Fitting growth curve model with ordinal indicators and imputed data

[Matti Cervin]

Hi all,

After some help here, I am working on fitting a growth curve model with ordinal indicators, but I can't get the syntax right. Below is a straightforward code that do not work. I impute data with mice and the use the list of imputed data sets as the input for lavaan. The ordinal indicators are x1-x4 and y1-y4 in this example.

imp<-mice(data, m=5, maxit=10, seed=1234)

milist <- miceadds::mids2datlist(imp)

model<-'

i1 =~ 1*x1 + 1*x2 + 1*x3 + 1*x4

i2 =~ 1*y1 + 1*y2 + 1*y3 + 1*y4

s1 =~ 0*x1 + 1*x2 + 2*x3 + 3*x4

s2 =~ 0*y1 + 1*y2 + 2*y3 + 3*y4'

fit<-runMI(model, milist, 

           estimator="WLSMV", 

           fun="growth", 

           ordered=c("x1","x2","x3","x4",

                     "y1","y2","y3", "y4"))

Anyone that knows how to fit a growth curve model with ordinal indicators? I would be very thankful for help with specifying the model!

Best wishes,

Matti

[Terrence Jorgensen]

Anyone that knows how to fit a growth curve model with ordinal indicators? 

As described in the article I posted on the other thread, you need to fix your thresholds in order to "squeeze interval information" from them.

x1 | 0*t1
...
y4 | 0*t1

And because you have no observed means, you must fix the mean of the latent intercepts to 0 for identification.

i1 ~ 0*1
i2 ~ 0*1

Terrence D. Jorgensen

Assistant Professor, 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

[Matti Cervin]

Thank you Terrence! I followed your instructions using this code:

model<-'

ifreq =~ 1*freq1 + 1*freq2 + 1*freq3 + 1*freq4

ipsy =~ 1*psy1 + 1*psy2 + 1*psy3 + 1*psy4

sfreq =~ 0*freq1 + 1*freq2 + 2*freq3 + 3*freq4

spsy =~ 0*psy1 + 1*psy2 + 2*psy3 + 3*psy4

ifreq ~ 0*1

ipsy ~ 0*1

freq1 | 0*t1

freq2 | 0*t1

freq3 | 0*t1

freq4 | 0*t1

psy1 | 0*t1

psy2 | 0*t1

psy3 | 0*t1

psy4 | 0*t1

'

fit<-runMI(model, milist, 

            estimator="WLSMV", fun="growth", 

            ordered=c("freq1", "freq2", "freq3", "freq4",

                      "psy1", "psy2", "psy3", "psy4"))

However, this error code appeared:

Error in lav_partable_flat(FLAT, blocks = "group", meanstructure = meanstructure,  : 

  lavaan ERROR: expected 4 threshold(s) for variable ‘freq1’; syntax contains 1

Any suggestions on how to proceed/fix the code?

Thanks!

--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/g-rV-D4LmV8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+un...@googlegroups.com.
To post to this group, send email to lav...@googlegroups.com.
Visit this group at https://groups.google.com/group/lavaan.
For more options, visit https://groups.google.com/d/optout.

[Terrence Jorgensen]

However, this error code appeared:

Error in lav_partable_flat(FLAT, blocks = "group", meanstructure = meanstructure,  : 

  lavaan ERROR: expected 4 threshold(s) for variable ‘freq1’; syntax contains 1

Any suggestions on how to proceed/fix the code?

Have you read the paper yet?  freq1 clearly has 5 categories, so you need to fix 4 thresholds to some kind of meaningful values, as the paper shows.

freq1 | 0*t1 + 1*t2 + 2*t3 + 3*t4

[Matti Cervin]

Thank you, Terrence. No, I have not read through the whole paper yet. I struggle with the math formulae so it will take me a while, but I will get through it.

--