According to new research guaranteed to bring out your inner nerd—the University of Pennsylvania released the most comprehensive analysis of language to date—some age and gender stereotypes ring surprisingly true. At least, on Facebook.
The Word Well Being Project (WWBP) suggests that a glance at your Facebook status says a lot about your age, gender, and even personality. The report, designed to shed “light on psychosocial phenomena using language analysis,” can, in theory, predict shifts in society’s attitudes towards different aspects of life and culture.
Are the words in our vocabulary really so telling? Yes, as far as the study is concerned.
For this study, 75,000 people were recruited to take part, using a Facebook app called “My Personality.” Users then provided some basic information—like age and gender—and took a quick personality test. From there, the app monitored the users’ status updates. Over time, researchers were able to discern which words and phrases were most frequently associated with people based on location, age, gender, or personality type.
Orwellian? Maybe. Interesting? Absolutely.
The study highlighted a number of surprising (and some not-so-surprising) findings—especially when it comes to their breakdown of gender. For example, when referring to one’s significant other, male participants were more likely to use possessive terms than the female participants In other words, the male participants were far more likely to say “my girlfriend/boyfriend/wife/husband/partner,” whereas the study’s female participants were more likely to simply say “girlfriend/boyfriend/wife/husband/partner” without the possessive “my.”
The study also found that female participants were more likely to make use of emoticons than their male counterparts, and much more frequently used singular, first-person pronouns (“I,” “he,” “his,” “she,” “her,” etc.). The male participants used fewer first-person pronouns, but more “formal, affirmation, and informational words,” as well as “object references” (“xbox,” “TV” etc.) and swear words. Women, on the other hand, were predisposed to more frequently discuss social interaction, and use terms to describe feelings.
The study’s analysis of the relationship between age and language turned up a number of interesting, if not obvious, results. For example, 19 to 22-year-olds were more likely to use terms like “drunk,” “hangover,” and “wasted,” to describe a night of beverages, while 23 to 29-year-olds stuck to the slightly more refined “beer,” “drinking,” and “ale.” When discussing school, 13 to 18-year-olds were more likely to use terms like “school,” “homework,” and of course, “ugh,” while 19 to 22-year-olds were more predisposed to use terms like “semester,” “college,” and “register.”
Does this study confirm stereotypes? Perhaps, in a roundabout manner, it does. Does it reinforce gender norms? As a purely observational study, this wouldn’t seem to be a concern in itself. However, should this type of data be used in marketing and advertising to craft gender-specific campaigns, it’s safe to say that we can all expect a little extra gender binary-supporting timeline filler in our futures.
“Online social media such as Facebook are a particularly promising resource for the study of people, as ‘status’ updates are self-descriptive, personal, and have emotional content,” reads the study. “Language use is objective and quantifiable behavioral data, and unlike surveys and questionnaires, Facebook language allows researchers to observe individuals as they freely present themselves in their own words. Differential language analysis (DLA) in social media is an unobtrusive and non-reactive window into the social and psychological characteristics of people’s everyday concerns.”
In other words, given the right amount of data, it’s possible to predict a fair amount about a person based on their status updates. Using the algorithm developed by the University of Pennsylvania, they were able to accurately predict someone’s gender based on the words and phrases they use on Facebook with a 91.9 percent success rate.
Similar data studies have been used to track seasonal shifts in mood, to detect a flu epidemic before the Center for Disease Control is even aware of its existence, and even to predict the stock market’s performance. The difference between those studies and this one was that they viewed people from a limited perspective without factoring in the social aspect of language, and the subtle variations between groups.
The researchers behind this study hope that using this new analysis, they will be better able to measure psychological well-being among groups, and track social change.
Whether you’re as predictable as the sunrise or you live the life of a wild card, one thing is certain: Like it or not, we will all find ourselves hit with a moment of self-consciousness before posting our next status update.
(Featured Image via)