Quote:
|
Originally Posted by sapiens
In a given statistical test, the p value is the probability that an observed difference between groups is due to chance, rather than real differences between the two groups. Some fields adopt more stringent cutoffs than .05.
Independent of the studies cited, it's generally the other way around. The p value of the test must be lower than .05 "to hold water".
|
You're right in both cases; I work in the social sciences, so .05 is often the standard for us. I was curious what was standard in the studies she was looking at, if that was known.
On the second part, yep you got me there, I was flipping it in my head (thank you, last summer's statistics course)
