Someone asked me recently about the difference between R and SOFA Statistics. In short, SOFA is aiming for a very different niche. One of the initial project slogans/messages is: “SOFA won’t replace sophisticated statistics systems like R, but there is a good chance it will do what you need and do it well.”
Major points of difference as I see it (open for discussion):
- R: statisticians and experienced quantitative researchers.
- SOFA: business analysts, secondary school statistics students and their teachers, social science students in the tertiary sector, experienced statisticians doing some quick exploration of data or wanting to create attractive output for a report or presentation, citizen activists wanting to use publicly available data to support their cause.
- R: statistical analysis – what are the very best tools available for understanding the data.
- SOFA: ease of use, simplicity, beautiful output (aesthetics as a value in its own right, not just a means for the communication of information)
Scope of statistical tests:
- R: everything and anything you might need
- SOFA: the main tests that most potential users of statistical analysis need. Favouring thoroughness of support for user over breadth of tests available. See the second screenshot here – http://www.sofastatistics.com/screenshots.php – for an idea of the philosophy being followed.
Of course, these are generalisations. R is not uninterested in ease-of-use or aesthetics and SOFA Statistics is intended to be extensible with plugins to allow more sophisticated analysis. But there is a difference in emphasis and there is room for both approaches as open source software increases its presence in the statistical analysis area.