Lately, across the statistical blogosphere, the repeating discussion of R vs. SAS has started up again. In this series of posts, I’ll offer my opinions of the programs, and supply some information. In this post, I introduce the series and say a little about where I am coming from, so you can see where my opinions come from.

I’m a data analyst/statistician. Mostly, I work with researchers in the social and behavioral sciences, education, and medical fields. I’ve been using SAS for about 15 years and R for about 5, and I use SAS more than R. I am **not** a programmer.

There are many statistical packages, but there are two (SAS and R) that I use regularly – in fact, I use both every day. I like both. I don’t want to give up either. But they are very different.

**Websites for SAS and R**

For more on SAS see their home page ; for more on R see the R project page

**Two basic differences between SAS and R**

Two uncontroversial differences are:

**1. SAS is commercial, R is free.**

That is, with SAS, you pay an annual license fee, which varies depending on many factors. R, on the other hand, is free. Anyone can download it. (R is a dialect of S, there is also a commercial version of S – called S plus, but I haven’t used it, and I don’t see it mentioned much; there is also at least one commercial version of R, see the comments.).

**2. SAS has tech support, R does not. **

This is, of course, related to the first point. One of the things you are paying for with SAS is tech support – available by phone or e-mail. I have found SAS tech support to be among the best of any software I’ve used.

**What I plan to cover in future posts**

1. Ease of learning

2. Getting help

3. Error messages

4. Speed

5. Available statistics

**Request for assistance**

This series is really for two groups of people: Those trying to learn a little about these two prominent statistical software packages, and those who already know a lot, but want to discuss things. It might get a bit heated in the latter group – people have strong opinions. Please keep it civil.

If you have particularly good links on any of the above, or ideas for more topics in this series, or anything else you’d like to contribute, let me know.

I plan to search using Google (of course) and also look through both SAS-L and r-help and stackoverflow quite a bit. If there are other good places for me to look, let me know that too.

[learn_more caption="Author Bio"] I specialize in helping graduate students and researchers in psychology, education, economics and the social sciences with all aspects of statistical analysis. Many new and relatively uncommon statistical techniques are available, and these may widen the field of hypotheses you can investigate. Graphical techniques are often misapplied, but, done correctly, they can summarize a great deal of information in a single figure. ** I can help with writing papers, writing grant applications, and doing analysis for grants and research.**

** Specialties:** Regression, logistic regression, cluster analysis, statistical graphics, quantile regression.

You can **click here to email** or reach me via phone at 917-488-7176. Or if you want you can follow me on Facebook, **Twitter**, or LinkedIn. [/learn_more]

Peter,

I just wanted to point out that your first “uncontroversial difference” (between SAS and R) is indeed controversial as you have stated the first difference. I would agree that the first difference would be “uncontroversial” if you had merely said “SAS has an annual license fee and R does not.” But you used the loaded adjective “expensive” which entails a host of implications. For one example: “commanding a high price and especially one that is not based on intrinsic worth or is beyond a prospective buyer’s means” (Merriam-Webster). I’m sure you will agree that the adjective “expensive” is context dependent and not an intrinsic attribute of SAS.

Joe

Good point Joe. I will edit that.

Peter,

I’m very much looking forward to your series.

I would like to point out that there is support for R, ranging from “People helping people” with community support on the R mailing lists, sites like StackOverfow and even Twitter, to paid commercial support through companies like REvolution Computing.

–___

Best Regards,

Joseph

http://Twitter.com/JAdP

Hi Joseph – thanks! I had planned to cover r-help and stackoverflow in my post on “getting help” – what other mailing lists should I cover? I didn’t know REvolution computing provided commercial support.

In this post, I just meant that R, itself, doesn’t have a ‘tech support’ line.

What R lacks is professional support. Also, since many statistical applications are supported only by a few dedicated researchers, bugs, backwards compatibility with old versions of the applications, and forwards compatibility of applications with new versions of R may or may not happen in a timely fashion.

I humbly suggest a book Nick Horton and I wrote as a way to move between SAS and R comfortably (see http://www.math.smith.edu/sasr/) and our blog with applications in each (sas-and-r.blogspot.com).

Other good resources are the quick-R web site and Robert Muenchen’s book on R for SAS and SPSS users.

Thanks Ken! I will look into that book, and also your blog.

Peter,

Just a clarification, S-Plus is not a commercial version of R, it is a commercial implementation of the S language. R is an open source implementation of the S language. This is not just quibbling. There are sufficient differences between the two implementations (for example, ‘scoping’ rules) such that the same code can produce different results R and S-Plus. As pointed out above, REvolution Computing sells a version of R and provides support.

One of the great things about blogging is learning things!

shoot sweet info dude.

Joseph A. di Paolantonio:

Hi, Joe:

I would be cautious to equate the “asking question on R help list” to the technical line offered by SAS. I see people on R help list who are instructed to learn “google” first before asking questions. Beginers’s questions may also be ridiculed due to their “failure” to adhereing to the “posting guideline”….. You rarely get stuff like that from professional tech support…..

Hi Peter,

Does R supports Arabic characters? And is there any technical comparison between R & SAS?

Hi Sherif

I don’t know if R supports Arabic characters …. I’ve never seen them, but I haven’t looked.

What do you mean by “technical comparison”?

Peter

Hi Peter,

Thanks,I meant if there is a full details comparison between both.

Hi Sherif

A “full details” comparison between SAS and R is not possible, for at least two reasons: 1) SAS and R are designed to do different things. 2) SAS is closed source, so there is no way to compare their algorithms to those of R

Hi Peter, just out of curiousity. I believe your info says you are using SAS, S-plus.. and here you are saying you are not using S-plus. Which one is it?

Hi Larissa

Which information says I use S-Plus? I did use it, years ago, and it is similar to R, but I don’t use it now.

Thanks

Peter

hi peter,

can you post me a brief comparison between R and SAS other than the 2 basic differences u stated @ the first

Hi Lahari – they are VERY different programs. There are a couple books comparing them, and there have been numerous threads on StackOverflow and on SAS-L comparing them.

Mr. Peter:

What are the names of the couple books you are refering to that compare R with SAS?

Thank you

Hi

SAS and R

R for SAS and SPSS users

Peter

In my experience the R cognoscenti do not like to involve themselves

with mundane matters like “quality control”. Recently, Zhang et al.

2011 published some simulation results indicating serious problems with

the lme4 package. I verified some of the results and posted to the

R list. There was absolutely no response whatsoever.

For comparison I used AD Model Builder which is free software. It got results close to those reported by Zhang et al. for SAS NLMIXED.

I certainly would not use R for any serious mixed model analysis.

The link is.

https://stat.ethz.ch/pipermail/r-sig-mixed-models/2011q4/006953.html