SAS vs SPSS: Pick one and forever hold your peace #MRX

Any serious researcher will fight you to the death to convince you that their favourite stats program is THE best stats program. But, there are good and bad things about each of them.

SPSS: Great for people who rarely use statistics, who don’t remember code, who are scared by code, have no time to learn code, who just need to use some of the basic processes with no special adjustment, who work with normal data that never changes. It’s quick and easy to learn and use, and you don’t 4 feet of manuals to find what you’re looking for. You can just click through the menu to find all the basics you’ll ever need. Compared to SAS, SPSS is major el cheapo.

SAS: Great for people who love coding. SAS does have a menu version but if you like menus, then you should be using SPSS. SAS is great for data manipulation as in creating brand new variables,  flipping cases into variables and vice versa, and re-running the same bit of code repeatedly with only a tiny change every time. The macros you can write are unlimited and awesome. And, if it doesn’t do the variation of a statistic you want, you can actually program that statistic into SAS. If you like collecting books, SAS can quickly contribute to that addiction. Buy them. You will need them.

R: What? A third choice? Oh yes. R is great for people who want to flaunt how anti-establishment they are. It’s open source and requires a lot of commitment to become competent. But it can do any statistic you’ve ever dreamed of and a billion more. And you can brag that you know hard core statistics programming. Do not attempt to learn it if Excel intimidates you. Otherwise, go and download it right now and get ready for a rip-roaring awesome 6 week holiday. It is free so prepare to salivate. Mmmmmm…. rrrrrrr. There are even a couple of self-help books now so you might want to find one of them. So, if you’re fresh out of school and no longer have the student version of SAS, R will be perfect for you.

What’s my preference? Well, I wish I was competent in R, but until then, SAS is the best thing since sliced bread.

7 responses

  1. Interesting piece and great discussion. When discussing the future of these packages, I think it’s also important to look at what is happenning in the big data analytics space. Much of the code from these packages are being re-written/adapted to exist in a multi-parallelized environement (e.g. SAS with HAP and LASR, R v3 and what RevoR is doing). The other trend I see is that folks (read software vendors) are developing to a visual interface and trying to look at things in context of a more production oriented environment. Coding is still critical especially in certain circumstances, but as we look to democratize the analytics and try and move our business users into more commodity type modeling approaches, these visual interfaces become critical. Also, as we move our data integration together with analytics, they are important here as well. I believe the future of these companies and packages have much to do with the battle between open source and proprietary solutions that is starting to play out and how the information architectures in firms develop.

  2. […] For an amusing discussion of whether to teach SPSS or SAS or R, check out the Lovestats blog. […]

  3. I would add another point about SPSS. Whilst you could stick with the menu interface, running one test or report at a time, there is a great deal more to it. First you can write pretty complex routines in SPSS syntax (repeated analysis, new variables, output to Excel etc). Second, you can use the programmability extensions to write programmes in either Python or R, both of which extend the functionality enormously (in fact, the code you will be writing will be a hybrid of Python/R and SPSS syntax). There are some severe limitations with SPSS, but the lack of a script-based interface isn’t one of them.

    Practically, where one might want to automatically generate large sets of charts, or to run a gazillion cluster analyses with different input conditions, I haven’t arrived at a brilliant solution yet. A solution, yes, but not brilliant. I’d like to think that if I find the right programme then it will do it for me. I am pretty sure this isn’t the case though, and that I’m going to have to simply improve my coding skills.

    As coders might well tell you, don’t worry about which language you choose to learn (think about which is best, by all means, but don’t drive yourself mad). Learn one properly, and then transferring to another is relatively simple – the principles tend to be very similar for all languages, and I suspect the same applies when thinking about stats programmes. Although an alarm in my head is immediately indicating that R might be an exception. I suspect the thrust of the argument stands, though.

    Thanks for the post.

    1. And thank you for the well thought out comment!

  4. one thing I would say is that SPSS is still pretty good for the vast majority of people in survey market research, and the deficiencies vs. SAS are not relevant to most. There are plenty of juicy stats and functions in there. And Custom Tables by itself lets people do what they need.

    For large scale data sets, SAS seems to be the way for a lot of people. For automation SAS is better too.

    I do think SPSS is relatively challenging to learn if you don’t have stats and data background. I feel as if a lot of vendors are offering data analysis solutions via web interfaces to the data. Good and bad, since it separates you from the raw data.

    1. You bring up an interesting point. I wish people could see that learning SPSS does NOT equate to learning statistics. Stats program are cool ONCE you know you’re doing a statistic and how you can interpret it or misuse it.

  5. I learned SAS, SPSS, and a few others in grad school, which was long before R was invented. My current fav is Statistica although I use SPSS too. Most of my work, though, yes … it’s in Excel. I have to consider the audience.

%d bloggers like this: