Date: Tue, 15 Jul 2003 11:28:19 +1000
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Paul Dickson <firstname.lastname@example.org>
Subject: Re: Cohen's Kappa for multiple judges
I have not used this but it may be of some use, and you can use it independently
of SPSS. I have however, found other aspects of the website quite useful
> Simon Freidin <email@example.com> wrote:
> Multi-rater Kappa statistics in SPSS
> Current SPSS procedures offer only Cohen's original Kappa for two
> (this is provided in the CROSSTABS procedure;
> Analyze->Descriptives->Crosstabs in the menus). There is a macro
> (mkappasc.sps) available from SPSS that computes one version of a
> multi-rater Kappa statistic. It is available from the SPSS web site
> Please make sure that you read the readme file with instructions on
> how to
> use the macro.
> The mkappasc.sps macro computes the multi-rater Kappa measure
> discussed in
> S. Siegel and J. N. Castellan's Nonparametric Statistics for the
> Sciences, 2nd edition (1988, McGraw-Hill). Results include estimated
> a standard error under the null hypothesis and a test of the null
> hypothesis of a 0 population value. Note that there are multiple
> definitions of agreement and chance agreement. This measure focuses
> agreement between pairs of raters. It does not produce the same value
> Cohen's original Kappa for the case of two raters. See Siegel and
> and the references cited therein for further information.
> See also:
> At 03:13 PM 7/14/2003 -0400, you wrote:
> >The average of all possible pair-wise Kappas. See Conger (1980),
> >Bulletin, 88, 322-328. I've seen SAS macros but nothing for SPSS.
> >-----Original Message-----
> >From: Stephen Cox [mailto:firstname.lastname@example.org]
> >Sent: Sunday, July 13, 2003 8:15 PM
> >To: SPSSX-L@LISTSERV.UGA.EDU
> >Subject: Cohen's Kappa for multiple judges
> >Folks - Does anyone know the computational formula for calculating
> >modified Kappa in the case of multiple judges (7 in this case)
> >objects into a nominal scale (6 categories - included/excluded)?
> >Is there a macro for this - I haven't been able to find one.
> >Thanks all.
> >Stephen Cox
> >Dr. Stephen Cox
> >Faculty of Business Research Advisor
> >Queensland University of Technology
> >Room: 1st Floor, 126 Margaret Street
> >Ph: 3864 9371
> >email: email@example.com
> >CRICOS No. 00213J
> >-- Even though this E-Mail has been scanned and found clean of
> >-- known viruses, OPM can not guarantee this message is virus free.
> >-- This message was automatically generated.
> Research Database Manager and Analyst
> Melbourne Institute of Applied Economic and Social Research
> The University of Melbourne
> MELBOURNE VIC 3010
> Tel: (03) 8344 3601 Fax: (03) 8344 5630