- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hi SAS Community,
I need to embark on some data analyses but am at a loss as to what makes the most sense. Hopefully, you bright minds can shed some light. It is a repeated measures design (participants are assessed at two time points). My outcome of interest is dichotomous (low need vs. high need). I have 9 independent variables: sex (male/female), age (will be treating this as categorical in analysis with 3 groupings), agency type and 6 continous measures on a variety of personality traits. What I am interested in answering are the following 3 questions:
1) Is there a difference in the outcome variable between Time 1 and Time 2? --> Was thinking a McNemar test, but open to suggestions.
2) Are sex, age, agency type and the personality indicators predictive of higher need?
3) How does data at the agency level (will be pulling data from one of the agency types on all 9 independent variables) compare to the whole picture?
It's questions 2 and 3 that are causing me the most grief. Not sure what is the best way to answer my research questions while being sensitive to the fact there is independence in observations between two time points. Any suggestions would be greatly appreciated.
If there are any questions to help clarify, please feel free to ask.
Thank you in advance!
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
I cannot tell what you want with respect to question 3, but that aside, I would consider a mixed model logistic regression, which will definitely address Q1 and Q2. For an example, see Multilevel Models for Categorical Data Using SASĀ® PROC GLIMMIX: The Basics or Hierarchical Logistic Regression Modeling with SAS GLIMMIX .
I hope this helps.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Thank you for your input and the resources!