st: RE: kappa command and kapci program

[Garry Anderson <g.anderson@unimelb.edu.au>]

Hi Jen,

The following two references may help with your query about high agreement but low kappa.

Feinstein AR and Cicchetti DV (1990) High agreement but low kappa: I. The problems of two paradoxes. J Clin Epidemiol 43(6) 543-549

Cicchetti DV and Feinstein AR (1990) High agreement but low kappa: II. Resolving the paradoxes, J Clin Epidemiol 43(6) 551-558

Kind regards, Garry


________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Jen Krenz [jkrenz@u.washington.edu]
Sent: Wednesday, 5 March 2014 6:54 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: kappa command and kapci program

Hello,

I am using the kappa command to evaluate the reliability of survey
questions, and have installed the kapci program to compute confidence
intervals for kappa values.  Some of the survey questions seem to have
moderate to high levels of % agreement but have low kappa values.  In
addition, some of the confidence intervals that are generated using the
kapci command do not have lower bounds.  Are there any suggestions?

I am computing unweighted and weighted kappa statistics and confidence
intervals.  Here is sample code (for comparing participant responses to
the same survey question asked at two different times-this was an
ordered categorical question):

kap q_1717 q_1801, tab
kapci q_1717 q_1801, reps(1000)
kapci q_1717 q_1801, w(w2) reps(1000)

Thanks,
Jen

--
Jen Krenz, MS MPH
Department of Environmental and Occupational Health Sciences
University of Washington
206-551-6342
jkrenz@uw.edu

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

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