Excel is so very basic; SigmaPlot and SPSS ok, but again no support for simple non-parametric plots, for eg. Box and Whisker. Wondering what others are using?
Hi Felix, short answer: R. It's free, very flexible, good plotting possibilities, very dynamic with a growing user base and transparent developments. Box-Whisker plots are not a problem at all. In R, there are many packages in which commands are organized according to a subject etc., so you will find many packages with which to make them. Yes, it is command line-based (except for yet another package which offers a modest user interface), but there are so many resources out there already, many free to download, plus extensive searchable email list, that doing something like a box plot is no problem at all. As far as I know, you can basically do everything in R that you do with SigmaPlot, SPSS, Excel, and in a much more flexible, and often more correct, way. Best, Susanne
Hi Felix, short answer: R. It's free, very flexible, good plotting possibilities, very dynamic with a growing user base and transparent developments. Box-Whisker plots are not a problem at all. In R, there are many packages in which commands are organized according to a subject etc., so you will find many packages with which to make them. Yes, it is command line-based (except for yet another package which offers a modest user interface), but there are so many resources out there already, many free to download, plus extensive searchable email list, that doing something like a box plot is no problem at all. As far as I know, you can basically do everything in R that you do with SigmaPlot, SPSS, Excel, and in a much more flexible, and often more correct, way. Best, Susanne
R is a really interesting option, although I am just learning how to use it and it is (so far) a little complicated. But, it is free and its use is growing really fast.
I am using GraphPad Prism at the moment; it is easy to use, gives nice graphics, but it is quite expensive.
As a free software, I love using R. It's quite flexible, Plots are not much problem. If you are a new user or have little knowledge on it, you can get help any time using google. There are many packages available and also there are a lot of users using it. So I believe for every problem, you can get help on the net.
R is the best, without any question, so I do almost everything with it - except cooking coffee. But if you don't like to become a programmer, you may give PAST (palaeontological statistics ) a try. It is a free and very handy software that can be downloaded from:
Although there is no doubt that R is the best, if you don't have the time to learn it, I find JMP (from SAS, http://www.jmp.com/) very good and easy to use.
Agree partly with Thomas Mohr on R (taken 2 courses in R with little success, it is more difficult in my mind than the DOS version of SPSS/PC of previous millennium, no experience in GP P). Started using SPSS since their manuals were possible to read.
For the simple box'n whiskers plots they are readily available in the SPSS "Examine" procedure. If such plots are indeed non-parametric or not is not clear to me, but they do use the absolute values rather than ranks?
I would recommend R. In the second place, somebody may wish to use OpenStat for all statistical tools ready to use in which menus and dialogs resemble those of SPSS.
I would recommend R. In the second place, somebody may wish to use OpenStat for all statistical tools ready to use in which menus and dialogs resemble those of SPSS.