SPSS Statistics

SPSS Statistics

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

 View Only

  • 1.  MANOVA Warning

    Posted Mon April 03, 2023 05:00 AM

    Greetings!

    H    ow to handle the unequal covariance assumption in RM MANOVA in SPSS?

     

    I am receiving a warning that the Boxes test of Equality of Covariance Matrices I not computed because there are fewer than two nonsingular cell covariance matrices.

     

    The data is normally distributed but among the three symptom severity subgroups the smaller group with severe symptoms has larger variability.

     

    How to resolve this situation in SPSS to have valid statistics?

     

    Thank you for your assistance to solve this situation.



    ------------------------------
    Viktoria Luhaste
    ------------------------------


  • 2.  RE: MANOVA Warning

    Posted Mon April 03, 2023 09:05 AM
    Viktoria;

    Messages that mention singularity have to do with correlations among the variables in your model being highly correlated.
    My guess, without seeing the data and the model, is that a couple of variables (or more) are highly correlated. This is not a problem with the software but with the data. You might want to look at correlations of the variables in the model and see if there are some variables that are strongly correlated. One or more are redundant and should be removed.


    ------------------------------
    Rick Marcantonio
    Quality Assurance
    IBM
    ------------------------------

    Original Message