The R program for statistics is an amazingly powerful and completely free program (under the terms of the Free Software Foundation’s GNU General Public License). If you have any need to do statistics, then you really must take a look at R or, more formally, “The R Project for Statistical Computing“.

What exactly is R? Simply put, “R is a language and environment for statistical computing and graphics.” It is a special open-source implementation of S which is one of the earlier statistical programming languages.

There are a lot of programs that can “do stats” out there that you might consider using such as Statistica™, SAS® (formerly the Statistical Analysis System), IBM’s SPSS Statistics, Mathworks’® Matlab®, plus many, many others. I’ve used several of these at one time or another, but I’m most familiar with Matlab.

So why use R? First, each of the options listed above comes with fancy symbols — things like ® or © or ™ which means, among other things, that they cost $ or € or £ or ¥ …you get the idea. Those options aren’t free or even cheap. You must have money — sometimes a lot of money — to use them. Now, there is nothing wrong with charging a fee for great software but R doesn’t cost anything at all and, if you are like me, that is a huge plus. Second, support for R is remarkable. Yes, the same can be said for the others but R is supported by a community of people who believe in the power of open source software. Free software that works. Generally, that means they are very receptive to helping others when assistance is needed. And they tend to be very good statisticians and programmers. Third, R is very widely used in academia both by faculty members and researchers, and students of course. Given the cost (did I mention it costs *nothing* to use R?), it makes sense that academics and students would be drawn to it particularly given the power of the program.

And R is not solely for statistics. It works with numeric data of all types. Digital images, for example, can be processed, analyzed and displayed. Such images are, after all, just matrices of numbers.

Give R a look. If you already know about statistical methods and experimental design, you will not have any difficulty using R. If you don’t know much about statistics methods or design, then it is going to take some time. R is not the easiest program to learn and it was not designed to ‘teach statistics’ to people.^{2} At least, not people who aren’t stats students. At the same time, there are plenty of tutorials and helpful people out there so you will succeed if you persevere. On the practical side of things, the main interface is a command line. But there are some good GUI interfaces as well if you don’t like the command line. Check out RStudio, for example. Finally, a lot of people have developed packages to do specific things within R. A complete listing of such packages can be found at the Comprehensive R Archive Network (CRAN) List of Packages (by name).

For many uses, R simply cannot be beat.