Measurement Invariance - Lavaan & MPlus

32 views

Skip to first unread message

C. M.

unread,

Mar 1, 2019, 5:06:28 AM3/1/19

to lavaan

Hi,

I'm following a course on structural equation models. We discussed measurement invariance yesterday. The lecturer went through some examples, but used MPlus.

He went over how to set a reference group, so that the latent factor means can be compared in relation to that group.

However when I follow some examples in Lavaan, all latent factor means are estimated to be zero.

I suppose my question is; I'm not setting something like a reference group (as far as I'm aware), so which parts of the output can I compare?

Terrence Jorgensen

unread,

Mar 6, 2019, 2:14:02 PM3/6/19

to lavaan

I'm not setting something like a reference group (as far as I'm aware)

No, Mplus MODEL: allows you to define aspects of a model that apply to all groups, followed by separate group-specific MODEL sections. In lavaan, you just define the model once to begin with, but you can apply fixed/free values and/or parameter labels in vectors.

http://lavaan.ugent.be/tutorial/groups.html

So if you have 3 groups, you can apply the same labels to indicator intercepts to constrain them to equality

ind1 ~ c(int1, int1, int1)*1
ind2 ~ c(int2, int2, int2)*1
...

Then you can free the latent means in all but one group

factor ~ c(0, NA, NA)*1

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam