In 2015, billionaire investor Marc Andreessen famously said, “software is eating the world.” In 2019, it might be more accurate to say social media is eating the world, or at least that social media is eating the internet. People, on average, spend over two hours per day on social websites. Two thirds of people get their news on social media. Many use it to land their next job or even find a significant other. And, most are active on multiple platforms.
But what can be inferred from all this social media use? For one, it is abundantly clear that people’s desire to form online communities overrides most concerns about personal privacy. In fact, it is now common for communities to form around the very topics (e.g., politics, economic status, etc.) that were once considered unseemly to discuss in some public settings.
New research published in the Journal of Personality shows just much can be learned about someone’s personality, beliefs, and character by examining their social media activity. Specifically, a team of researchers led by Dandan Pang at the University of Bern in Switzerland examined approximately four million tweets from over 4,400 individuals to test how accurately they could predict people’s underlying character strengths.
To perform this analysis, the researchers first asked participants to complete the Values in Action Inventory of Strengths (VIA-IS). The VIA-IS is a self-report questionnaire consisting of 240 items such as “I never quit a task before it is done.” Together, these items measure 24 dimensions of character. They are: appreciation of beauty and excellence, bravery, creativity, curiosity, fairness, forgiveness, gratitude, honesty, hope, humor, judgment, kindness, leadership, love, love of learning, modesty, perseverance, perspective, prudence, self-regulation, social intelligence, spirituality, teamwork, and zest.
Next, the researchers requested the Twitter handles from all 4,423 participants who completed the survey (participants were pre-screened to make sure they were active users of Twitter). Using a Twitter API, the researchers collected up to 3,200 of the most recent posts by each participant. They relied on two sophisticated machine learning algorithms to classify the contents of the tweets into categories and word clusters that could then be mapped onto the character dimensions from the VIA-IS.
The researchers’ primary goal was to see if and which dimensions of character could be predicted with a high degree of accuracy from people’s Twitter activity.
Here’s what they found. First, they showed that character strengths could, in fact, be reliably inferred from one’s Twitter use. They write, “The current study demonstrates that social media can be used to further characterize and predict character strengths. The prediction results suggest that language‐based assessments of character strengths may well serve as a cost‐effective and scalable alternate measurement system.”
Specifically, they found that the character dimensions of “spirituality” and “love of learning” were by far the easiest character dimensions to predict. Other character dimensions showing high prediction accuracy were “zest,” “appreciation of beauty and excellence,” and “gratitude.” Character strengths that were most difficult to predict from people’s Twitter activity were “prudence,” “judgment,” and “fairness.”
The researchers speculate that reason behind the differing levels of prediction accuracy across character traits may have more to do with the nature of the exercise than with the traits themselves. They write, “The difference in prediction values might be that certain character strengths were more manifest on social media platform, while other strengths were more hidden. For example, love of learning indicates a certain degree of openness, which is linked to more social media activities (e.g., more “likes” and larger network) and also a tendency to post more content, whereas prudence indicates a degree of introversion which may correlate with less social media use.”