Re: Implementing Multiple Group Model with Negative Weights in lavaan

47 views

Skip to first unread message

Message has been deleted

Terrence Jorgensen

unread,

Dec 8, 2021, 10:06:12 AM12/8/21

to lavaan

Aren't sampling weights relative? You could rescale the probabilities by adding a constant (negative one times the negative probability), then dividing by the sum of those shifted probabilities. Actually, division shouldn't be necessary because by default, lavaan rescales any weights to sum to the sample size (i.e., average = 1). See "sampling.weights.normalization" on the ?lavOptions help page.

No idea if anyone has done a simulation study to verify that would be okay. But designing one yourself would be a nice publication that also validates the accuracy of your software.

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

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