Social Media Posts Help Build Predictive Model Of Personality
September 26, 2013

Facebook Used To Determine Age, Gender And Personality

Lee Rannals for – Your Universe Online

People don't have to read your profile to know you; new research indicates that you can find out quite a lot about someone through observing their online personality.

Penn Researchers wrote in the journal PLOS ONE that they were able to use Facebook data to determine a user's age, gender and personality traits. The study involved 75,000 volunteers, who completed a common personality questionnaire through a Facebook application and made their Facebook status updates available.

The team looked for overall linguistic patterns in the volunteers' language and generated computer models to predict an individual’s age, gender and response to the questionnaire. These models were able to predict the user's gender based only on the language of their status update 92 percent of the time. They were also able to predict a person's age within three years more than half of the time.

“Our personality predictions are inherently less accurate but are nearly as good as using a person’s questionnaire results from one day to predict their answers to the same questionnaire on another day," Hansen Andrew Schwartz, a postdoctoral fellow in computer and information science and the Positive Psychology Center, said in a press release.

To predict a personality, the team used an open vocabulary approach instead of a "closed" approach for analyzing the data. In a closed vocabulary approach, psychologists might pick a list of words they think signal positive emotion and then look at the frequency of a person using these words to measure how happy they are. This closed approach has several limitations and does not always measure what is intended.

“For example, one might find the energy sector uses more negative emotion words, simply because they use the word ‘crude’ more. But this points to the need to use multi-word expressions to understand the intended meaning," Lyle Ungar of Computer and Information Science said in the release.

The open vocabulary approach derives important words and phrases from the sample itself. The team sifted through more than 700 million words, phrases and topics, taken from Facebook status messages. They isolated the words and phrases clustered around the various characteristics that were self-reported in the volunteers' questionnaires. The team also used the "Big Five" personality traits, which include extraversion, agreeableness, conscientiousness, neuroticism and openness.

Researchers created word clouds that summarized the language that statistically predicted a given trait, which helped provide a window into the psychological world of people.

“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," said Martin Seligman, director of the Positive Psychology Center.

By building a predictive model of personality based on the language of social media, the team can use it to explore possibilities like whether a neurotic individual would become more emotionally stable if they played more sports.

“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," Johannes Eichstaedt, a postdoctoral fellow, said in the statement.