What your Facebook Updates Say about You, and How Social Media is Changing Research

What do you post about on Facebook?

Maybe you’re the kind to post updates about the daily details — what you ate for breakfast, what errands you ran, and what you wore to run them. Maybe you’re the kind to post only when you have brag-worthy news, like your recent promotion or the rare tickets you scored to see a massively popular band. Maybe most of your posts are about your feelings, or your kids, or your job, or your romantic relationship, or your hobbies.

But what does your Facebook status update really say about you? And can others suss out your personality type, age, and gender based solely on the words you use?

Spoiler alert: yes. Yes, they can.

An interdisciplinary team of researchers from the University of Pennsylvania recently gave 75,000 volunteers a common personality quiz and then mined their Facebook status updates to make connections between personality and psychological traits and behaviors.

After identifying linguistic patterns in the volunteers’ language, the researchers then built computer models that managed to predict the their age, gender, and personality questionnaire responses with surprising accuracy. For example, the models correctly predicted gender based on language usage 92% of the time, and pinpointed the volunteer’s age within three years more than 50% of the time.

This study was different than previous psychological language studies in that it drew from volunteers’ own natural words, phrases, and topics — over 700 million of them, to be exact — rather than a preselected set. The researchers believe that using this “open” method of analysis as opposed to the “closed” methods that came before results in qualitatively better language data. And with better language data comes better understanding of personality traits, and deeper insight into the relationship between the two.

All of which was made possible thanks to the reach and engagement of a giant social network, I might add.

But here’s where it gets really interesting. From their data, the researchers created word clouds based on specific personality traits, age, and gender. Perhaps unsurprisingly, words like homework, school, tomorrow, and emoticons like XD, :D, and <3 topped the usage scales for volunteers aged 13 to 18. For those aged 23 to 29, the predominant words were at work, new job, wedding, office, beer, and drinks.

The more extraverted among us most often used words like party, can’t wait, girls, baby, and love you, while the less extraverted used internet, anime, manga, computer, and sigh.

From the outset some aspects of the word clouds may seem obvious, but researchers were able to glean more profound information, such as those who post about sports are generally more emotionally stable. In the future, psychologists and psychiatrists could use these predictive models of personality to more succesfully measure mental states — and provide targeted help those who need it the most.

Martin Seligman, director of the Positive Psychology Center, said in a press release: “When I ask myself, ‘What’s it like to be an extrovert?’ ‘What’s it like to be a teenage girl?’ ‘What’s it like to be schizophrenic or neurotic?’ or ‘What’s it like to be 70 years old?’ these word clouds come much closer to the heart of the matter than do all the questionnaires in existence.”

Johannes Eichstaedt, a grad student on the research team, agreed: “It may seem obvious that a super extraverted person would talk a lot about parties, but taken all together, these word clouds provide an unprecedented window into the psychological world of people with a given trait. Many things seem obvious after the fact and each item makes sense, but would you have thought of them all, or even most of them?”

“Researchers have studied these personality traits for many decades theoretically, but now they have a simple window into how they shape modern lives in the age of Facebook,” Eichstaedt pointed out.

I love studies like this because they highlight the good of social media, and emphasize the human side of “big data,” which is often solely seen as the friend of the marketer and enemy of the consumer. Will we see more research moving away from surveys and towards anonymized social media feeds? It’s likely.

Would you let your Twitter or Facebook posts be used for science? Let me know in the comments below!