On 10 Apr 2002 12:40:56 -0700, [EMAIL PROTECTED] (JC) wrote: > Does anyone have a suggestion on what procedure to use in order to > report effect size from a Mann-Whitney?
Effects are average ranks? (or can be expressed that way) - so you have the difference of the average ranks. Mention the SD of the ranks. The resulting numbers are pretty unappealing unless you start with continuous variables and a pretty good difference between groups that are fairly small. But then, you probably shouldn't be leaning on rank-statistics unless that is what you have.... -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
