I have thought about this also. I know these pollsters are experts, but I question what conclusions can be drawn. An additional consideration…if I am shown such a list, I might also rank inflation or the economy high, as that is relevant to my life right now. But it is also true that if a candidate supports abortion bans, I’m done. It’s …
I have thought about this also. I know these pollsters are experts, but I question what conclusions can be drawn. An additional consideration…if I am shown such a list, I might also rank inflation or the economy high, as that is relevant to my life right now. But it is also true that if a candidate supports abortion bans, I’m done. It’s a deal breaker.
Inflation is an issue to which there can be multiple, debatable approaches. Abortion is a right. They’re just not parallel “issues” that belong on the same list, in my mind.
Agree but pollsters are not experts, people don’t always answer honestly and it depends on how the question are phrased and the data base used for the survey. Lots of moving pieces.
I have a hard time finding the original wording of many polls, and when I can there are usually fairly important issues with the way they are phrased and (as you point out) in the contexts and forced choices. Have you ever read something you wrote so that you have absolute knowledge about what you meant by it and then after someone took it the wrong way you realized some small word choice or even punctuation really Did give it a second possible interpretation? Add to that, poor reading comprehension (trait or state) and many polls are badly lacking. Those who construct them may be "professionals" at polling, but there are plenty of professional doctors I'd never see, professional painters I'd never hire... I've become convinced pollsters need to go back to using techniques more like what scientists use building anchored rating scales. They need to construct their polls using small sample pools first and debrief their test subjects for interpretations they may not have intended, to identify biased language and to better cluster issues. Just knowing abortion is a top issue means... what? The respondent cares strongly about it either pro-choice or anti-abortion, and that's not much on which to base any sort of prediction.
Also there is rarely a “why” involved particularly with questions like “is the country going in the wrong direction?” No clue how many answered Yes because Trump lost in 2020 and how many answered Yes because they feel that our democracy is in peril due to the current turn toward authoritarianism or some other factor.
I have thought about this also. I know these pollsters are experts, but I question what conclusions can be drawn. An additional consideration…if I am shown such a list, I might also rank inflation or the economy high, as that is relevant to my life right now. But it is also true that if a candidate supports abortion bans, I’m done. It’s a deal breaker.
Inflation is an issue to which there can be multiple, debatable approaches. Abortion is a right. They’re just not parallel “issues” that belong on the same list, in my mind.
Agree but pollsters are not experts, people don’t always answer honestly and it depends on how the question are phrased and the data base used for the survey. Lots of moving pieces.
I have a hard time finding the original wording of many polls, and when I can there are usually fairly important issues with the way they are phrased and (as you point out) in the contexts and forced choices. Have you ever read something you wrote so that you have absolute knowledge about what you meant by it and then after someone took it the wrong way you realized some small word choice or even punctuation really Did give it a second possible interpretation? Add to that, poor reading comprehension (trait or state) and many polls are badly lacking. Those who construct them may be "professionals" at polling, but there are plenty of professional doctors I'd never see, professional painters I'd never hire... I've become convinced pollsters need to go back to using techniques more like what scientists use building anchored rating scales. They need to construct their polls using small sample pools first and debrief their test subjects for interpretations they may not have intended, to identify biased language and to better cluster issues. Just knowing abortion is a top issue means... what? The respondent cares strongly about it either pro-choice or anti-abortion, and that's not much on which to base any sort of prediction.
Also there is rarely a “why” involved particularly with questions like “is the country going in the wrong direction?” No clue how many answered Yes because Trump lost in 2020 and how many answered Yes because they feel that our democracy is in peril due to the current turn toward authoritarianism or some other factor.