Is it possible to estimate latent mean differences with meas.Eq.syntax?

[Erik Ruzek]

Hi all,

If utilizing the Wu & Estabrook approach for testing measurement invariance in categorical items, is there an option to specify in the meas.Eq.syntax to get latent mean differences?

I am utilizing the data and syntax from the Svetina et al. (2019) tutorial paper in the SEM Journal. The data can be found at https://figshare.com/s/3f1d195da6c78195dd70 

# read data

dat<-read.table("BULLY.dat", header=FALSE)

names(dat) <- c("IDCNTRY", "R09A", "R09B", "R09C", "R09D")

# model 

mod.cat <- 'F1 =~ R09A + R09B + R09C + R09D'

# Thresholds + loading invariance

thresh_load <- measEq.syntax(configural.model = mod.cat,

                       data = dat,

                       ordered = paste0("R09", LETTERS[1:4]),

                       parameterization = "delta",

                       ID.fac = "std.lv",

                       ID.cat = "Wu.Estabrook.2016",

                       group = "IDCNTRY",

                       group.equal = c("thresholds", "loadings", "means"))

model.thresh_load <- as.character(thresh_load)

fit.thresh_load <- cfa(model.thresh_load, data = dat, group = "IDCNTRY", ordered = paste0("R09", LETTERS[1:4]))

Thanks, Erik

[Terrence Jorgensen]

If utilizing the Wu & Estabrook approach for testing measurement invariance in categorical items, is there an option to specify in the meas.Eq.syntax to get latent mean differences?

(differences in) latent means become estimable conditional on scalar invariance (i.e., equality constraints on thresholds, loadings, and intercepts).  See the ?measEq.syntax help page.

Terrence D. Jorgensen    (he, him, his)
Assistant Professor, Methods and Statistics
Research Institute for Child Development and Education, the University of Amsterdam
http://www.uva.nl/profile/t.d.jorgensen