Tools for the armchair statistician

I enjoy doing statistics, and do it both for pay and fun. Recently I did a “four-part series”: critically looking at the statistics done by a certain Mark Geier, MD and David Geier, who “examined two databases”: and found a link between the removal of -m-ethylmercury-based thimerosal from vaccines and drops in reports of new autism cases. I concluded that their methodology was not valid and that correct methodology might make their case stronger. Parts I-III were a bit heavy on the statistical analysis (“in Klingon” as one blogger put it).

The tools I used cost me a total of $25. (And that $25 will be used for other things, too, so don’t worry that I shelled out $25 just to analyze one paper. I did pay $30 one time for a journal article that I needed to write a paper for another journal, but it was a critical piece of work that I couldn’t do without.) As it turns out, if you have a computer and understand statistics, you can do pretty sophisticated statistical analyses on the cheap. And if you don’t have a computer, well, get off your friend’s computer and get one! However, I will not guarantee the results of these software products, nor should you just let it rip without checking the results.

First up is the statistical package “R”: R is a lot like “S-plus”:, just free and open source. It is actively maintained and developed with a group of very smart people behind the project. The downside is that R is a programming language and has a bit of a learning curve. There are quick-start guides and some books out for the program, so there is support. I believe universities are starting to use R in their departments as well. The graphics and analysis capabilities of the program are very strong.

Second is the Java program “Datathief”: Datathief has been around for a while for the Macintosh, and I’ve used it on several occasions. Its purpose is to convert graph data into numbers that you can input into a statistical package such as R or “SAS”:, and does so very powerfully. It can handle different axis scalings (e.g. linear, logarithmic, semi-logarithmic), polar graphs, date-based and categorical axes, and so forth. It can trace curves or find points manually. It’s shareware and $25. A demo version lets you handle basic graphs.

Then there is the program “Easycalc”: This is a “Sourceforge”: project released under the GNU(GNU’s Not Un*x) license. This extremely powerful calculator is for Palm machines and includes scientific, integer, statistical, matrix, integration, solving, and graphing capabilities. It seems to have not been updated for a while, but it’s great as it is.

Finally, I’d like to give a shoutout to the program “Octave”: Octave is designed to be a free alternative to “Matlab”:, which does lots of numerical computation. The learning curve is high, but the rewards are great. I don’t use this so much anymore because R does nearly everything I need.

%d bloggers like this: