SPSS Statistics

SPSS Statistics

Your hub for statistical analysis, data management, and data documentation. Connect, learn, and share with your peers! 

 View Only
Back to discussions
Expand all | Collapse all

Wrong value of Cohen's kappa instead of error message

  • 1.  Wrong value of Cohen's kappa instead of error message

    Posted Thu November 02, 2023 03:31 PM

    Dear SPSS experts,

    I tried to use Cohen's kappa on a 2x2 table in which answers for one method were coded as +/- and for the other method as yes/no (data set with four rows, and frequencies as observation weights). SPSS 29 reported Kappa=0.000 with a standard error of zero. I find this very weird, because I would have expected to either get the correct answer or an error method because of a mismatch between the rater values.

    After changing the yes/no entries to +/-, the correct kappa value was reported.

    What is going on?

    Best regards,
    Ulrike Grömping



    ------------------------------
    Ulrike Grömping
    ------------------------------


  • 2.  RE: Wrong value of Cohen's kappa instead of error message

    Posted Fri November 10, 2023 02:07 AM

    Good morning,

    I am a bit irritated that there is no answer whatsoever regrding an SPSS behavior that I consider a serious bug: customers get a wrong answer without any warning.

    Please confirm that someone is looking into this problem. 

    Best regards, 

    Ulrike Grömping 



    ------------------------------
    Ulrike Grömping
    ------------------------------

    Original Message