APIM model

[DAVID VILARIÑO]

Dear Prof Jorgensen;

I am conducting a research about violence in couples and I am aplying APIM model. I have 2 exogenous variables, 2 mediators and 3 dependent variables (5 endogenous). Each variables has several subscales?. It is necessary to split all variables by gender to conduct APIM model.

My strategy is :

- Collapse all items of subscales in a observable variable. I mean each subscale only will have a value (the average of values of people's answer)

- All exogenous variables will be correlated

-Residual error will be correlated.

- Each variable (subscale) has to be related with endogenous variable both for man and women of each couple(following APIM model)

- Assess each exogenou variable in different models, one for each exogenou variables.

I am coming up that both models are saturated and I can't know if the model converges or fit.

I tried it with the 2 exogenous variables at the same time and I have the same problem. The model is saturated.

Apart of this, can I extract conclusion about "effects" between variables" despite to the model is saturated?

Is this plan correct?

Any thought to improve the model?

Would be necessary to simply it, eliminating any variables?

Can I do this kind of model (APIM) with so amount of variables?

Thanks in advance.

David Vilariño

Universidad de Jaén

Spain