A discussion in fall 2010 considered the extent to which purely software-related questions would be welcome on this site. It didn't really reach a conclusion, but one useful suggestion that arose is to collect a set of links to online support resources (such as user groups and list servers) for the various statistical computing platforms. I guess that would let us close some of these questions in a constructive and relatively guilt-free manner. I, for one, would like to help the people who come by with questions about SAS macro syntax or table access, even though these questions have no direct statistical interest (and would interest only small subcommunities here) and therefore ought to be closed or migrated.

Could we organize replies in the present thread by software platform? The ones of most immediate use are those that keep showing up: R, SAS, SPSS, Stata, Excel.

The (main) FAQ now links to this thread. – whuber Apr 7 '11 at 6:51

18 Answers 18


R-help, and the various R Mailing Lists or SIGs, welcome any questions (provided they conform to the Posting Guide). Answers generally come within one or two days.

Quick-R gives a gentle overview of most of the basic R syntax for people coming from SAS, SPSS, or Stata. Stack Overflow also provides strong support for R questions. Additionally, rseek.org provides a custom Google search that facilitates queries related to R code, packages, articles, etc., while crantastic.org features useful reviews of current packages.

If you're looking for visualization ideas, visit the R Graph gallery and the Learn R blog, both of which feature a wide variety of plots and the accompanying code. The R Graphical Manual also provides a visualization of all CRAN R package example plots, and is searchable by topic. The Cookbook for R page provides multiple examples and recipes for plotting data (mostly using ggplot2) plus some additional information on using R.

To add the excellent resources list above, I (@michelle) have found the R Tutorial web site to be helpful. Also I have R bloggers as a feed. That has lots of useful posts from various bloggers and is an excellent way to keep up with new packages and new ways of using existing packages. If you're coming from SAS or SPSS, check out R for SAS and SPSS Users; there is a book with more information in it. An equivalent book for Stata users coming to R is R for Stata Users. Nice introductory tutorial can be found on R Tutorials page - it covers introduction to basics of R, using statistical tools such as t-tests, ANOVA, regression and other topics.

There are also some resources listed on our site here: Resources for learning R, and on our R tag wiki.

For learning on more advanced topics in R programming the best resource that is available online is Advanced R site by Hadley Wickham. It is an online version of book under the same title. Another resource covering programming issues is The R Inferno by Patrick Burns available as pdf file. Those two cover topics that are negligible to most people that use R for statistics but can be crucial if you do actual programming in R and can be helpful in understanding how R works 'under the hood'. If trying to understand better how some R function works, you can always check their source code as R is open-source.

The link to the RGraph Gallery appears to be dead. – and0rsk Oct 1 '15 at 1:12


Although this is not a statistical package per se, it has extensive statistical capabilities.

In addition, people who prefer Python for scripting, but would like access to R's wide-ranging statistical capabilities can call R from Python with rpy2 (see also: A Slug's Guide to Python).

NumPy and related questions get jummped on pretty quickly on Stack Overflow ... As a python programmer, I can actually recommend Stack Overflow as a default. – Tritium21 Sep 25 '15 at 4:50


Statalist is the place to go with questions about how to do things in any version of Stata. It is very active: Stata developers from StataCorp and many experienced users are leading members. Questions cover basic Stata use, Stata programming, and statistical practice.

Talk Stats occasionally includes Stata questions. Beginner questions are common. The number of experienced users lurking there is far smaller than on Statalist.

Stata Forum.De is dedicated to Stata and is conducted in German.

The official FAQs are extremely useful as well. The help files for specific commands are available online; for direct access, form the url as http://www.stata.com/help.cgi? appended with the command name, like regress: http://www.stata.com/help.cgi?regress. The full pdf documentation is also available online, e.g., the User Guide and the Reference manuals are usually a good place to start for general commands.

StataCorp also maintains a YouTube training channel and offers registered users free technical support via e-mail.

ATS/UCLA has an entire section dedicated to Stata. Start from there with the learning modules, the FAQs or the "links by topic."

@lejohn Thanks for adding the links. Welcome to our site! – whuber Aug 8 '11 at 19:40
I'm surprised no one mentioned Stata's official command documentation, which can be found online. Normally a google search for the command will bring it up in the first few results. To narrow the search, you can add "site:www.stata.com/help.cgi" after searching for the name of the command, e.g. regress site:www.stata.com/help.cgi. – Ricardo Altamirano Nov 7 '12 at 19:40
The pdf documentation that comes bundled with Stata is also quite excellent and well-written. Some of these manuals can even be found online. Lots of examples, formulas, and syntax. For some reason, many people seem to be unaware of their existence, even when they have used Stata for years. The link at the very top of Stata's documentation that you get when you type -help XXX- takes you there. – Dimitriy V. Masterov Apr 13 '13 at 3:48
As from Stata 13, the entire manuals are on-line and accessible to all. – Nick Cox Oct 25 '13 at 23:59
Typing -help XXX- into an internet search engine usually spits out the Stata help file as the first or second search result (at least for Google and Yahoo, anyway) – marquisdecarabas Apr 13 at 5:08


SAS questions do get asked and answered on StackOverflow; SAS also runs a community forum, which is very active.

The proceedings of SUGI are a tremendous resource. Also invaluable if you need material for a corny SAS stand-up comedy routine.

The UCLA Academic Technology Services provide really fantastic resources for SAS.

Another resource is SAS-L (site has archives and information on how to join).

The online documentation at the SAS support page is a great resource. It includes the SAS User's Guide which contains quite detailed information on SAS procedures, including syntax, theoretical details, and examples. As an example, here's the page for proc glm.

