So apparently SAS likes to use the last variable as the reference variable for logistic regressions. All of my variables are coded as one would expect, with the smallest/first variable as the reference, so I would like to override this odd SAS default. I've figured out that I can use (ref=first) next to variables listed next to CLASS in order to account for my binary variables (0=no, 1=yes) but I'm not sure how to account for my continuous variables. Here is the present state of my code for reference: PROC SURVEYLOGISTIC DATA=brfss19s.brfss19shortvar; CLUSTER _PSU; CLASS DIABETE42 (ref=first) SOFEMALE2 (ref=first); MODEL DIABETE42 = ACETOTAL _AGE80 SOFEMALE2; STRATA _STSTR; WEIGHT _LLCPWT; run; Where ACETOTAL and _AGE80 are continuous variables that should be ordered normally (instead of using the highest value as the reference). Any guidance to address this hurdle would be greatly appreciated!! P.S. I believe I'm using SAS 9.4
... View more