Kappa for more than 2 raters

[Placide-2]

Dear All,

I would like to know if SPSS provide a macro for computing Kappa for multiple raters (more than 2 raters) .

Thank you for your help

Best regards

Placide

[bdates]

Cohen’s kappa was built for two raters only.  That’s why SPSS only has that option.  There are several expansions by other authors.  Light expanded Cohen’s kappa by using the average kappa for all rater pairs.  Davies and Fleiss used the average Pe for all rater pairs rather than the average kappa.  This makes somewhat more mathematical sense because Pa was never the issue.  Berry and Mielke approached a multirater kappa from a disagreement standpoint and for the most part duplicated Davies and Fleiss.  The only way to get a multirater kappa from SPSS is to compute kappa for each rater pair and then average as Light did.  Otherwise, syntax is necessary.

 

Brian

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of placide poba-nzaou
Sent: Friday, February 05, 2010 5:27 PM
To: [hidden email]
Subject: Kappa for more than 2 raters

 

Dear All,

I would like to know if SPSS provide a macro for computing Kappa for multiple raters (more than 2 raters) .

Thank you for your help

Best regards

Placide

 

[Michael Kruger]

Here's link for downloading from SPSS support:

http://support.spss.com/Tech/default.asp

If you need I can send it to you as an attachment with doc file...

M. Kruger
>

=====================
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

[E. Bernardo]

Dear M. Kruger

 The link is protected by a password. Hope you send the file offlist?

 

Thank you.

Eins

--- On Fri, 2/5/10, Michael Kruger <[hidden email]> wrote:

From: Michael Kruger <[hidden email]>
Subject: Re: Kappa for more than 2 raters
To: [hidden email]
Date: Friday, 5 February, 2010, 10:47 PM

Here's link for downloading from SPSS support:

http://support.spss.com/Tech/default.asp

If you need I can send it to you as an attachment with doc file...

M. Kruger
>

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

---

Surf faster.
Internet Explorer 8 optmized for Yahoo! auto launches 2 of your favorite pages everytime you open your browser.Get IE8 here! (It's free)

[Bruce Weaver]

Placide-2 wrote

Dear All,
I would like to know if SPSS provide a macro for computing Kappa for multiple raters (more than 2 raters) .
Thank you for your help
Best regards
Placide

The default intraclass correlation computed by SPSS (two-way mixed, single measures, consistency) is equivalent to a weighted kappa (with quadratic weights).  See Norman & Streiner's "Biostatistics: The Bare Essentials" if you need a reference to convince someone (look near the end of the chapter on repeated measures ANOVA).  Norman & Streiner's comment is that they may compute an ICC, but report it as a kappa if that's what the reviewers or editor want to see.  

http://books.google.ca/books?id=8rkqWafdpuoC&printsec=frontcover&dq=Biostatistics+Bare+Essentials&client=firefox-a&cd=1