CFA endogenous vs exogenous variables

[João Amaral]

Hi!

It's my first post here, hope I'm not asking something obvious. Reading the lavaan documentation, I realized that the ordered argument in the cfa() function is related only to endogenous variables, right?

sdq_model_sr <- '
Social =~ sdq6_sr + sdq11_sr + sdq14_sr + sdq19_sr + sdq23_sr
Internalizing =~ sdq3_sr + sdq8_sr + sdq13_sr + sdq16_sr + sdq24_sr
Externalizing =~ sdq5_sr + sdq7_sr + sdq12_sr + sdq18_sr + sdq22_sr
Hyperactive =~ sdq2_sr + sdq10_sr + sdq15_sr + sdq21_sr + sdq25_sr
'

Considering the model above, where all variables are likert-type items from the Strengths and Difficulties Questionnaire (SDQ), are those variables considered endogenous? Should I worry about exogenous variables only when doing  a more complex SEM model with regressions involved?

Best  regards,

João

[Yago Luksevicius de Moraes]

Exogenous variables have no arrows pointing to them. 

Because you used your items as indicators of a latent variable, they are considered endogenous.

If you use an ordinal exogenous variable (e.g., 'sex~Social'), lavaan will give a warning and treat it as intervalar (or will give an error, and will not fit your model. Not sure now). You can easily solve this problem by adding a covariate (e.g., 'sex~Social; sex~~age'). It works if you set the covariation to zero too (e.g., 'sex~Social; sex~~0*age')

Best regards,

Yago

[João Amaral]

Thanks a lot, Yago!

--
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/0cca1bb2-71fc-45ea-a02c-c289a7d9db50n%40googlegroups.com.