| statistics-ratio of variances In looking through some of my statistics notes (deciding whether to use pooled or unpooled formula for variances) I have that when the variance of sample 1/variance sample 2 is less than or equal to 3 then we can treat our variances as similar. I don't really understand where this statement comes from-can you direct me to a resource that will give me a clue as to why this is so? |