st: Re: Skilmack across group

["Joseph Coveney" <jcoveney@bigplanet.com>]

Jonathan Lam wrote:

I want to compare agreement between k measurements cross groups. I am aware of 
the non-parametric Skillings-Mack test, which can handle repeated measures with
unblanaced observations. However, I don't know how to execute the corss-group 
comparision (invention vs. control) in this test which is similar to what I did
using ANOVA:

For skilmack, this is what I have:
Skilmack hba1c, id(id) repeated(survey)

For repeated measure ANOVA, this is what I have:
Anova hba1c group / id|group survey group#survey, repeated(survey)

Can you show me how to compare cross group with the skilmack test? 

--------------------------------------------------------------------------------

Neither the Friedman test nor the Skillings-Mack test is intended to be used 
with individuals nested within groups, and so you're not going to be able to use
them in a split-plot design like for your repeated-measures ANOVA.

I'm not sure that I understand what it is that you want to do.  Do you want to
compare the magnitude of inter-rater agreement in Group A to the magnitude of
inter-rater agreement in Group B?  That is, are you seeking to use Kendall's 
coefficient of concordance from -friedman- (or some surrogate of it from 
-skilmack-) as a nonparametric index of agreement?  If so, then you might be
able to use -skilmack- on the groups separately in conjunction with a 
resampling method.

Joseph Coveney


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/