Saving Factor Scores Using LavPredict w/ 2 Groups

[Allie Choate]

Hi everyone,

I am trying to save factor scores using the LavPredict function from a multiple-group CFA.

I am able to save the factor scores successfully, but because I have a differing number of males and females in my groups, I am trying to thus save just the male factor scores into its own subset and then same with the female scores. I'm assuming this should be a easy fix, just haven't been able to find much on the topic. 

My code is below. Any help would be very much appreciated! 

Thanks in advance!

###

model1<- cfa(bifactormodel, data = main.data, group="Gender", 

                 std.lv=TRUE, orthogonal = TRUE, 

                 estimator = "MLR", missing="fiml")

summary(model1, fit.measures = TRUE, standardized = TRUE)

fscores <- lavPredict(model1,type="lv",method="regression")

[Terrence Jorgensen]

I am trying to thus save just the male factor scores into its own subset and then same with the female scores.

lavPredict() already returns a list (one matrix per group) of factor scores for multigroup models.  Using the ?cfa help-page example:

example(cfa)
model1 <- update(fit, group = "sex")
fscores <- lavPredict(model1, type="lv", method="regression")
str(fscores)
lapply(fscores, head)
sapply(fscores, nrow) # group sample sizes

If you are wondering how to merge them back into your data, you can use the case indices:

(ID.list <- lavInspect(model1, "case.idx"))
IDs <- do.call(c, ID.list)
fitted.data <- main.data[IDs, ]
allFscores <- do.call(rbind, fscores)
fscores.data <- cbind(fitted.data, allFscores)

Terrence D. Jorgensen

Postdoctoral Researcher, Methods and Statistics

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

UvA web page: http://www.uva.nl/profile/t.d.jorgensen

[Allie Choate]

I figured it was an easy fix that I was just missing!

Thank you so much.