Quote:
Originally Posted by sheephunter
Not that I know alot about stats but I think you might be calculating the error margain for an opinion poll as opposed to an information survey and I don't think that number you cite takes into account that fact that the survey was conducted for multiple years rather than one.....but I could be mistaken.
Actually just ran the numbers through a couple confidence level calculators and to acheive 99% certainty with a 1% margain of error, you need to survey 16,753 people if you include all resident Wildlife Certificate buyers. Most polls only run a 95% certainty so these numbers are pretty restrictive. I'd guess that is pretty close to the sample size that SRDhas; perhaps it's a bit lower. That didn't factor in multiple year results either. Something to consider when looking at spending large amounts of money on a mandatory survey.
|
16,753 is the aggregate. You need to apply a distribution factor (perhaps tags allocated per WMU for draws at least) to determine how many people are required per WMU for that kind of accuracy. That would be ideal anyway.