A new poll shows President Joe Biden leading Trump 44% to 37%, with Kennedy notching 16%.
Released by Marist in partnership with NPR and PBS Newshour on Tuesday, the poll shows a five-point drop among Democrats for Biden with Kennedy in the race. Meanwhile, the survey indicates a 10-point drop among Republicans for Trump with RFK Jr. on the ticket.
Again, national polls are meaningless since we don’t run national elections.
They give a general idea of how the public will vote when it comes to spoiler candidates. No, they don’t consider regional differences, but I wouldn’t say they are meaningless.
Not saying polls don’t have their worth, but you can make polls say basically whatever you want.
“poll of 2,000 people of varying ages, genders, backgrounds” when they stood out in front of a music theater to get opinions on modern rap music. The results are biased because of how they collected them. Yeah, technically everyone coming out of that theater fits your “different ages sexes” and so on, but they’re all going to have strong opinions based on why they have gathered in common interest.
That’s not how polling works.
Small sample size national polls are always the first line of polling.
They are not meaningless, even if they don’t have the same precision as exit polling.
What I mean is, thanks to the electoral college, running a national poll as though it means anything is pointless.
We saw this in 2016 with Clinton. National polling showed her winning, and as far as the popular vote was concerned, she won.
Which means jack all in the electoral college.
Polling showed Clinton as being most likely to win. The fact that she didn’t win doesn’t mean the polling is necessarily meaningless. Even if someone has a 90% chance of winning, it means they can not only lose, but 1 in 10 times you expect them to lose.
Seems to me that your mistake is that you believe the purpose of polls is to predict an outcome, and/or tell you who is “winning” or “losing” at a given point in time. That is not their purpose.
Their purpose is to gauge the relative effectiveness of different campaign messaging strategies, and to give a rough order of magnitude of a campaign’s trajectory.
Here’s the most important part: polls contain no actionable data for voters. They shouldn’t influence whether or how much you volunteer or donate, and they absolutely must not influence how you vote.
So true! Also, even if this data is 100% accurate, that means 16% of people prefer him to Trump (or Biden). But come election day, the one thing Republicans can be counted on to do is to check the box next to the ‘R’ candidate, no matter what.
It will be interesting to see what effect he has on turnout, if any…
