Fixing Variance Ordinal Data

[karchjd]

Should this not return a warning that the fixing of x8 variance is ignored because its declared ordered?

library(lavaan)
data <- HolzingerSwineford1939
data$x8 <- ifelse(data$x8 > mean(data$x8), 0, 1)
HS.model <- ' visual  =~ x1 + x2 + x3 
              textual =~ x4 + x5 + x6
              speed   =~ x7 + x8 + x9
              x8 ~~ 2*x8'

fit <- cfa(HS.model, data = data, ordered = "x8")

[yrosseel]

Indeed, this is now silently ignored. A warning would be nice, if we use parameterization = "delta" (the default). I will leave this issue open until this is done. (It may take a while).

[yrosseel]

Ok. I have added warning (in the github version). It looks like this:

Warning message: lavaan->lavParTable(): variances of ordered variables are ignored when parameterization = "delta"; please remove them from the model syntax; variables involved are: x8