How to add Control Variable in SEM Mediation Analysis

[Adeela Farqan]

Hello,

 

I want to add a control variable to this sequential mediation syntax below. Can someone please guide me as to how to include a control variable, say ‘C’ to this model below:

 

My model is: X -> M1 -> M2 -> Y

 

 

Model <- ' M1 ~ a*X

M2 ~ b*M1

Y ~ c*M2

Y ~ d*X

Indirect Effect := a*b*c

Direct Effect := d

Total Effect := (a*b*c) + d

'

 

fit<-sem(Model, data=ade_data,se="bootstrap", bootstrap=1000)

summary(fit,fit.measures=TRUE,standardized=TRUE,rsquare=TRUE)

[harold kincaid]

Others are no doubt more expert, but a "control variable" just needs to be another causal link I think. So you have to decide what link you find plausible or want to test. Best Harold

Séanadh Ríomhphoist/
Email Disclaimer
Tá an ríomhphost seo agus aon chomhad a sheoltar leis faoi rún agus is lena úsáid ag an seolaí agus sin amháin é. Is féidir tuilleadh a léamh anseo. 
This e-mail and any files transmitted with it are confidential and are intended solely for use by the addressee. Read more here. 

--
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/3594b92b-b303-4bf8-be67-5a4f77009292n%40googlegroups.com.

--

Harold Kincaid

Professor

School of Economics

University of Cape Town

kincaidh...@gmail.com

[Pat Malone]

Adeela,

If by 'control variable' you mean something analogous to a covariate in ANCOVA, the usual way is to regress all your model variables on it.  

So

X ~ C

M1 ~ a*X + C

etc.

If that's not what you have in mind,  please say more about what role you have for it conceptually. 

Pat

Patrick S. Malone, PhD
Sr Research Statistician
FAR HARBΦR
+1 803.553.4181 | pat@ | farharbor.com
This message may contain confidential information; if you are not the intended recipient please notify the sender and delete the message.