It also depends on the context and the purpose of the survey. With a sample size of n=1000, you can get a general sense of the population’s opinions or characteristics if the sample is representative of the population and the survey is well-designed. However, larger sample sizes may be needed if you want to make inferences with more precision or if your population is very diverse, or if you want to drill into sub-segments within the population that may be niche and hard to reach (e.g., minority ethnic groups).
In that article. I am saying religious groups have niche minorities similar to minority ethnic groups.
The survey isn’t about the “niche” subset of the population that is religious. It is about the composition of the entire population. Not a subset of the population so that isn’t relevant.
Edit: to be clear, I understand that there may be some niche subsets within this survey that may not be represented because there are only 20 people in the US that believe in that weird religion, but again, that has nothing to do with the larger, non-niche subsets which are absolutely represented with enough accuracy to draw statically significant conclusions with a sample size of 1000
I said for religious selection 3,300 is too small a sample size. Their previous studies show they have much larger sample sizes for religious diversity. There’s over 4k religions in the world and while those aren’t all represented in the US I think the sample size should exceed the possible variety.
Statistical analysis 101. You don’t really get to have you own opinion.
Okay I’ll go off facts instead: Pew’s Religious Landscape Study in 2014/2017 says they surveyed 35,000 Americans.
https://www.pewresearch.org/religion/religious-landscape-study/
https://www.mili.eu/learn/is-a-sample-size-of-n-1000-sufficient-for-accurate-survey-results
In that article. I am saying religious groups have niche minorities similar to minority ethnic groups.
The survey isn’t about the “niche” subset of the population that is religious. It is about the composition of the entire population. Not a subset of the population so that isn’t relevant.
Edit: to be clear, I understand that there may be some niche subsets within this survey that may not be represented because there are only 20 people in the US that believe in that weird religion, but again, that has nothing to do with the larger, non-niche subsets which are absolutely represented with enough accuracy to draw statically significant conclusions with a sample size of 1000
What’s the fact here? That this study had more than ten times the sample size?
That doesn’t make 3,300 an insufficient sample size. It just means this study had a larger sample size.
And yeah, that study may have different numbers, but it is based on decade-old polling, so of course it does.
I said for religious selection 3,300 is too small a sample size. Their previous studies show they have much larger sample sizes for religious diversity. There’s over 4k religions in the world and while those aren’t all represented in the US I think the sample size should exceed the possible variety.