Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Kappa and test of significance


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Kappa and test of significance
Date   Thu, 3 May 2012 09:34:17 +0100

The title of your post although not the text alludes to -kappa- but
that seems quite wrong here. The technology of -concord- (-findit- for
locations) seems a bit better fit but much of it was developed with
measured data, not counted data, in mind.

Poisson modelling would seem a more natural starting point. A minimal
descriptive approach includes plots of A vs B, ln A vs ln B, (ln A -
ln B) vs (ln A + ln B)/2. The last two require all counts to be
positive.

Nick

On Thu, May 3, 2012 at 1:56 AM, Aminu Shittu <ameen_vet@yahoo.com> wrote:

> I have a data showing number of parasite counts from composite animal fecal samples from 23 farms. Two individuals (A & B) analysed each sample independently and reported different egg counts presented in the data set as vars acomp, bcomp, and totalcomp = acomp + bcomp. I was wondering whether I could test for agreement and associations between the 2 observers. I will appreciate any help on commands to be used, being a non-proficient Stata user. I use Stata IC v. 10.

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


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index