Quote:
Originally Posted by RogueGypsy
Oh and what would ever make you think 2000/300000000 or .000006% is in anyway a good cross section of the nation. 
That's like saying if a hair falls out of your head, the rest of you is gonna die now too. Better lie down and wait for it. |
Unless you've done the calculations to determine how sample size affects study power with respect to this data, you don't know what you're talking about.
Assuming that there isn't too much variability in the data, and that there wasn't a lot of bias in the sample selection process, 2000 might be more than adequate to characterize the experiences of a significant portion of the population. It's likely a significantly larger sample size than the sample size generated by your own personal experiences.