1
$\begingroup$

I'm examining if the Big Five personality traits have an effect on the relationship type someone is in.

The Five personality traits are latent with continuous indicators, while the dependent variable is manifest and categorical. If possible, I would also like to check if the personality traits are correlated.

Which technique is most appropriate for this specific model?

Thank you for your time.

SEM model

$\endgroup$
  • 1
    $\begingroup$ Sounds like multinomial logistic regression to me. You can do that within a latent variable modeling framework, using some programs (e.g. Mplus). $\endgroup$ – Jeremy Miles Jul 24 '16 at 19:16
  • $\begingroup$ Thanks for your answer! Should I analyze the CFA and multinomial logistic regression separately? Or does it make more sense to combine the two into a SEM model? $\endgroup$ – Transdisciplinary Jul 26 '16 at 15:40
  • $\begingroup$ It's better to combine the two, but it's a much harder model to fit, and you'd need a larger sample size. $\endgroup$ – Jeremy Miles Jul 26 '16 at 18:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.