Lex Jansen's site indexes not only all of SAS Global Forum (formerly SUGI) but the regional meetings as well.

May I suggest we remove the reference to runsubmit.com, or at least notate it that the community has been dead for the better part of the year now? That community is basically just spam now, and I can't think of the last time I found a worthwhile answer on it. – Joe Jul 21 '14 at 15:47
@Joe Thanks for the suggestion - done! – Matt Parker Jul 21 '14 at 20:05
The link to 'various SAS blogs' via Chris Hemedinger's blog is also no longer usable. Google reader no longer exists, so there's no way of getting to the list. – Rob Penridge Nov 12 '15 at 16:48


Several forums devoted to SPSS software usage are;

SPSSX and the google group forum receive a fairly wide variety of data manipulation questions and questions related to statistical analysis. Developer central is sometimes a more appropriate place to go if you are doing anything not related to actions that can be accomplished through traditional syntax or the GUI (e.g. scripting in python or vba).

Make sure to check out the SPSS tag wiki for a more complete list of resources, many of which are freely accessible online.



MATLAB (MATrix LABoratory) is a multi-paradigm numerical computing environment and fourth-generation programming language. It is developed by MathWorks. MATLAB has a free, open-source counterpart named Octave that is distributed on GNU-GPL license and offers access to a subset of MATLAB's original functionality.

MATLAB questions get routinely answered in SO. In addition to that one can check:



Julia is a new language with MATLAB-like syntax but Lisp-like semantics and a Lisp-style macro language. Julia has growing capabilities for statistics, and, its main advantage, is blazing fast! To learn about Julia start with http://julialang.org/ and, especially, http://julialang.org/community/.

Julia has several online discussion groups including julia-users and julia-stats.



Stan is an open-source, probabilistic programming language implementing full Bayesian statistical inference and penalized maximum likelihood estimation.

Stan has very good and detailed manual and active Stan users mailing list.





JAGS is Just Another Gibbs Sampler. It is a program for analysis of Bayesian hierarchical models using Markov Chain Monte Carlo (MCMC) simulation not wholly unlike BUGS. JAGS was written with three aims in mind:

  • To have a cross-platform engine for the BUGS language
  • To be extensible, allowing users to write their own functions, distributions and samplers.
  • To be a plaftorm for experimentation with ideas in Bayesian modelling

(source: http://mcmc-jags.sourceforge.net/)

Online forum for JAGS users can be found on JAGS sourceforge page.



Learn more about Minitab on Wikipedia



Wizard is desktop statistics and data visualization package for Mac OS X. The primary support channel is the Wizard User Group hosted on Google Groups.

Disclosure: I am the developer of the software.



JMP is a desktop statistical exploration tool from SAS.

The primary source of online support for JMP is the JMP User Community site which hosts discussion forums and a file exchange for add-ins and data sets.

Other online resources include JMP documentation, a semi-technical JMP blog and weekly live webcasts.

Too bad this isn't nested. JMP script (JSL) is very particular and demanding. It might qualify as its own sub-heading. – EngrStudent Jan 20 '15 at 21:08


Weka is a collection of machine learning algorithms for data mining tasks. The algorithms can either be applied directly to a dataset or called from your own Java code. Weka contains tools for data pre-processing, classification, regression, clustering, association rules, and visualization. It is also well-suited for developing new machine learning schemes.

A mailing list, forum, & IRC channel are listed at the Weka site under Getting Help



Fortran is one of the main languages in which statistical algorithms have been coded.

Comp.lang.fortran is an active Usenet group.

Stack Overflow often has Fortran questions.

The Open Directory has links to many Fortran codes in statistics and econometrics.


Taverna Workbench

It is a open source tool that has been more used for helping in data mining recently. In other words, you can create pieces of the workflow to code extracting data from different databases, analyze them, do some processing and it also supports output in R.

"Taverna is an open source and domain-independent Workflow Management System – a suite of tools used to design and execute scientific workflows and aid in silica experimentation".



A fortran based non-linear mixed effects modelling software with very powerful algorithms including ODE solvers. It is a commercial software made by ICON plc that is widely used in the pharmacometrics community. It has a steep learning curve but once you get used to it, you'll probably not need to use another non-linear mixed effects modelling software. It needs to be used alongside other statistical tools like R to analyse and visualise modelling results. It can also be used alongside PsN for automated covariate analysis and visual predictive checks, simulation, etc. For more on NONMEM, visit ICON plc.

There is a NONMEM user group dedicated to users of NONMEM here.

This short tutorial gives a very concise introduction to pharmacometrics and non-linear mixed effects modelling.



ROOT is a general purpose data-analysis framework written in C++. It is the de facto choice for any kind of analysis in the particle physics community, although it is not limited to that community. It provides many specialised functionality through libraries, e.g.:

  • Minuit is a minimisation library original written in Fortran, now reimplemented in C++,
  • RooFit is a fitting framework focused on maximum likelihood fitting,
  • RooStats is a statistics package built around RooFit used to provide tools for statistical significance calculations, limit setting, etc,
  • TMVA is a multivariate analysis library which implements numerous machine learning algorithms like neural networks, decision trees, etc.

ROOT provides language bindings in other programming languages like, Python, Go, and Ruby. It also provides various I/O facilites to handle very large datasets (multi-gigabyte), as well as visualisation facilities.

The package is developed as a joint effort by Fermilab and CERN. The homepage has extensive documentation that include getting started tutorials, code examples, as well as an extensive reference manual. To get help, one can go to the web forum or the mailing list. Unfortunately to signup for the mailing list, one needs to get a lightweight CERN account (no restrictions though, anyone can get it).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .