Hi! I would like to evaluate a parenting intervention using the Linear Mixed Model in SPSS. I believe I know how to do this, but then I also need to control for a categorical variable (education) and a continues variable (stress). Can I put these two variables in the covariate(s) box? Thanks!
Yes you can although I never used mixed effects model using SPSS. You also need to see what other variables are important and how their omission can confound the estimates.
I think you need some theoretical background or background information on which variables are important and therefore how to construct your model.
An advantage of using mixed effects model is that you also get estimates of random effects of parenting information and this quite important too.
Dear Raziye I think you should first see the independent predictors for the outcome and then significant factors can be used for the mixed effect model.
Hi Raziye! I am not knowledgeable enough to propose anything specific, but the suggestions made by the three experts above seem important. So you might develop a theoretical background from which to derive possible predictors, analyse the independent associations of predictors with outcomes and then decide which variables to use as independent predictors and covariates, respectively.
Can you give more details about your models? What variables would you use?
I guess that if one of your control variables is categorical, you will have to include it in the "factors" section, whereas all others continuous variables should be entered in the "covariates" box. If you include a categorical variable in the covariate box, it will be treated in the model as a continuous variable, and you probably do not want that.
- Badges
- Science topic
- Similar topics
- Correction