Multilevel SEM with categorical + nonnormal data

[Michael Filsecker]

Dear colleagues,

after doing some research I realized that lavaan does not support multilevel with categorical variables. I supposed that the only way is to use MPlus for now...I also read Stan could make it. Any experience with Stan on this?

Thank you!!

Michael.

[Stas Kolenikov]

Do it in Stan. It democratized Bayesian computing considerably, all you need to care about is increasing adapt_delta and max_tree_depth until everything converges cleanly.

-- Stas Kolenikov, PhD, PStat (ASA, SSC) 

-- Principal Statistician, NORC @NORCnews

-- Opinions stated in this email are mine only, and do not reflect the position of my employer

-- Social media: @StatStas [ Twitter | mastodon.online ]

-- http://stas.kolenikov.name

--
You received this message because you are subscribed to the Google Groups "lavaan" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lavaan+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/lavaan/CACXLW8_aky3X8wPxC4Y69x6BU_j%2Bns9637LzEdwtx37Lr0v-wg%40mail.gmail.com.

[Michael Filsecker]

Thank you Stas I will try.

Kind regards,

Michael.

To view this discussion visit https://groups.google.com/d/msgid/lavaan/CACHJnQCBo0-hT%3D4v6ZZstk0KnSbdWMpUbPt_bKpc4QEvDiCR9w%40mail.gmail.com.

[Jarrett Byrnes]

You can also do it in brms. Which uses stan. Check out my tutorial https://jebyrnes.github.io/bayesian_sem/bayesian_sem.html

On Aug 1, 2025, at 9:05 AM, Stas Kolenikov <skol...@gmail.com> wrote:



To view this discussion visit https://groups.google.com/d/msgid/lavaan/CACHJnQCBo0-hT%3D4v6ZZstk0KnSbdWMpUbPt_bKpc4QEvDiCR9w%40mail.gmail.com.

[Michael Filsecker]

Thank you Jarret, I will try, hope I can get it done!

Kind regards,

Michael.

To view this discussion visit https://groups.google.com/d/msgid/lavaan/7DA1AED6-2315-4922-8B79-D95A4D76F2F2%40gmail.com.