4
$\begingroup$

I estimated measurement invariance using

measEq.syntax(
  parameterization = "theta",
  ordered = ...,   
  ID.cat = "lisrel",
  group.equal = c("thresholds", "loadings", "intercepts")
)

By default, it fixes the factor variance in the first group (first year) to 1 for identification.

My question: is it possible to instead freely estimate the factor variances to compare them and identify the model by fixing the loading of the first item to 1 in each year?

In other words, can I change the identification strategy in measEq.syntax, or is it not supported?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
5
$\begingroup$

You can use the ID.fac = argument in measEq.syntax to change the method for identification.

Setting ID.fac = "std.lv" fixes the factor variances to 1 (which is the default). However, using ID.fac = "auto.fix.first" will fix the first loading of each factor to 1 instead.

Multiple aliases for "auto.fix.first" (and "std.lv") are available, which you can find in the help page: ?semTools::measEq.syntax under the ID.fac argument.

Improve this answer
$\endgroup$

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.