SPSS Statistics

SPSS Statistics

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

 View Only
  • 1.  EFA results different in SPSS vs Jamovi

    Posted Thu September 03, 2020 05:25 PM

    Hi everyone,

    I am conducting a scale development study for my dissertation and I have run into quite a barrier. When I run the analysis in Jamovi, 6 factors are extracted. When I run SPSS, 14 are extracted. I have to figure out which program is correct (or at least better for the purpose of scale development) before I can proceed. I have 159 participants, no missing data, 126 items, and I am running Principle Axis Factoring with Oblimin rotation and using eigenvalues > 1 as my cutoff criteria. I ran the analysis in R as well to compare, and it is much closer to Jamovi compared to SPSS, though it is still a little bit different.

    My overall question is why this is so drastically different, and which is best to use, but I understand it is not as simple as asking that and getting an answer. Any insight is helpful.

    1. I checked to see if it the Kaiser normalization. It is not. When it is shut off, the results barely change and they are still drastically different from Jamovi. I ran the same EFA in R with Kaiser normalization and specified 14 factors, and things don’t align there with SPSS either. What else could be going on?

    2. I ran an EFA (PAF, eigenvalues > 1) with no rotation so I can compare it to Jamovi. Should the number of eigenvalues > 1 ever change in total from the initial values to the total after extraction? For example, in SPSS it gives me 14 initial eigenvalues > 1, but then under “extraction sums of squared loadings” there are 8 eigenvalues > 1, and the eigenvalues got smaller (which is alright from what I learned about common variance vs all variance mentioned above). Which is appropriate at that point? Do I pay more attention to initial or after extraction?

    3. How are eigenvalues calculated in SPSS? The initial eigenvalues are different in Jamovi and so I am wondering if this is where the problem might start.

    4. Any other recommendations on where the issue might lie?

    Thanks in advance for your help in building my understanding.






    #SPSSStatistics
    #Support
    #SupportMigration


  • 2.  RE: EFA results different in SPSS vs Jamovi

    Posted Thu September 03, 2020 05:36 PM

    You can see how SPSS Statistics calculates the factors in the Algorithms documentation, which is here

    ftp://public.dhe.ibm.com/software/analytics/spss/documentation/statistics/26.0/en/client/Manuals/IBM_SPSS_Statistics_Algorithms.pdf

    The difference in starting values should not matter.






    #SPSSStatistics
    #Support
    #SupportMigration


  • 3.  RE: EFA results different in SPSS vs Jamovi

    Posted Thu September 03, 2020 07:44 PM

    Thanks for that! I will start digging in.

    Can you please elaborate on why starting values for the eigenvalues shouldn't matter? Since I am using eigenvalues > 1 as my criteria it seems like it would impact that.






    #SPSSStatistics
    #Support
    #SupportMigration


  • 4.  RE: EFA results different in SPSS vs Jamovi

    Posted Thu September 03, 2020 07:44 PM

    Thanks for that! I will start digging in.

    Can you please elaborate on why starting values for the eigenvalues shouldn't matter? Since I am using eigenvalues > 1 as my criteria it seems like it would impact that.






    #SPSSStatistics
    #Support
    #SupportMigration


  • 5.  RE: EFA results different in SPSS vs Jamovi

    Posted Thu September 03, 2020 07:46 PM

    Whoops. I responded inline below.






    #SPSSStatistics
    #Support
    #SupportMigration


  • 6.  RE: EFA results different in SPSS vs Jamovi

    Posted Thu September 03, 2020 08:30 PM

    Note this (from the CSR)

    MINEIGEN(n). Minimum eigenvalue used to control the number of factors extracted. If

    METHOD=CORRELATION, the default is 1. If METHOD=COVARIANCE, the default is computed as (Total

    Variance/Number of Variables)*n, where Total Variance is the total weighted variance principal

    components or principal axis factoring extraction and the total image variance for image factoring

    extraction.

    So you can set the threshold as desired. How do the other statistics in that table compare with other sources?






    #SPSSStatistics
    #Support
    #SupportMigration