If done right, they took representative samples. Many potential things that can go wrong in these processes but the way sampling works makes it totally plausible that you live somewhere and never get asked. Exit polls for elections are also not made up and I’ve never been polled. And if such polls makes sense is yet another topic.
Who is the group here? There are surveys that do not try to prove a point but to survey and get a picture of whatever. I guess you could even find out about the methodology used in the poll from this post and check it.
Like I said there are issues. No survey can completely get the perfect picture but as long as we do not read out the brains of every person on the planet, methodically good social sciences are the best apprpximation we have.
methodically good social sciences are the best apprpximation we have.
Yes and No. In many cases, the methodology is questionable at best and the results become a part of “settled science” which - unless you’re talking specifically to academics and researchers - is highly problematic.
Should we do nothing then? In many cases - yes. Which sort of illustrates the problem in another way - if we had nothing to say we’d just make something up right? Well, instead of doing that if we have nothing to say we run some polls or surveys and find something to say in them. That’s a huge problem. (Admittedly that’s less to do with social science than media, but the Venn diagram of the two is ugly.)
When talking about science I mean science and nothing else. Clear, good methodologies, peer reviews etc. Random street polls etc. are trash and surveys between science and street polls vary across a broad spectrum.
If done right, they took representative samples. Many potential things that can go wrong in these processes but the way sampling works makes it totally plausible that you live somewhere and never get asked. Exit polls for elections are also not made up and I’ve never been polled. And if such polls makes sense is yet another topic.
That’s a pretty large if. Haha
No group has ever cherry picked the sample set to prove their point right?
Who is the group here? There are surveys that do not try to prove a point but to survey and get a picture of whatever. I guess you could even find out about the methodology used in the poll from this post and check it.
All surveys are flawed. Quantitative social science is a rats-nest of scams, schemes, and popularity contests. If you like that kind of thing.
Like I said there are issues. No survey can completely get the perfect picture but as long as we do not read out the brains of every person on the planet, methodically good social sciences are the best apprpximation we have.
Yes and No. In many cases, the methodology is questionable at best and the results become a part of “settled science” which - unless you’re talking specifically to academics and researchers - is highly problematic.
Should we do nothing then? In many cases - yes. Which sort of illustrates the problem in another way - if we had nothing to say we’d just make something up right? Well, instead of doing that if we have nothing to say we run some polls or surveys and find something to say in them. That’s a huge problem. (Admittedly that’s less to do with social science than media, but the Venn diagram of the two is ugly.)
When talking about science I mean science and nothing else. Clear, good methodologies, peer reviews etc. Random street polls etc. are trash and surveys between science and street polls vary across a broad spectrum.