You can use formats to have character values displayed for numeric values.  If you already have the data set created, you can run code similar to the following in a Program Editor: libname myfmtlib " /* path to folder to save format definition */ "; proc format library=myfmtlib; value myscale 1='Extremely likely'   5='Kind of likely'   9='Extremely unlikely'   other='n/a'   ; run; options fmtsearch=(myfmtlib); data libref.newdata; format /* name of numeric columns you want format applied to */ myscale.; set libref.olddata; run; After creating this data set libref.newdata that has the format applied, you will need to run and save this code in your Project Start Code in Enterprise MIner: libname myfmtlib " /* path to folder to save format definition */ "; options fmtsearch=(myfmtlib); Then you should be able to create a data source in Enterprise Miner for this new data set, and run flows as usual. Hope that works for you! ... View more

If you're familiar with logistic regression and need the SAS details these are good references: UCLA, their annotated output is useful though a bit out of date: SAS Topics: Logistic And of course the documentation: Example 51.2 Logistic Modeling with Categorical Predictors SAS/STAT(R) 9.2 User's Guide, Second Edition Read up on the class statement and how it dummy code/effect code variables as this is a common source of errors. FAQ: What is effect coding? FAQ: What is dummy coding? Search logistic regression on lexjansen.com for many relevant papers. My biggest suggestion is to find a set of data/analysis and figure out how to replicate it in SAS before doing your actual analysis on your data. It will save you many mistakes. ... View more

And here's a tip from Shunping on Spectral Clustering in SAS® Enterprise Miner™ Using Open Source Integration Node. ... View more