One-way analysis of variance by ranks (Kruskal-Wallis Test) -----------------------------------------------------------
^kwallis2^ varname [^if^ exp] [^in^ range]^, by(^groupvar^) control(^#^ > )
Description -----------
^kwallis2^ is an extremely useful test for deciding whether k indepedent samples are from different populations. The null hypothesis is that k samples come from the same population or from identical populations with the same median. When a significant value is found, ^kwallis2^ determine which groups ar > e different.
varname refers to the variable recording the outcome and groupvar refers to the variable denoting the population. Note that the ^by()^ "option" is not optiona > l.
With the ^control()^ option, you may specify which group is to be considered as a control group or a standard group. For example, when a researcher wishes to assess the effects of various drugs on behavior.
Example -------
. ^kwallis2 medage, by(region)^ . ^kwallis2 behavior, by(drug) control(1)^
Also see -------- Manual: ^[R] kwallis^ On-line: help for @nptrend@, @oneway@, @runtest@, @signrank@ Book: Siegel S. & Castellan J. "Nonparametric Statistics for the Behavioral Sciences" Chap. 8 (1988) McGraw-Hill