limiting logistic regression to one group

[gejoachim99]

I'm working with a dataset that has two control groups for reasons that I won't get into. I'm trying to do logistic regression on each of those separately - so, cases vs control group 1, then cases vs control group 2. What command do I use within my proc logistic to limit it to one group? Thanks!

[Tom · Accepted solution]

How is your data structured?

Assuming you have a variable with three possible values then just use a WHERE clause to eliminate the observations from one of the control groups.

title 'cases vs control group 2';
proc logistics data=....
where group ne 'Control Group 1';
...

title 'cases vs control group 1';
proc logistics data=....
where group ne 'Control Group 2';
...

[gejoachim99]

Yes, that worked! I have a variable exactly as you described, and that fixed it. Thank you very much!! 🙂