Penn researchers use Facebook as psychology tool
Every day, millions of people post billions of words on Facebook, Twitter, and other social media sites. Far from seeing this type of communication as mere ephemera, an interdisciplinary research team set to find out whether the information could be used as a veritable window to the soul.
In a new study, which included Martin Seligman, director of the Positive Psychology Center in the School of Arts & Sciences, Lyle Ungar, a professor in the Department of Computer and Information Science in the School of Engineering and Applied Science, and H. Andrew Schwartz, a postdoctoral fellow in both departments, researchers had 75,000 volunteers complete a common personality questionnaire through a Facebook application. The volunteers then allowed the researchers to monitor their Facebook status updates and look for overall linguistic patterns in the language they used on the site.
For decades, psychologists have looked at the words people use as a way of gaining insight into what is going on in their minds. This kind of linguistic analysis, however, has historically faced two main challenges. One is that researchers needed to decide beforehand which words are associated with certain personality traits. The other is that the most common words make up the majority of people’s vocabulary; gathering writing or speaking samples large enough to find truly meaningful correlations between words and traits was prohibitively difficult. The “closed” nature of the vocabulary and correlations in this approach were inherent limitations.
The researchers saw Facebook as a way to test an “open” approach to this kind of linguistic analysis, drawing on the language volunteers with different traits used to inform what words were important.
“The billions of words available in social media allow us to find patterns at a much richer level,” says Schwartz, who is the study’s lead author.
To visualize their results, the researchers created word clouds that summarized the language that statistically predicted a given trait, with the correlation strength of a word in a given cluster being represented by its size. For example, a word cloud that shows language used by extroverts prominently features words and phrases like “party,” “great night,” and “hit me up,” while a word cloud for those who scored low on the neurotic scale used a greater number of words that referred to active, social pursuits, such as “snowboarding” or “basketball.”
The latter case demonstrates one of the advantages of the researchers’ “open” approach—it reveals correlations that wouldn’t be obvious beforehand.
“This doesn’t guarantee that doing sports will make you less neurotic; it could be the neuroticism causes people to avoid sports,” Ungar says. “But it does suggest that we should explore the causation in the link.”
The researchers feel this “big picture” view of people’s internal states could be an important new tool for psychological studies.
“When I ask myself, ‘What’s it like to be an extrovert?’ or ‘What’s it like to be neurotic?’ these word clouds come much closer to the heart of the matter than do all the questionnaires in existence,” says Seligman.