R (http://www.r-project.org/) is a free, open-source, cross-platform system for statistics, data analysis, and graphics. While very powerful, R can have a steep learning curve, as it functions more like a programming language than a menu-driven system and may feel less familiar to users of SPSS, SAS, Systat (which I used and loved for many years before discovering R), Stata, and other more familiar, point-and-click statistics packages. A Web site called Quick-R helps users of those packages to quickly become familiar with R and its subtleties. From the author of Quick-R:
“R is an elegant and comprehensive statistical and graphical programming language. Unfortunately, it can also have a steep learning curve. I created this website for experienced users of popular statistical packages such as SAS, SPSS, Stata, and Systat (although current R users should also find it useful). My goal is to help you quickly access this language in your work. I assume that you are already familiar with the statistical methods covered and instead provide you with a roadmap and the code necessary to get started quickly, and orient yourself for future learning. I designed this web site to be an easily accessible reference.”
Learning R is well worth the effort. As just one example, my favorite statistical graphic, the notched box plot, is not available in SPSS, but can easily be produced in R with a short line of code. (See McGill et al., Variations of Box Plots, American Statistician, 1978, Vol. 32, No. 1., pp. 12-16.) The notches in notched box plots for grouped data display simultaneous 95% confidence intervals around the medians and function as a visual nonparametric analysis of variance, with the notches providing a measure of the rough significance of differences between the values. And having a free statistical package that runs under Windows, Mac OS, and Unix/Linux, with free upgrades and NO annual license fees, is very attractive as well.