from my escort survey, self-reported age and estimated monthly earnings.
sample size kinda small, each bin is n=~40
9
5
134
Conversation
Replying to
As someone with a degree in economics and data science, 40 is TOO small a sample size for the data you are trying to measure considering how large of potential sample pool there actually is (ie- millions of SWers)
5
7
You have to calculate what the relevant sample size for your population is in regards to what you are trying to measure. Know how to calculate variance. Know what that “variance” even means. What’s her calculated variance for the data regarding age range and monthly income?
1
3
The vast majority of research is conducted with non-representative convenience samples (like this). Real pop variance is unknown. Reducing your margin of error means little if you begin with a nonrepresentative sample.
3
4
So you could increase your sample from 40 to 4000 and there would be no guarantee that it would be any more representative. In fact given nonrepresentative sampling to begin with you could even expect it to be worse, less representative.
1
1
I’ll just keep saying it again…you need to widen the demographics of the sample pool. There’s only a very small portion of SWers someone is reaching in Twitter. Try polling providers who ONLY use Tryst or P411 or PD or OF or TER or Silxa. Most SWers aren’t marketing on Twitter.
1
1
i used all my platforms to share my survey with SWers, not just twitter.
You can look at the raw data here
