|
Pathfinder,
That is funny but it is also interesting. If you know nothing about cats (you are an alien visiting earth), then the answer is clearly no. The minimum number of subjects required in an all or nothing response for statistical significance is about 7 patients, even if for example every cat meowed and every dog barked (or whatever).
Your post sounds funny because you already know cats meow. Your sample size is huge. Every cat you've ever seen.
---------------------------------
So anyway, say you have a group of 7 people in a wonder-drug study and 7 in a placebo group (control group). All 14 are terminally ill and are expected to die within a month. A month goes by and the 7 given the wonder-drug live and the 7 given a placebo die. This is the minimal study size and control group necessary to be "statistically significant". If one of those 7 that lived died or one of the ones that died lived, it would actually not be significant yet. You would need more data.
__________________
skype = "adultdatelink"
|