R is a software package that is used for statistical analyses. It has a syntax-driven interface which allows for a high level of control, many add-on packages, an active community supporting the program and it’s users and an open structure. All in all, it aims to be statistical software that goes beyond pre-set analyses. Oh, and it is free too.

The R software is developed by the R Core Development Team, presently having seventeen members. New versions of the R software are coming out regularly, so apparently progress is made. The source code of the R software is open-source. This means that everybody is allowed to read and change the program code. The consequence of this is that many people have written extensions to R which are able to nest itself in the fundaments of the software. For instance, it can interact with programs such as (WIN)BUGS or have extensions based on C or Fortran code.

A typical R session can be characterized by its flexibility. The software is set up in such a way, that functions or command can interact and thereby be combined to new ones. Obviously many statistical methods are already available, but if a command just doesn’t do exactly what you want it to do, it can easily be altered. Or, you build your analyses from the ground up using the most basic of functions. If you can think of it, you can create it.

So basically, you can invent your own set of wheels. Of course, many wheels have been invented yet, so it is not necessary to do it again yourself. Snippets of R-syntax are readily available on the internet. They can even be combined into ‘packages’, which can easily be downloaded from within the R-software itself or the R-website. Many of these packages are actively maintained and constantly improved. So don’t worry about being confronted with outdated software.

Another distinguishing aspect of R is its data-structure. All data are assigned to objects. And since R can handle more than just one object, several (read: virtually unlimited) sets of data can be used simultaneously. Functions are stored in objects too and finally the output of functions are stored in an object as well. This opens up many possibilities. For instance, when we want to compare several statistical models, we can store these models in different objects, that can be compared using the right functions (ANOVA, for instance). In a later stage, these objects can be used to extract data, for instance to graph the results.

So, in order to answer the question what R actually is, it can be stated that R is a very open-structured and flexible software package, readily available and very suitable for statistical analyses.

– – — — —– ——–

*– – — — —– ——–
R-Sessions is a collection of manual chapters for R-Project, which are maintained on Curving Normality. All posts are linked to the chapters from the R-Project manual on this site. The manual is free to use, for it is paid by the advertisements, but please refer to it in your work inspired by it. Feedback and topic requests are highly appreciated.
——– —– — — – –
*