Proc GLIMMIX with 3 random intercepts - univariate analyses

[KPCklebspn]

Hi all,

I've got a hierarchical model with 3 random intercepts and some fixed effects.

I want to present the full model (3 random intercepts plus room- and unit-level factors fixed effects). Outcome is a dichotomous var. 

If this were a simple logistic regression, we'd present the univariate analyses prior to presenting the full model.

How to univariate analyses work with multilevel modeling? When I'm doing the univariate analysis for each room- and unit-level factor, do I include all 3 random intercepts?

Thx!

[Community_Guide]

Hello @KPCklebspn,

Your question requires more details before experts can help. Can you revise your question to include more information? 

Review this checklist:

• Specify a meaningful subject line for your topic.  Avoid generic subjects like "need help," "SAS query," or "urgent."
• When appropriate, provide sample data in text or DATA step format.  See this article for one method you can use.
• If you're encountering an error in SAS, include the SAS log or a screenshot of the error condition. Use the Photos button to include the image in your message.
  
• It also helps to include an example (table or picture) of the result that you're trying to achieve.

To edit your original message, select the "blue gear" icon at the top of the message and select Edit Message.  From there you can adjust the title and add more details to the body of the message.  Or, simply reply to this message with any additional information you can supply.

SAS experts are eager to help -- help them by providing as much detail as you can.

This prewritten response was triggered for you by fellow SAS Support Communities member @sld

.

[Haris]

Sounds like you need to report univariable effects of your FIXED factors while controlling for your random effects.  Random effects usually reflect sampling differences that you want to control for rather than focus on.

It may also be interesting to know if the effects of your fixed variables are different across the levels of your random effects.  If that is the case, random intercepts model may not be appropriate and you may need to model cross-level interactions.

The final answer is somewhat context-specific.