:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/49263179/514585162.0.jpg)
One of the biggest differences in the way I experience the 2016 Democratic primary compared with the 2008 or 2004 version is through the rise of Twitter.
Twitter, which launched in 2006 but has grown to become a crucial part of any 2016 campaign, gives journalists like me quick and easy access to something it used to take hard legwork to get — direct and unmediated takes on the campaign from rank-and-file voters.
We have constant, direct access to passionate supporters of candidates who aren't full-time professional political pundits, campaign operatives, or interest group leaders. It's a window into what voters really think.
But this turns out to be one of those scenarios where a little knowledge is a dangerous thing.
Twitter is not all that popular with the American people as a whole, and most people who use Twitter don't tweet much. Surveys show that most Americans don't talk about politics online at all. There is no reason to believe that the people who do argue about politics on Twitter are in any way a representative subset of people as a whole, and a mountain of evidence to suggest the opposite.
The overall result is that pundits watching their mentions are mainlining a deeply misleading view of the Democratic primary — one that features a passionate and bitter race that simply doesn't exist in the real world, complete with starkly polarized views on issues that don't exist among actual voters.
Most people follow the motto, "Never tweet"
Only 23 percent of adults who have an internet connection use Twitter at all, and in general, discussions about politics are not a popular use case for the internet.
A majority of every education demographic has *never* talked about politics on the internet (via @MattBruenig) pic.twitter.com/IQbKyo2DvD
— Derek Thompson (@DKThomp) April 6, 2016
Of course, if the subset of people who do talk about politics on Twitter were a random subset of the overall population, that would be one thing. But it pretty clearly isn't. The internet user population skews young. The talks-about-politics population skews more educated.
And most obviously, the people shouting on Twitter about the virtues of this candidate and the evils of that one are kind of obnoxious. The reason the arguments seem bitter is that having extended internet arguments with quasi-strangers is something you're only going to do if you're really amped up. Last but by no means least, people who spend their spare time arguing about politics can (and do!) rehearse much more sophisticated and elaborate arguments than normal people.
Sanders supporters and Clinton supporters have similar views
In contrast to casual sample taking on Twitter, statistically valid surveys done by places like the Pew Center consistently show that Sanders supporters and Clinton supporters have similar views on the issues.
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/6302947/overview_2.png)
One can, of course, quibble with these findings. Alternate wordings on questions about trade, for example, tend to produce a picture of a more trade-skeptical public. But for our purposes, Pew's relevant finding is that Clinton supporters and Sanders supporters express a similar opinion.
Another notable fact is that the small differences that do exist do not line up well with the candidates' stated issue positions.
Clinton, for example, has taken pains to emphasize her endorsement by Planned Parenthood and suggested that Sanders's single-minded focus on economic inequality reflects a lack of commitment to the pro-choice cause. But Clinton supporters are somewhat less uniformly pro-choice than Sanders supporters. And Sanders has highlighted his support for creating a huge new federal health insurance program, but his supporters are somewhat less likely to say that ensuring universal health care is a government responsibility.
Sanders supporters don't hate Clinton, and vice versa
On Twitter, Sanders supporters think Hillary Clinton is a monstrous right-winger who spends her days and nights plotting to throw poor kids off welfare benefits while outsourcing their fathers' jobs to China in order to free up scarce public resources for her warmongering. Clinton supporters, by contrast, think Sanders is a Trojan horse fake Democrat sent by the Koch brothers to destroy the Democratic Party.
In polls, things look different. At peak campaign season before either candidate has done anything to try to unify the party, the median Clinton supporter thinks Sanders would be an okay president and the median Sanders supporter thinks Clinton would be an okay president.
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/6303197/Section4_5.png)
There really are embittered haters out there, but they are a minority, and in all likelihood their numbers will shrink when the contest ends.
The real story: demographics
If you want to see a stark divide in the Democratic race, you need to look to demographics. White Democrats feel the Bern to a much greater extent than black ones. And young Democrats are voting for Sanders in overwhelming numbers, even as senior citizens back Clinton by a margin that's almost as large.
If the election pitted an elderly African American against a young white person, it would be common sense to understand the election as primarily about demographic divides rather than policy ones. But since both candidates are old white people, the extent of age and racial polarization is a bit hard to process.
Given Clinton's long association with the centrist Democratic Leadership Council and Sanders's avowed socialist ideology, it's natural to understand the race in primarily ideological terms. And among political obsessives on Twitter, that is how it's understood — providing a nice, easy lens through which the media can watch a titanic ideological struggle.
Survey evidence, however, suggests that the majority of rank-and-file Democrats simply don't see it that way.
