I'f go as far as to say almost mandatory to treat them as two pairwise comparisons. There may be a case to use the independent t test. There may also be a case to still use two-tailed tests. The only caveats are i)whether you really are interested in only those two tests and only results in one direction - one couldn't say without knowing the context, and ii) whether or not a specific contrast/trend test might be more appropriate.
Thom cblack wrote: > > I am trying to compare 3 groups of people, each group exposed to a different > situation. Data consists of question responses giving ratings from 1 to 7. > I am only interested in one-tailed comparisons between group 1 and group 2 > and between group 1 and group 3. Kruskal-Wallis seems to waste power by > doing a 2-tailed test and considering all possible pairs of groups. > What else can I use? Is it appropriate to treat the comparisons as two > independent Mann-Whitney tests? > > Thanks...if possible, please copy responses in e-mail to: > cblack (at) unix (dot) tamu (dot) edu . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
