warning messages with satorra 2000 estimator

[Mark]

Hi all, 

I'm trying to compare two models using the scaled chi-square test. Specifically, I'm using the lavTestLRT function. 

For some reason I get warning messages "In if (p0$free[p0.idx] == 0L) { ... : the condition has length > 1 and only the first element will be used" when using the  "satorra.2000" method. 

I'm not getting this warning message when I'm using the  "satorra.bentler.2010" method. 

The results are very different too. 

Would really appreciate any insight into this issue. Thanks in advance! 

lavTestLRT(model2pgenconfigfit,model2pgenloadingfit, method = "satorra.2000")

Scaled Chi Square Difference Test (method = "satorra.2000")

                      Df   AIC   BIC  Chisq Chisq diff Df diff Pr(>Chisq)

model2pgenconfigfit  520 23677 24373 1073.8                              

model2pgenloadingfit 539 23658 24280 1092.7     16.374      19     0.6322

There were 50 or more warnings (use warnings() to see the first 50)

> warnings()

Warning messages:

1: In if (p0$free[p0.idx] == 0L) { ... :

  the condition has length > 1 and only the first element will be used

2: In if (p0$free[p0.idx] == 0L) { ... :

lavTestLRT(model2pgenconfigfit,model2pgenloadingfit, method = "satorra.bentler.2010")

Scaled Chi Square Difference Test (method = "satorra.bentler.2010")

                      Df   AIC   BIC  Chisq Chisq diff Df diff Pr(>Chisq)

model2pgenconfigfit  520 23677 24373 1073.8                              

model2pgenloadingfit 539 23658 24280 1092.7      17.31      19     0.5689

[Yves Rosseel]

> For some reason I get warning messages "In if (p0$free[p0.idx] == 0L) {
> ... : the condition has length > 1 and only the first element will be
> used" when using the "satorra.2000" method.

Is this with 0.6-1? If not, try the option A.method = "delta" in
lavTestLRT().

If this is with 0.6-1, I would like to investigate further. Could you
send me your R script and the data (or a snippet of the data)?

Yves.

[Mark]

You are right - it was a version problem. Problem went away after updating to 0.6-1. 

Thanks, Yves!