4 ms·
This is quite interesting. One question that springs to mind, though: who exactly are the respondents thinking of when they're asked to estimate? Do they hear "
by wool_gather 7y ago
This is quite interesting. One question that springs to mind, though: who exactly are the respondents thinking of when they're asked to estimate? Do they hear "Are (Democrats|Republicans) $SOME_TRAIT?" and think "Is (Nancy Pelosi|Mitch McConnell) $SOME_TRAIT?", or "Is my kid's study buddy's (Democrat|Republican) parent $SOME_TRAIT?" There's always a big difference in how you judge someone you know personally as opposed to a face on the news. This doesn't seem to have been asked, or at least it's not mentioned in the article.
- jerf 7y agoI was thinking that probably played a role as well. The extreme partisans on one side will be exposed to the opinions of the extreme partisans on the other, and ignoring the namecalling for a moment, it would be interesting to see if the extremists are accurate about what the other side's extremists believe. I don't think that would change very much about this report overall; I'm sure both extremists would still overestimate the size of the extremists groups on both sides quite significantly. But it would be an interesting data point.