Understanding and Resolving Convergence Issues in SEM/CFA

Поділитися
Вставка
  • Опубліковано 4 лис 2024

КОМЕНТАРІ • 3

  • @Butterbirne0716
    @Butterbirne0716 8 місяців тому

    Dear Mr. Regorz,
    thank you very much for that informative video - I was struggling with the convergence problem for a couple of days but I finally found the problem. Namely, two (of my three) factors have a correlation close to 0. Furthermore, the correlation between the items of these two factors (which are also three each) are also close to 0.
    As this is mentioned in your second point (Model Underidentified), I understand that the factors, that I assumed to be correlated, are not correlated. What exactly would you recommend to handle this problem? I, unfortunately, am not not able to change my model because this was derived theoretically but I still want to "confirm" that my Items are loading on these factors.
    Thank you very much in advance for your help!

    • @RegorzStatistik
      @RegorzStatistik  8 місяців тому +1

      One possibility could be to change to PLS-SEM, a very different SEM-technique which has, as far as I know, not that many convergence problems.

    • @Butterbirne0716
      @Butterbirne0716 8 місяців тому

      Thank you very much for your answer. I already had a look at PLS SEM but it doesn’t seem to offer a good guidance on factor analysis - or at least I haven’t found any good papers yet. In case someone has found any, I would highly appreciate if they could be shared!