Among the prevailing explanations for the still-surprising victory of Donald J. Trump is the idea that the liberal media, and the entire Clinton campaign, never saw it coming because we have insulated ourselves ideologically. The coastal elites in Washington and New York, the conventional wisdom goes, have been ensconced in like-minded bubbles, oblivious to the populist upsurge across the country.
This line of thinking has come to characterize much of how we engage with the world politically in general, on both the left and right. For every liberal bubble, there is now an equal and opposite conservative bubble; for every Huffington Post a Breitbart. We see this in our daily personal interactions online, as we construct then inhabit our own private Filter Bubbles, as Eli Pariser laid out in his 2011 book of the same name, unperturbed by opposing ways of thinking. More recently, as Cass Sustein explains in #Republic: Divided Democracy in the Age of Social Media, these sorts of information cocoons aren't simply to our individual detriment, but have become a real threat to democracy itself. In essence, this is what the entire debate about Fake News hinges on. Anything that isn't being reported by our pre-selected group of trusted sources is immediately dismissed.
Social media platforms have become the chief culprit for this bifurcation, and for sound reason. When we can literally choose to tune out anyone with a different ideology, it makes sense that the constant reinforcement would only compound itself over time. But, as a just published study from a trio of economists at the National Bureau of Economic Research offers, that may not be as clear cut an explanation. In fact, as the study points out, political polarization is on the rise regardless of levels of Internet use.
The concept of polarization is taken as a given in the study, referencing the above mentioned authors, among others, but its findings come from other recent research. On Monday, Pew released data that shows an exacerbated level of political divide among generations, with younger people increasingly identifying as further left and older ones further right.
The NBE study, led by a trio of economists from Stanford and Brown, also looked at political divide on a generational level, but reached a conclusion that bucks the conventional wisdom about the effect the Internet has on our ideology.
"In a nutshell, what we find is the groups that have the greatest propensity to get their news online are not the groups that have shown the biggest increase in polarization," Jesse Shapiro of Brown University explained to me by phone Monday.
"A lot of people believe polarization is on the rise in the U.S., and a lot of have attributed that to the roll of online news and social media, in particular with the idea that online media creates echo chambers where people only hear their own opinions," he said. "[Our] evidence does not square with the simplest version of that narrative."
Instead, they found, it's people who rarely, if ever go online—those 75 and older—that exhibit the furthest swing to the extreme of political ideology.
Using data from the American National Election Study, Shapiro's team examined nine measures of political polarization, including things like a propensity for straight-ticket voting, issues consistency, and religious affiliation, then broke down respondents into demographics that have been shown to predict Internet and social media use. Less than 20 percent of those 75 and older reported using social media in 2012, they explain, whereas 80 percent of people ages 18-39 used the services.
"I was personally surprised at how consistent the pattern was across the different measures of polarization," Shapiro said. "Pretty much for all of them, the older groups are seeing faster polarization. I thought we would see more variability."
The paper doesn't offer a suggestion for why it might appear that those less likely to be online seem to be the most likely to entrench themselves politically. "My own view is the reason is probably nothing to do with changes in media and technology, but maybe in broader socio-economic forces like inequality," he said.
Of course it's possible, at least anecdotally speaking, that all of this could become a moot point in the current political climate, where the concept of self-selected information bias in the age of Trumpism has seemingly exploded. They'll continue to update their work over time, Shapiro says. "We'll try to update as soon as new data is out and see if the story is different," he said. "I think given how much the media environment has changed [in the past few months] it has already shed some light in the role of online media, but it will be interesting to see what we get."
And once we get it, it will be interesting to see if any of us believe it.
