On Wed, 18 Jun 1997 14:15:48 GMT, Hans Kingma <[hidden email].NL_>
wrote: >Hi, I'm trying to calculate a Cohens Kappa on two observers. The problem is >that they didn't used all the values available so the crosstab isn't >square. I try to specify an integer mode but didn't succeed: the missing >columns and rows aren't displayed and so isn't the Kappa. > >Someone got a clue? > >greetings >hans > >PS. the command I use is: > >CROSSTABS VARIABLES=fcomiscr (1,10) hcomiscr (1,10) > /TABLES=fcomiscr BY hcomiscr > /FORMAT= AVALUE TABLES > /STATISTIC=CORR KAPPA > /CELLS= COUNT . > >-- >Please remove the last _ from my reply to address when replying This link is no longer working. http://www.spss.com/tech/macros/Nskappa.html. Could somebody help with the computation of Kappa when the crosstab is not squared. Thanks Placide ===================== 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 calculate Kappa when the crosstab is not squared, you can use a macro
I wrote in 2006. This Macro and some notes and references can be found on Levesque's site: http://www.spsstools.net/SampleSyntax.htm Please scroll to the header "Matrix" . Best Regard, Pieter van Groenestijn RadboudUniversiteit Nijmegen Faculteit Sociale Wetenschappen Research Technische OndersteuningsGroep Thomas van Aquinostraat 4.00.51 tel: 024-3612035 fax: 024-3612351 email: [hidden email] hp: http://www.ru.nl/fsw/rtog/ Placide wrote: > On Wed, 18 Jun 1997 14:15:48 GMT, Hans Kingma <[hidden email].NL_> > wrote: > > >> Hi, I'm trying to calculate a Cohens Kappa on two observers. The problem is >> that they didn't used all the values available so the crosstab isn't >> square. I try to specify an integer mode but didn't succeed: the missing >> columns and rows aren't displayed and so isn't the Kappa. >> >> Someone got a clue? >> >> greetings >> hans >> >> PS. the command I use is: >> >> CROSSTABS VARIABLES=fcomiscr (1,10) hcomiscr (1,10) >> /TABLES=fcomiscr BY hcomiscr >> /FORMAT= AVALUE TABLES >> /STATISTIC=CORR KAPPA >> /CELLS= COUNT . >> >> -- >> Please remove the last _ from my reply to address when replying >> > > This link is no longer working. > http://www.spss.com/tech/macros/Nskappa.html. > Could somebody help with the computation of Kappa when the crosstab is not > squared. > Thanks > Placide > > ===================== > 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 |
In reply to this post by Placide-2
Hi Placide, Apologies for the moving links. The link http://support.spss.com/ProductsExt/SPSS/Documentation/Statistics/index.html has links to the macros available from our macro library, including the nskappa macro. David Nichols SPSS, an IBM Company
On Wed, 18 Jun 1997 14:15:48 GMT, Hans Kingma <[hidden email].NL_> wrote: >Hi, I'm trying to calculate a Cohens Kappa on two observers. The problem is >that they didn't used all the values available so the crosstab isn't >square. I try to specify an integer mode but didn't succeed: the missing >columns and rows aren't displayed and so isn't the Kappa. > >Someone got a clue? > >greetings >hans > >PS. the command I use is: > >CROSSTABS VARIABLES=fcomiscr (1,10) hcomiscr (1,10) > /TABLES=fcomiscr BY hcomiscr > /FORMAT= AVALUE TABLES > /STATISTIC=CORR KAPPA > /CELLS= COUNT . > >-- >Please remove the last _ from my reply to address when replying This link is no longer working. http://www.spss.com/tech/macros/Nskappa.html. Could somebody help with the computation of Kappa when the crosstab is not squared. Thanks Placide ===================== 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 |
Administrator
|
David, is a macro necessary? In the two-rater case, one can simply use the integer mode for CROSSTABS, can they not? Here's an example using data from http://www.john-uebersax.com/stat/saskappa.htm#ex1.
data list list /r1 r2 r3 (3f2.0). begin data 1 2 1 1 2 1 1 2 1 1 2 2 1 3 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2 3 3 2 3 2 2 3 3 2 3 3 1 3 3 2 3 3 2 end data. crosstabs r1 by r2 / stat = kappa. * Table is not square, so kappa is not computed. * Now use INTEGER mode to ensure table is square . crosstabs VARIABLES = r1 r2(1,3)/ table = r1 by r2 / stat = kappa. Cheers, Bruce
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. |
Hello,
the question of a non-square table struggles me at the moment, and I was happy that recently an answer was given on the list. Great! But to my disappointment Bruce Weaver's solution does not work. Running the syntax with data etc as provided leads in both cases with this message: "Kappa statistics cannot be computed.They require a symmetric 2-way table in which the values of the first variable match the values of the second variable". It seems as if the integer mode does not work. Any ideas what could be the point I'm missing? I'm using SPSS for windows release 11.0.1. Thanks Harald Bailer -----Ursprüngliche Nachricht----- Von: SPSSX(r) Discussion [mailto:[hidden email]] Im Auftrag von Bruce Weaver Gesendet: Dienstag, 2. Februar 2010 18:17 An: [hidden email] Betreff: Re: Kappa, missing rows/colums David, is a macro necessary? In the two-rater case, one can simply use the integer mode for CROSSTABS, can they not? Here's an example using data from http://www.john-uebersax.com/stat/saskappa.htm#ex1. data list list /r1 r2 r3 (3f2.0). begin data 1 2 1 1 2 1 1 2 1 1 2 2 1 3 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2 3 3 2 3 2 2 3 3 2 3 3 1 3 3 2 3 3 2 end data. crosstabs r1 by r2 / stat = kappa. * Table is not square, so kappa is not computed. * Now use INTEGER mode to ensure table is square . crosstabs VARIABLES = r1 r2(1,3)/ table = r1 by r2 / stat = kappa. Cheers, Bruce ===================== 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 |
Free forum by Nabble | Edit this page |