Error with invariance testing

[Micah N]

Hello everyone,

I am having some trouble sorting out an error I am getting when performing measurement invariance testing. I am using measEQ.syntax in semTools and the problem arises when I try to incorporate complex survey design characteristics using lavaan.survey. Here's the code I ran:

model <- '

latent =~ v1 + v2 + v3 + v4 + v5 + v6

v1 ~~ v6

v5 ~~ v6

'

baseline <- measEq.syntax(configural.model = model,
                              data = data,
                              ordered = c("v1", "v2", "v3", "v4",
                                          "v5", "v6"),
                              parameterization = "delta",
                              ID.fac = "std.lv",
                              ID.cat = "Wu.Estabrook.2016",
                              group = "group",
                              group.equal = "configural")

model_baseline <- as.character(baseline)

fit_baseline <- cfa(model = model_baseline, data = data, group = "group", ordered = c("v1", "v2", "v3", "v4", "v5", "v6"))

This code works and summary(fit_baseline) provides error free results. But the error occurs when I try to account for the survey design:

survey_weight <- svydesign(ids = ~psu, strata = ~strata, weights = ~weight, data = data, nest = TRUE)

fit_baseline_surv <- lavaan.survey(lavaan.fit = fit_baseline, survey.design = survey_weight, estimator = "DWLS")

Error in start[th.idx] <- th.values[match(th.names.lavpartable, th.names.sample)] :
  replacement has length zero

Any ideas about how I could fix this issue?

Thanks for your help,

Micah

[Terrence Jorgensen]

the problem arises when I try to incorporate complex survey design characteristics using lavaan.survey

That package does not work with categorical outcomes.

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

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

http://www.uva.nl/profile/t.d.jorgensen