A new open source option that has only recently become available is SOFA Statistics (
http://www.sofastatistics.com). The emphasis is on ease of use, learn as you go, and beautiful output. It is not aimed at the same niche as R, SAS or SPSS but it should be useful for many users and purposes. [Disclosure - I am the lead developer]. SOFA is cross-platform and will work anywhere Python will run. The deb and Windows packages can be found at sourceforge (
https://sourceforge.net/projects/sofastatistics/). SOFA Statistics allows the user to connect directly to (as opposed to import) MySQL, SQLite, MS Access, and MS SQL Server, and data can be imported from csv or MS Excel. In addition to nested tables with row and col %'s, sd, mean, median etc there are Independent and Paired t-tests, Mann Whitney U, Wilcoxon Signed Ranks, Pearson's Chi Square, one-way ANOVAs, Kruskal Wallis H, Spearman's and Pearson's R. In addition to the GUI interface, processes can be automated using Python - whether from exported scripts or manually crafted ones. SOFA Statistics is released under the AGPL and is under active development.
Bookmarks