as a life long cheeseparer i can confidently say that the intersecetion between cheap/free and worthwhile is not big. I am reminded of this when i survey the three brokedown 'budget lawn-mowers' in my shed. As Cormack Mccarthy put it "the worth of any game is not inherent in the game itself, but the value of that which is put at hazard." Nothing ventured, nothing gained.
R is as good as it gets on the free lunch side. PSPP is cool though, for the name alone.
@fed2 in the aughts I read all of McCarthy's works after loving "The Road". Nothing is of limits in his books. Was Blood Meridian the one where they were on a hill top urinating and weighting for it to dry in order to make a bomb or gun powder while their adversaries were approaching?
@EnochKuhlman - as I was trying to solicit some context from you, I use SAS 50%, R 45%, and Python 5%. SAS is structured and secure, R open source and relative for the medical research I conduct. Python I have dipped into for higher level machine learning (e.g., long short-term memory recurrent neural networks).
PAST is used in different research fields and is tailored particularly for palaeontology. Lots of modules; very good graph facilities. Used by scientists world-wide. Worth keeping on your computer.
JAMOVI: an free and open source alternative to the more cumbersome SPSS. Modules can be added to expand its functionalities; not so many currently available. Output tables in APA style (which can prove handy); charts and plots' quality depends on the module used. It builds (under the hood) on R, which is good. I like the integration with R (which can be executed from within Jamovi). I use JAMOVI for two study units I am currently lecturing. Not bad. I hope that the list of modules will grow in the near future. Can be used as a "bridge" to R.
I would add Orange: lot of modules here as well, including text mining. I like the fact that the analytical steps are build via widgets. Can be handy and fun for teaching. I do not really like the way data have to be fed into the program. For the rest, I think it is another software worth installing.
R of course (otherwise @Dason will start complaining): steep learning curve, but its functionalities and potential are unbounded.
JMP I really like this; its concept is different from other software as it rests on the idea of data exploration. Lots of graphical and analytical options are "hidden" into sub-menus; this allows to digg really deeper into data. I would suggest this to people who are already more than familiar with stats.
Anybody ever try JMP? I had it installed as a part of teaching undergrad students. I think it may be owned by SAS (not free, of course). I remember it having really nice features for experimental designs. And point and click things.
Yes. I actually like JMP more than any other gui system I tried. We taught the two intro business stats courses using it. Was good enough for regression and anything simple. I actually really liked it for time series too - was pretty powerful and had some really nice multilevel time series modeling capabilities.