|
|
CFA with non-normal continuous data |
|
Message/Author |
|
|
Hi, I'm trying to run a CFA model with non-normal continuous indicators. I think MLM or MLMV is usually recommended, but I'm wondering if it's OK to use WLSMV in this case. Also, if I have a mix of non-normal continuous and ordinal indicators, WLSMV might be a good choice, right? Thank you. |
|
|
Both MLR and WLSMV are good choices. When there are categorical outcomes and many factors, ML (MLR) is computationally heavy due to numerical integration being required. |
|
NR posted on Monday, March 23, 2009 - 6:40 pm
|
|
|
Hi, Dr. Muthen Is there any reference on the performance of WLSMV for (non-normal) CONTINUOUS data? |
|
|
Not that I am aware of. |
|
Back to top |
|
|