Serious question, my naive impression of this response rate makes me skeptical of most political polling, at least in isolation.
Is that a reasonable interpretation, or are their methods such that you think it accounts well for that?
I think polling vs. election results gives us our answer. Basically "we're + - 10%" for any one poll.
So many variables with various methodologies to me it's amazing when results are within 3% of a poll.
given the drop in response rate it's remarkable that the polls were more accurate in '22 than any other cycle going back to '98.
ofc there's no reason this has to translate to '24 but I'm cautiously optimistic.
That’s been my impression, I don’t know the methods well enough to judge how well the adjustments are working.
I guess no one really does until after the fact anyway.
Most research methodologies are hampered by the fact that people who will participate in research are likely a kind of person.
Personally I value research where participants are paid a bit more for this reason because “people who like or need money” is a larger group than “people who like polls”
Last time one of those market research things sent me a 50+ page survey and a $5CDN bill. I took out my phone, and set a timer for $5 worth of work. Never finished. They called me and were quite irate.
I calmly told them my time is valuable, and if they want me to finish, I need $20 more (minimum).
Of note; I did not agree to do that survey, they sent it to me without my consent.
In the end, I got $5, the research company got nothing, and have never sent me a survey again since. win/win/win!
