[Winstonm]

Does anyone actually think presidential polling is accuarate or even remotely accurate? I don't remember the state now, but in 2020 Biden was shown in the polls as leading by 9 points--he won that state by 0.5. That's well outside any margin of error, more like inside the margin of our best guess is....

[kenberg]

Winstonm, on 2024-September-25, 09:56, said:

Does anyone actually think presidential polling is accuarate or even remotely accurate? I don't remember the state now, but in 2020 Biden was shown in the polls as leading by 9 points--he won that state by 0.5. That's well outside any margin of error, more like inside the margin of our best guess is....

I just thought of an experiment. We could do a poll to find out how many people are going to vote. Not a poll about who to vote for, just a poll to see if they will vote. Would such a poll be a reasonably accurste predictor of turnout? People do not always do what they say they will do. I planned to go to the next meeting of the library book club but then Necky noticed they meet on the day a grandchild is having her birthday party so we will go to that instead.

I have voted in every presidential election starting in 1960, I think it is important to do so, so It's a good bet that I will be voting this November. But not everyone feels that way. I gather that around 2/3 of eligible voters actually vote. I doubt that this means that the population divides into 2/3 that al;ways vote and 1/3 that never vote. Probably there is a sizable fraction, maybe 1/5, who vote sometimes, don't vote other times, with no particular reason as to when they do and when they don't.


If it is difficult to predict how many will actually show up to vote then it is certainly difficult to predict how the vote will go.

Polls have their uses, so I believe, but caution seems sensible.