1
$\begingroup$

I saw a paper comparing correlations between the latent factors using Wald's chi-square test of parameter constrains. Anyone knows how to get the Wald's chi-square test result with correlation matrix, chi-square and p-values for the latent factors with 'lavaan' package? I was able to use inspect(fit,'cor.lv') to get the correlation matrix for the latent variables but now sure where to find the complete test results including chi-square and p-values. Thank you.

$\endgroup$
  • $\begingroup$ What do you want to test? $\endgroup$ – Jeremy Miles Feb 27 '17 at 18:45
  • $\begingroup$ @JeremyMiles Test the correlations between the latent factors and report p-values, chi-square. Thanks. $\endgroup$ – Joe Feb 28 '17 at 13:07
  • $\begingroup$ rdocumentation.org/packages/lavaan/versions/0.5-22/topics/… $\endgroup$ – Jeremy Miles Feb 28 '17 at 14:33
  • $\begingroup$ @JeremyMiles Thank you. That is testing the coefficients of the individual variables. How to test the correlations between the factors such as visual, textual and speed? $\endgroup$ – Joe Feb 28 '17 at 17:40
  • $\begingroup$ @JeremyMiles Thank you! I used the code to test the correlations between these three factors (visual ~~ r1 * speed+ r2*textual'). Does it look right? ' HS.model <- ' visual =~ x1 + x2 + x3 textual =~ x4 + x5 + x6 speed =~ x7 + x8 + x9 visual ~~ r1 * speed+ r2*textual' fit <- cfa(HS.model, data = HolzingerSwineford1939) summary(fit, fit.measures = TRUE) con = 'r1-r2==0' lavTestWald(fit, constraints = con) inspect(fit,'cor.lv') $\endgroup$ – Joe Feb 28 '17 at 20:52
1
$\begingroup$

If you have named a parameter in your model, you can use the lavTestWald function to test it.

In your model:

"visual ~~ r1 * speed"   

Then to use the lavTestWald function:

lavTestWald(fit, constraints = "r1 == 0")
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.