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

RE: st: Kappa programming challenge...


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Kappa programming challenge...
Date   Thu, 20 May 2004 11:58:05 +0100

I'd happily interpret continuous 
here as including any discrete 
scale which you are willing to 
treat as approximately continuous. 

Even with a 7 point ordered scale, 
it may well work much better than you 
fear. It is difficult to add rigour
to that, but many useful procedures
depend on treating ordered scales 
literally (meaning numerically) 
through scores, whatever the 
measurement purists say. 

Nick 
n.j.cox@durham.ac.uk 

Joseph Coveney
> Sent: 20 May 2004 02:07
> To: Statalist
> Subject: Re: st: Kappa programming challenge...
> 
> 
> I'm not familiar with Lin's coefficient of concordance.  The 
> help file for 
> -concord- states that Lin's coefficient is for continuous variables.  
> 
> The command does allow frequency weights, which I suppose 
> would warrant 
> inference that it's suitable for use with 
> limited-dependent-value data, such as 
> rating scales.  
> 
> Kendall's coefficient of concordance has always sprung to 
> mind in the past as 
> the alternative to Cohen's weighted kappa coefficient in 
> situations of inter-
> rater agreement with ordered categorical variables.  
> 
> Should I switch--is there a reason to prefer Lin's 
> coefficient in these kinds 
> of situations?  
> 
> The reason that I'm asking is that I couldn't find any 
> guidance on this in its 
> description in the _Stata Technical Bulletin_.
> 
> Joseph Coveney
> 
> --------------------------------------------------------------
> -----------------
> 
> Ronan Conroy wrote:
> z
> > stephen.kay@adelphigroup.com wrote:
> 
> > All ratings are on a 1 to 7 scale with end anchors along
> > the lines of 'rubbish' and 'brilliant'.
> 
> Since the scale is ordered, what about Lin's concordance 
> coefficient? I
> prefer it, somehow, to weighting kappa.
> 
> --------------------------------------------------------------
> ------------------
> 
> 
> 
> *
> *   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/
> 

*
*   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