I cannot get the output as my main predictor is dichotomous (in input I highlighted this variable as categorical). The information I have is that the predictor cannot be categorical. How could I sort this issue? Any ideas, please? Is mediation possible when predictors are dichotomous? Mediators are continuous and outcome (dependent) variables are latent (based on observed items). Thank you
Hi Witold,
You can absolutely have dichotomous predictors in the mediation analysis. MPlus SEM allows you to test such model. You should use CATEGORICAL option of the VARIABLE command to specify your predictors.
Thanks,
Sakhavat
Hello Sakhavat,
Here is what Mplus 'tells' me:
The CATEGORICAL option is used for dependent variables only.
The following variable is an independent variable in the model.
Witold
What happens in your syntax when you take your predictor variable off the CATEGORICAL list? I know that the scale of your predictor variable should not change your model estimation.
Hello Sakhavat,
Then I have the output. Is that ok? I thought I have to set this as categorical (this is the only categorical variable in my model).
Witold
The output you have now should be correct. If you have an MPlus User's Guide, I suggest you double check to see if there is no error in the variable specifications.
Hello Sakhavat,
I think I am all sorted now, thank you for your help
Regards, Witold
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size