I’ve had a student complete a piece of work with some great data. I am considering re-analysing the data for publication. SEM makes sense, but the model is both formative and reflective. How can I overcome this issue?
-
1$\begingroup$ What do you mean by "formative" and "reflective"? $\endgroup$– Giuseppe Biondi-ZoccaiApr 26, 2018 at 8:20
-
1$\begingroup$ Thanks for your comment. Really good explanation found here: stats.stackexchange.com/questions/184947/… $\endgroup$– user205985Apr 26, 2018 at 16:17
-
1$\begingroup$ One example of a model with both kinds of indicator is a MIMIC model, see, for example: stats.stackexchange.com/questions/189462/… But there's nothing special that you need to do - just put them both in. $\endgroup$– Jeremy MilesApr 26, 2018 at 17:06
-
$\begingroup$ Please clarify what "the issue" is here. $\endgroup$– Gregg HApr 27, 2018 at 20:17
-
$\begingroup$ Thanks Jeremy Miles, this is exactly what I was looking for :-D :-D $\endgroup$– user205985Apr 28, 2018 at 21:45