Twitter will turn 10 next year, meaning it’s been a long time since it was the Internet’s shiny new thing. We now take for granted that it’s an important—if often vitriolic—platform for public policy debate, including the high-pitched battles over education reform.
What is new is the use of Twitter as an analytic tool. For example, the Consortium for Policy Research in Education (CPRE) made a splash in February with an innovative study of how the Common Core debate is playing out on Twitter; scholars found, among other things, that proponents tend to make policy points while opponents use “political language” in their tweets.
Innovative though that research may be, it’s still fundamentally about Twitter. Another strand of research uses data from Twitter to measure other phenomena. Perhaps the best-known example is the work of University of Pennsylvania psychologist Johannes Eichstaedt, popularized earlier this year by an article in the New Yorker. He started with the Linguistic Inquiry and Word Count (LIWC) program, a decades-old algorithm developed by psychologist James Pennebaker and colleagues that can determine people’s psychological states and personality traits by analyzing their speech or writing. He and his team then analyzed hundreds of millions of tweets and used them to code the emotional state of the U.S. counties from which they originated. Remarkably, Eichstaedt et al. found higher death rates from heart disease in counties where residents’ tweets tended toward words related to “hostility, aggression, hate, and fatigue.”
This got me thinking: What does Twitter say about the tone of the education policy debate? And in particular, the emotional state of its combatants? To find out, I started with a list of the 25 top education-policy people on Twitter that I published on the Education Next blog in August 2014. (I’ve been producing similar lists annually since my Fall 2011 article, “All A-Twitter about Education.”) Then my colleagues used Pennebaker’s analyzewords.com website—an offshoot of his LIWC—to analyze the recent tweets of those 25 individuals. Table 1 presents the results.
One methodological note: The tool only analyzes a handful of tweets at a time (which themselves are very short, at 140 characters or less), so we ran the test twice, a few days apart in late March, to ensure consistency. Individuals are only included in the findings if they scored “high” or “very high” on the various domains during both tests.
So what might we conclude? It appears that many of the leading tweeters in education policy are “arrogant/distant,” meaning we are “well read” and “use big words.” Good for us! Before we think too highly of ourselves, though, we should consider that many people use Twitter to comment on pop culture, sports, or their daily routines; surely any public-policy debate will look smart compared to that.
It’s interesting to see how similar some of the tweeters score. Jose Vilson and Sabrina Stevens, two of the few teachers on the list, both score high or very high on upbeat, plugged-in, and analytic. And it appears that Randi Weingarten and Michelle Rhee were separated at birth, with similarly high scores for upbeat, plugged-in, and arrogant/distant. Maybe they have more in common than they think.
Fun though this is, this exercise, like the CPRE analysis, is still about how people behave on Twitter. Might education scholars start to use Twitter to study how people behave or feel in the real world? For example, what if we could examine high school students’ tweets to determine their level of engagement in school? And correlate that to student achievement? Might social media provide an answer to the elusive question of how to measure noncognitive skills? Or could social media provide an effective means to determine teachers’ emotional states, and how they relate (or not) to their states’ or districts’ accountability regimes?
Speaking personally, I can say that using Twitter has become a lot less fun over the years, thanks to the nasty behavior that is so pervasive now on social media. But using Twitter to study education reform—now that could be a whole new adventure.
Michael J. Petrilli is president of the Thomas B. Fordham Institute and an executive editor at Education Next.