual orientation, personality or whether or not you smoke.

As the study states: “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.”

The accuracy of the “personality model” varies depending on the characteristic the researchers were looking at: the researchers claim 95 percent accuracy at using “Likes” to distinguish caucasions from African-Americans, but only 65 percent accuracy in identifying drug users. At 88 percent, gay males are easier to identify from “Likes” than lesbians (75 percent); Democract-versus-Republican alignment was predicted with 85 percent accuracy, they claim, while gender is easy at 93 percent.

Some of these may well be revealed by users in their profiles, of course, but the research is illustrating that even without profile information, the “Like” button can get a fair way under the skin of someone's public persona.

According to Associated Press, Facebook dismissed the findings as nothing new: "The prediction of personal attributes based on publicly accessible information, such as ZIP codes, choice of profession, or even preferred music, has been explored in the past," Facebook's Frederic Wolens said in a written statement.

Still, it's probably a timely reminder that the digital footprint persists. As the study notes: “the predictability of individual attributes from digital records of behavior may have considerable negative implications, because it can easily be applied to large numbers of people without obtaining their individual consent and without them noticing.” ®

Email delivery: Hate phishing emails? You'll love DMARC