[R - Lavaan] Which estimator for confirmatory factor analysis?

#1
Hi,

Currently I am using Rstudio with the Lavaan package in order to perform a confirmatory factor analysis (CFA) on a dataset (N = ~ 700, 23 observed variables, 2 factors). The problem, however, is that the data are not even close to being multivariate normal, with a Mardia's kurtosis of 378 and a Mardia's skew of 1140.

I am aware there are several estimation methods designed to handle deviations from multivariate normality (e.g. MLM in Lavaan, ADF). Which estimation method would be best for my analysis? And is it even sensible to perform a CFA with this kind of kurtosis and skew?

Thanks in advance for your help!

Niels