John M Price wrote: > omitted > Thus, if > : there is an apparent causal relationship at the p<100 level, that means > : the liklihood that the results are a fluke are less than 1 in 100. > > BTW the proper notation would be p < .01. And it does not really talk of > by chance, rather, given the gamma function of your particular statistic's > dfs, it tells you where its value lies in the function of all possible > values of the same statistical test with the same dfs. > L OOO OOO K K L O O O O K K L 0 O O O KK L O O O O K K LLLLL OOO OOO K K [280 lines left ... full text available at <url:http://www.reference.com/cgi-bin/pn/go?choice=message&table=05_1997&mid=2952293&hilit=HYPNOSIS> ] -------------------------------- Article-ID: 05_1997&2847477 Score: 78 Subject: REPOST: The Training Pts 1 & 2