Nitin Jain, MBBS, MSPH
> Thank you for the response. But the concord and other
> tests contained in it are only for 2 observers/raters.
> I am looking for something that calculates
> interobserver agreement for a continuous variable WHEN
> THERE ARE MORE THAN 3 RATERS AND THE number of raters
> vary for each subject ?
That wasn't clear to me.
As I see it, you may be torn
between two conflicting aims here: you
will want to try to epitomise the level
of agreement in a few numbers, but there is
likely to be a lot of fine structure here
which will be missed or even misrepresented by
any summary measures.
As Alvine Bissery has suggested, some kind
of anova with observers as factors seems
possible. Alternatively, consider principal components
on the largest complete subset of values.
And don't omit trying a scatter plot matrix,
for which you a need a data structure in which
observers are variables.
It is also possible that -concord- retains
some use if you think in terms of producing
one or more matrixes of measures. The
Stata construct for doing that could be
two nested -foreach- loops, analogous to
a detailed example in Stata Journal 2(2), 202-222 (2002).
Nick
[email protected]
*
* 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/
