SPSS Statistics

SPSS Statistics

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

 View Only

  • 1.  STATS EXRASCH

    Posted Mon August 12, 2024 10:10 AM

    Hi,

    I am a newcomer to rasch analysis, and I am trying to run this extension for spss for the first time. I have tried different data sets, with 10 items and 200 cases, however none of which were succesful. I did a rasch analysis of partial credit model method on this data, but I get a time out message after 10 minutes. 

    Can anyone help?

    NB sorry Jon, I don't know if I missed your response. 



    ------------------------------
    Christian Kobbernagel
    ------------------------------


  • 2.  RE: STATS EXRASCH

    Posted Mon August 12, 2024 04:19 PM
    Sorry.  I hadn't had time to look at this, but when I did, I got some error messages, and I'm not sure what is needed, since I'm not familiar with your data.

    The latest message says 
    Error in datcheck(X, W, mpoints, groupvec, model) :
      Subjects with only 1 valid response must be removed!

    but it looks like all your cases have only one response.  It also complained earlier about cases with no valid data.

    If you are getting no output at all from the procedure, you may be stuck on the phantom OMS bug.  Occasionally, SPSS gets stuck in a state where OMS is suppressing all Viewer output.  Run the command OMSEND. in a syntax window to make sure that that doesn't happen.

    The underlying package documentation says


    Input data matrix or data frame with item responses (starting from 0); rows represent individuals, columns represent items. Missing values are inserted as NA.

    If I run this with Kid1, Kid2, and Kid3, I get this output.
    image.png

    --



    Original Message


  • 3.  RE: STATS EXRASCH

    Posted Thu August 15, 2024 09:16 AM

    Dear Jon,

     

    Thanks a lot for taking time to look at this for me. It is greatly appreciated.

     

    Your output makes sense in some way, however in another it does not. How come there is a separation of the values in the data? Those are likert scale items with response cats of 1 to 4. The output arrange them as binary items, how is that? Also the CI intervals are somehow distorted because estimates are outside them.

     

    But, it is really good to see that it can be done. I don't understand how I cannot get it to run at all.

     

    I will try the bug thing you mention and then try and run it again. Is there any other places it is possible to get support on technical stuff with spss?

     

    Mh Christian

     

    --------------------------------------

    Christian Kobbernagel
    Ekstern lektor Kommunikation, IKH

    cand.comm, ph.d

    RUC

    Roskilde Universitet
    Institut for Kommunikation og Humanistisk Videnskab/Center for N

    Universitetsvej 1
    DK-4000 Roskilde
    Telefon: +4546743881 / 61408066
    ruc.dk

    facebook.com/ruc.dk
    twitter.com/roskildeuni
    instagram.com/roskildeuniversitet

     




    Original Message


  • 4.  RE: STATS EXRASCH

    Posted Thu August 15, 2024 09:18 AM

    HI  - sorry about the CI values I see them correct now. Best, Christian

     

    --------------------------------------

    Christian Kobbernagel
    Ekstern lektor Kommunikation, IKH

    cand.comm, ph.d

    RUC

    Roskilde Universitet
    Institut for Kommunikation og Humanistisk Videnskab/Center for N

    Universitetsvej 1
    DK-4000 Roskilde
    Telefon: +4546743881 / 61408066
    ruc.dk

    facebook.com/ruc.dk
    twitter.com/roskildeuni
    instagram.com/roskildeuniversitet

     




    Original Message


Related Content