Banya, Winston
>
> I am working on a small dataset and most of the variables are skewed. In
> this data 2 'observers' give an opinion and in some instances
> they do agree.
> I want to find out the difference in response when they both agree on a
> variable.
>
> For example in this dataset the age of the people was given, and the two
> observers are asked to state a preferred method of treating the subjects
> and I want to find out using non parametric methods if there is any
> difference in age when the 2 observers choose the same method of
> treatment.
> That is if the method is 'A' for example, I am interested in any
> difference
> when they choose method A. Note that in some cases they both
> agree on A and
> in some cases they do not and so the groups cannot be mutually
> exclusive. Is
> there a way of handling this in STATA such that if observer 1 says yes 15
> times and no 12 times, while observer 2 says yes 18 times and no 9 times
> then I can only compare when they said yes. However I want to treat the
> responses as if there were no times when they said ties together
> even though
> they did.
>
This would, I guess, be easier to answer if you gave examples of
the data structure you are using or intending to use.
What are your variables? What defines your observations?
However, observations can be identified as agreeing on
two variables by something like
... if var1 == "yes" & var2 == "yes"
or
... if var1 == 1 & var2 == 1
Nick
n.j.cox@durham.ac.uk
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/