In article <[EMAIL PROTECTED]>, James Choe <[EMAIL PROTECTED]> wrote:
>I am trying to show that precipitation level in 6 selected states are >statistically different. Is Kruskal-Wallis test appropriate for this? >My data consists of annual average precipitation county level. If all you have is one number for each county (the average annual precipitation), there is no way of doing any sort of significance test. You would need data for many years, and would have to account for temporal autocorrelation. Testing whether precipitation differs from state to state is pointliess in any case. We KNOW that precipitation differs from place to place, and hence from state to state. A test of a directional hypothesis (eg, does Wyoming have more precipation than Colorado?) would make more sense, in that at least we don't already know the answer, but it is difficult to see why anyone would be interested. State boundaries have no ecological significance. Radford Neal ---------------------------------------------------------------------------- Radford M. Neal [EMAIL PROTECTED] Dept. of Statistics and Dept. of Computer Science [EMAIL PROTECTED] University of Toronto http://www.cs.utoronto.ca/~radford ---------------------------------------------------------------------------- . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
