FROM THE APA SCIENCE STUDENT COUNCIL
Free Stats Software for Students!
The Science Student Council is a group of nine graduate students who spend a couple of weekends a year with the Science staff, advising on programs and activities that would benefit graduate students in psychological science. In this column, the students will present useful information that other graduate students need to know! Visit the Science Student Council to learn more about the activities of the SSC.
Free Stats Software for Students!
by Felix Thoemmes
Major statistical packages, like SPSS or SAS, offer a wide variety of data analytic techniques. Sometimes however, researchers are interested in more uncommon methods for which they require special statistical software. Fortunately, many programs that perform these operations are freely available on the Internet.
Here is a (non-comprehensive) list of some of the freely available statistics software:
A great freeware resource to calculate statistical power for various research designs is GPower, developed by Franz Faul and colleagues, currently in Version 3. It is freely available for download from Heinrich-Heine-Universität. Gpower calculates required sample size, minimum needed effect size, and expected power, for a wide variety of experimental designs. The user-friendly and intuitive interface along with the graphical presentation makes Gpower a great choice for power analysis.
Multiple Imputation, a new and important technique to handle missing data, is not supported by many general use packages yet. Joe Schafer and his team at Penn State University have developed the freeware program NORM, which delivers this state-of-the-art technique with ease. It can be downloaded from Pennsylvania State University.
Structural Equation Models (SEM) are ever increasingly used but need special software to be estimated. Many schools have site-licenses or computer labs equipped with SEM packages, but it is also possible to freely download student demo versions of the most popular programs. These versions are necessarily limited in its functionality but are sufficient to aquatint yourself with the software. AMOS, LISREL, and Mplus all offer limited free version on their websites. Other programs, like MX are completely free.
An invaluable resource for the Bayesian statisticians, WINBUGS, is freely available from the Bugs Project. It offers a wide range of general statistics in the Bayesian framework but is generally more geared towards more advanced users. Available online tutorials can help you get started.
R is perhaps the most versatile and complete statistical package available at no cost. R. R has a very active community that frequently updates the program with so called “packages” that are add-ons geared at performing a specific statistical task. Examples include Item Response Theory, Structural Equation Models, or Optimal Matching, just to name a few. R is also aimed at more advanced users, because it requires knowledge of the command syntax and does not by default have a point-and-click interface. Again, tutorials exist online, and some add-on packages (e.g. R-Commander) offer a graphical interface to perform basic operations.
Hopefully, this list will be helpful next time you are in need of some statistical software that answers the research questions that you are interested in.