|
Message/Author |
|
|
I'm trying to run a model with mediation...my IV and DV are normally distributed, but my mediator is count. Is there a way to do this? I get errors when I leave my mediator as count. I've tried setting it to categorical and using theta parameterization, though this doesn't seem appropriate. The syntax follows. Thanks in advance, Steve VARIABLE: NAMES ARE totcomps bdiscont intrntot; categorical IS totcomps; USEVARIABLES ARE intrntot totcomps bdiscont; Missing = all(-1234) ; ANALYSIS: TYPE = GENERAL MEANSTRUCTURE; parameterization = theta; BOOTSTRAP = 5000; MODEL: bdiscont ON totcomps* (p1); bdiscont ON intrntot; totcomps ON intrntot* (p2); totcomps*; [intrntot*]; intrntot*; MODEL CONSTRAINT: NEW(ind_effect); ind_effect = p1*p2; OUTPUT: CINTERVAL(bcbootstrap); |
|
|
Please send your input, data, output, and license number to support@statmodel.com so I can see the problem you get when you have the mediator variable a count variable. |
|
Back to top |
|
|