Only one suggested modifications (using modindices/ modificationIndices)

[Anabel]

Dear all, 

I have a simple model with two correlated factors which are composed of 6 indicators each (χ2  : 873.881, df= 54, CFI: .896, RMSEA = .075). When I request the modification indices using modindices/ modificationIndices without any further specification, I get only one suggestion for correlated residuals between two items, which does not even improve the model fit. However, if I randomly try out another pair of correlated residuals, the model fit significantly improves, meaning that there are several modification options which are just not displayed. Does anyone have an idea how to solve this problem?

Some additional information: the model I’m referring to is specified at level 1 in a multilevel model with a saturated model at level 2. I do not have the problem when I request modification models for other models at level 1 or for the same model at level 2 (i.e., in every other case several modification options are displayed). 

Many thanks in advance
Anabel

[Anabel]

Should I perhaps rephrase the question or is it just an uncommon problem?

[Terrence Jorgensen]

More information might help (e.g., your model syntax, lavaan() call, and summary() and modindices() output)

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

[Anabel]

 

Hi Terrence,

Thanks for your reply! As I’m only interested in which residual correlation to include, I thought I could try out all different combinations to see which improves the chi-square value the most. Yet I’ve already encountered a new problem while doing this, which I’ve posted here: https://mail.google.com/mail/u/0/#inbox/FMfcgzGkZkNBMhMzcjDMWfVSJxdXdfsr

Kind regards
Anabel

--
You received this message because you are subscribed to the Google Groups "lavaan" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lavaan+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/a6527b56-5266-4557-b1ad-e3bed028ca9fn%40googlegroups.com.