Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Omega cofficient CFA or PFA

Hi all,

I need to estimate coefficient omega. I am wondering what is the difference when estimating omega using the "Advanced Options", CFA or PFA?

Thank you in advanced,

Nikos

Comments

  • Hi Nikos,

    so in order to find the reliability, one approach is to estimate a common factor underlying the variables. CFA and PFA are two different ways to estimate such a factor. In practice, which of the two you choose will make little difference, but for small samples PFA can yield results while CFA may not yield results at all.


    Julius

  • Thank you Julius for your response,

    Is there any reference that the two approaches CFA and PFA produce similar results? In addition, should I select non parametric bootstrap (the default option) or parametric? My experience from empirical data with small samples (around 50) suggest that PFA results in equal or higher omega values.

    Nikos

  • I do not have a reference, that is just my experience. My paper here does show that the PFA-based omega yields accurate results for point estimation as well as intervals with non-parametric bootstrap https://doi.org/10.1080/00273171.2021.1891855.

Sign In or Register to comment.