3 ms·

I don't think I've misrepresented the outcome of the poll at all. Even after lumping together 18-49 year old respondents, an unusually coarse cohort spanning s

• URSpider94 7y ago
  First, I don't see any table that lumps all responses 18-49. I do see a column for <45. Second, the tables that show responses for all respondents do include values (not N/A) for all age groups, meaning that across the n=1000 poll there are > 125 respondents in each age group, meaning that at minimum 250 people out of the 1000 in the poll are under 50. The categories that do have N/A are in questions that select for only Democratic or Democratic-leaning independents who are registered to vote -- which is only 411 respondents out of the poll. And that one does have a lot of N/A's, including in the < 45 column. Having N < 125 under 45 in a sample size of 411 people is disappointing, but hardly the same as "no responses from people under 50", which is what you wrote. It also doesn't mean that those responses weren't included in the poll, only that they didn't feel confident to break them out separately.