Why ? Basically because the ANOVA testing suggests that there is no significant mean difference betwen the samples given the specific factor tested; i.e., they probabably come from the same population distribution as far as that factor is accounted for.
So it doesn't make sense to ask which sample mean is greater if they're not that different.
Now there are those who would dispute this interpretation, but then that's one reason why we have a variety of multiple comparison tests (probably what you're really asking...)
========================================================
Now I have a related question. Are there multiple comparison tests for time series data ? Do we basically test with a form of repeated measures, with, say, the covariance modeled by an AR structure, and then compare means ?
Happy Holidays,
Al
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now
