Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: Interobserver agreement for continuous variables


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: Interobserver agreement for continuous variables
Date   Fri, 18 Oct 2002 09:48:59 +0100

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 
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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index