PSPP is an open source statistical analysis and data mining tool. It was designed as a free alternative to IBM’s SPSS tool. PSPP is very similar to SPSS and includes most of it’s features. PSPP is capable of processing up to 1 billion cases and variables; offers both a graphical and terminal user interface and facilitates data import from spreadsheets, text files and databases. Most noteably PSPP has no license fees or expiration period. It can be run in Windows, Linux or Mac OSX environments.

This article reviews some of PSPP’s statistical tests; based on the server logs for this blog recorded in October 2010. The sample data focuses on 1503 unique visitors to the blog. The variables recorded include total hits to the blog, unique page hits, kilobytes of data downloaded, country of origin, time spent on the blog and the search term used to find the blog.

Read the rest of this entry »