XPost: rec.arts.sf.written, rec.arts.sf.science   
   From: b.scott@csuohio.edu   
      
   On Thu, 23 Jan 2014 19:31:31 -0700, Chrysi Cat   
    wrote in   
    in   
   rec.arts.sf.written,rec.arts.sf.science,rec.arts.sf.misc:   
      
   [...]   
      
   > Statistics work. If you couldn't extrapolate from a proper   
   > sample (which would probably be closer to 500 or 5000 for   
   > that 50 mil), then we'd pretty much be up a creek for   
   > knowing what people do and think...   
      
   Actually, the margin of error depends almost entirely on the   
   absolute sample size; in real-life settings there is almost   
   no dependence on the ratio between population and and sample   
   sizes.  A sample of 50 out of 50 million is virtually as   
   good as a sample of 50 out of 50,000.  It *is* a bit too   
   small; I’d want a few hundred, and a sample of 1000 would be   
   good.   
      
   Brian   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)