I would like to learn some SAS because I am interested in a few industries that tend to use it exclusively. However, I don't want to get stuck with a resource that assumes I know nothing about statistical programming. Is there a good guide for programmers with statistics experience in R?
Thanks, Steven
The Little SAS Book is the only SAS book I have seen. I liked it as an introduction but you may still find it useful. There's a preview on Google Books at the link.
SAS for Dummies is supposed to be pretty good. I haven't read it, but people here at work seem to like it.
Robert A Muenchen read your mind. He's published a free manuscript that later became a Springer book called R for SAS and SPSS users. It's really written for folks who know SAS or SPSS but would like to learn R. Luckily he gives examples of how to do things in all three languages. It will also work well if you know R and want to back into SAS. It started as a free manuscript and then Springer helped him flesh it out into a full book. Here's some links:
In addition, here's a paper on the topic of passing data back and forth between R and SAS.
When you talk to people in industry keep in mind that to many end users 'SAS' may mean a GUI interface into one of the SAS tools. I learned SAS programming in graduate school and wrote programs in it for years. To me SAS was a language. I remember having a really awkward cocktail talk with the wife of a friend who told me she was learning SAS. I was excited and started talking about PROC statements and DATA STEP programming. She tried to be polite but pretty soon I could tell she had no idea what I was going on about. She was a forecast analyst and was using a SAS forecast tool that had a GUI on top which she was learning to use. I soon realized that when people talk about using SAS it is about as specific as 'using Microsoft.' So it is worth brushing up on the suite of SAS BI tools, Data Mining tools, etc. I think they all have SAS language underneath them, but don't assume someone is talking about SAS programming just 'cause they say 'SAS.'
most sas manuals are on line. here is the list of everything: http://support.sas.com/documentation/
this may be a good place to start: http://support.sas.com/documentation/cdl/en/basess/58133/PDF/default/basess.pdf
it depends on how much of an R guru you are, but if you tend to think in terms of vectors, then you will feel more at home with sas/iml or stat studio 3 (which will be renamed to iml studio), which has a matrix based scripting language. No passing functions, nor closures, though. sas institute has announced that they will provide interfaces to R in the iml studio. see: http://support.sas.com/rnd/app/studio/Rinterface2.html
The problem with 'SAS for Dummies' is that it's written for people getting data out of SAS; it's a limited book. I strongly, strongly urge you to get 'The Little SAS Book'. It's sweet. When I was a graduate student learning SAS, I'd have gladly paid twice the cover price for it.