Best option in cfa function for Discrete Data

57 views

Skip to first unread message

Henrique Oliveira

unread,

Jul 5, 2020, 5:22:58 PM7/5/20

to lavaan

Hi,

I have a model composed of 26 itens, where the values in each item are discrete, following a five points likert scale (1 to 5).

I have smooth range values of skewness and kurtosis: -0.934<sk<0.298 & -1.226<ku<1.301.

For cfa function in lavaan, estimator = "WLSMV" is the best option to be used?

Best,

Henrique

Carlos Ossio

unread,

Jul 5, 2020, 5:58:49 PM7/5/20

to lavaan

There is an interesting paper comparing DWLS and WLSMV:

DiStefano, C., & Morgan, G. B. (2014). A comparison of diagonal weighted least squares robust estimation techniques for ordinal data. Structural Equation Modeling, 21(3), 425–438. https://doi.org/10.1080/10705511.2014.915373