be
shouldn't
liberals,
Two bits says if i could get to the actual raw data and methods, i debunk= that study.
be
shouldn't
liberals,
Two bits says if i could get to the actual raw data and methods, i debunk= that study.
study.
Why don't you determine what the confidence level is with a sample size of 46 people?
that study.
...so narrowly selected.
that study.
Yes, ones with strong political views, out of a random larger bunch.
They were only able to get 23 conservatives into the test group. The rest were too fearful to participate.
screening,
never be
shouldn't
my
=20
liberals,
debunk that study.
Because that is not where the problems involved occur. =20
screening,
never be
shouldn't
pull my
to
the=20
liberals,
debunk that study.
size=20
Naw, there only that many so darned c*ck-sure of themselves out of over =
1000.
that study.
No, that's not it. The answer is... Because we don't know how.
I found this page
and put 46 in the right side and then put 15 in the left side and it came back with Sample size needed: 43
If that startles you, then you figure something out.
screening,
never be
shouldn't
pull my
to
the=20
liberals,
debunk that study.
size=20
It does not startle me at all. Though the sample size for the stated=20 conditions does seem large.
The fundamental error was in the selection of the subsets in the first=20 place. That guarantees invalid results. Face it selecting for 40 to 50=20 extremist position to represent the consensus of over 1000 is exactly=20 what the politicians do. It ain't science anymore, it is politics.
But that was the whole point of the test.
Valid for members of the subsets. Less so for those who were excluded.
See it like a bell curve, with 1000 moderates lumped toward the middle, tapering down to 23 extremists at either end. How is testing the middle 1000 for the differences that set them apart going to reveal anything? They're hardly apart.
that study.
Here's a visual that helps
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.