Wednesday, February 24, 2016

The Post's semantic analysis of Sanders supporters on Twitter is a good step forward

Rebekah Tromble and Dirk Hovy at The Washington Post have finally done the obvious thing and conducted a semantic analysis of how Sanders supporters talk about Hillary Clinton on Twitter. This is a dramatic step past the lazy, anecdote-driven punditry that's dominated the discussion for months; but of course, there are clear reasons why the people who've spammed the internet with Bernie Bro articles didn't want to look at the issue with any rigor:
...of a total 52,181 tweets mentioning @HillaryClinton, just 606, or 1.16 percent, contained [gendered slurs]...The vast majority of the slurs were associated with Twitter users on the right — particularly self-identified Trump supporters. But 14.7 percent came from those backing Sanders...That is a mere 0.17 percent of all the tweets mentioning @HillaryClinton that we examined....our analysis does provide a better understanding of the extent and character of the attacks lodged against Hillary Clinton online. And it seems relatively little abuse originates from the left.
This is of course exactly what you would expect. It is actually much smaller than my ballpark estimate of around 4%. While women running for office may face institutional and systematic sexism, popular sexism is not presenting them with a significant challenge - and in fact among Democrats popular biases favor women.

Presumably, Clintonites are going to meet this study with three responses:

  • Misrepresentation (EG quoting caveats like "we do find some evidence of Bernie Bros’ bad behavior" as if they are final conclusions)
  • Methodological quibbling (EG complaints about how various words should have been flagged as "negative"); and
  • Silence
Suffice to say, whatever grievance you have with the study, it's an astronomical step forward from the ridiculous parade of anecdote and innuendo that preceded it. If Sanders critics want to insist that their perspective on the question is more credible than what's been supplied here, they should present their own study, and then we can compare them. Otherwise, this should - although it certainly will not - put the matter to rest.