Political polarization? Don't blame the web

"Our findings don't rule out that the internet has played some role in the recent rise in polarization," said Jesse M. Shapiro, study author and professor of economics at Brown. "But they cast doubt on some common narratives linking polarization to online news and social media."

Shapiro coauthored the study out today in the Proceedings of the National Academy of Sciences, "Greater Internet use is not associated with faster growth in political polarization among US demographic groups," with Levi Boxell and Matthew Gentzkow from Stanford University. The study expands on research that appeared as a working paper in March, and includes data for the 2016 presidential election.

The authors note that many studies and measures indicate that Americans have become increasingly polarized in recent years, and that numerous researchers and commentators attribute increased polarization in part to the rise of social media and the internet. A common concern among those researchers, Shapiro and his coauthors wrote, is the tendency for online news sources or social media circles to create "echo-chambers" of like-minded individuals who paint the opposition as perpetrators of outrages and close off opportunities for conversation.

To test the hypothesis that the web is a primary driver of rising polarization, Shapiro and his coauthors used data from the American National Election Study (ANES), a nationally representative, face-to-face survey of the voting-age population that has been conducted both pre- and post-election since 1948. The ANES collects data on Americans' social backgrounds, political predispositions, social and political values, perceptions and evaluations of groups and candidates, and other issues, according to the ICPSR, a data archive of research in the social and behavioral sciences where the studies are accessible.

The authors also utilized survey microdata on social media use from the Pew Research Center that covers the years 2005, 2008, 2011, 2012 and 2016, as well as from every presidential election year between 1996 and 2016.

Shapiro and his coauthors assessed whether demographic differences, in particular age, impacted trends in eight measures of political polarization, ranging from straight-ticket voting to partisan affect polarization -- the tendency of people identifying as Republicans or Democrats to view opposing partisans negatively and co-partisans positively.

For all but one of the eight measures of polarization, the increase was larger for the older group than for the youngest group. This is important because age is a strong predictor of internet and social media use. Less than 40 percent of those 65 and older and less than 20 percent of those 75 and older reported that they obtained information about the 2016 presidential campaign online. In contrast, over 75 percent of 18 to 39-year-olds got information about the 2016 presidential election online.

Within the 65+ age group, partisan affect polarization, which the authors identify as an especially important measure, grew at three times the rate it did for those aged 18 to 39.

"We find that the groups least likely to use the internet experienced larger changes in polarization between 1996 and 2016 than the groups most likely to use the internet," the authors wrote.

"Under appropriate assumptions, these facts can be shown to imply a limited role for the Internet and social media in explaining the recent rise in measured political polarization," the authors wrote.

Any explanation identifying political polarization as an outgrowth of internet or social media use, Shapiro notes, would have to account for the rapid increase in partisanship among those with limited internet use and negligible use of social media.

"I think the main culprits in explaining the rapid rise in polarization are probably to do with forces broader and deeper than the digitization of the news," Shapiro said.