Posted by
bdates on
Nov 16, 2011; 4:15pm
URL: http://spssx-discussion.165.s1.nabble.com/Fleiss-Kappa-Macro-for-SPSS-tp4994433p4998307.html
Kristina,
Ordered categories contain ordinal data rather than nominal data, so
there is a natural rank among the categories.
Brian
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of
Kristina Georgiou
Sent: Wednesday, November 16, 2011 11:03 AM
To:
[hidden email]
Subject: Re: Fleiss Kappa Macro for SPSS
There are 9 categories which 3 raters(judges) can choose from, and each
rater
has rated over 200 different objects. I am not sure what you mean by the
categories being 'ordered'.
------------------------------------------
Kristina, I don't think you've said how many categories are available
for
raters. If the categories are ordered, or if there are only two
categories,
you could compute the the most common form of intraclass correlation,
which
is equivalent to weighted kappa (with quadratic weights). The following
is
from a newsgroup post I made (to comp.soft-sys.stat.spss) back in 2002.
--- start of old post ---
The following is from Norman & Streiner (2000), Biostatistics: The Bare
Essentials:
------------------------------------------------------
RELATION BETWEEN KAPPA AND THE INTRACLASS CORRELATION
One reason why Cicchetti was fighting a losing battle is
that the weighted kappa using quadratic weights has a
very general property--it is mathematically (i.e.,
exactly) equal to the ICC. (p. 221)
------------------------------------------------------
There are several varieties of ICC. The one Norman & Streiner are
describing is this one (p. 100 in their book, 2nd Ed):
MS_subj - MS_residual
ICC = ----------------------------
MS(subj) + (k-1)*MS_residual
--- end of old post ---
HTH.
-----
Kristina Georgiou
Research Student
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Fleiss-Kappa-Macro-for-SPSS-tp4994433p4998187.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD