Anyone use any of these free statistical packages? Good/bad experiences?
Thanks
Gene Shackman
All Answers (99) Show full discussion
-
Also i use weka...a nice software -
R and only R for programming, writing my own algorithm and of course my love Statistics!
- The BEST Statistical software -
Is there any free software for Enzyme kinetics analysis... -
R is the best ... i have been using this... -
R is the best open-source statistical software around. It provides the most comprehensive set of features and there is a large repository of open-source packages targeted to specific fields (biostats, social sciences) that can easily be imported into the base R package. There is a large community of users to help you along and all the code is always available to adapt methods for your specific needs.
That being said, there is quite a steep learning curve since it's more of a programming language and there is no full featured GUI like a JMP, SPSS or Minitab. However, if you are going to be doing regular statistics work in a particular field and publishing research in academic journals it is worth learning.
Here are some resouces to get you started:
-- RStudio is an open-source IDE (development environment) that can help you get started and make the programming part less intimidating: http://www.rstudio.org
-- Cookbook for R is a nice introduction: http://wiki.stdout.org/rcookbook/
-- There is a large R support community at StackOverflow with a large knowledgebase and R experts provide fast support for any level of questions: http://stackoverflow.com/questions/tagged/r -
The "R Cookbook" is definitely useful, but I would recommend "R in Action" for beginners, for me "R Cookbook" is more useful as a reference, and to refresh forgotten knowledge -
In my opinion, the evaluation of a software has to be done from a user perspective i.e., depending on the application/use od the software. For methodological purposes, R is as far the best. This is because it considers mathematical objects (matrices, etc) as what they actually are . If you are not interested in methodology but just in application, I would suggest SPSS for simple models and descriptives (nice tables) and STATA or SAS for more complex modeling.
-
Friends
can i do population diversity analysis using R -
>Edwin Raj • UPASI Tea Research Foundation Tea Research Institute
>Friends
>can i do population diversity analysis using R
If you meant an analysis of genetic or morphological variation in a series of populations, then yes; hdpg {ade4} could be used for this. -
I have R for the last 4 yrs. Its flexible and reliable in all types of analysis.
-
Kind attn. Gaétan Moreau
Thanks for the same, im looking into it -
I agree withVictor that using R would be more useful for researchers and academics but for simple analysis SPSS would be fine with easy graphical user interface and also STATA and SAS. -
I also used R for the analysis of field data on arthropods in biosafety research on genetically engineered maize. It is highly flexible, with lots of possibilities including graphics. -
Although R is powerful package, but it seriously needs GUI to get popular. -
It would really be wonderful if we can have R with GUI which would be very useful for beginers and early career researchers.
-
In fact, there are many GUI for R available!
http://www.linuxlinks.com/article/20110306113701179/GUIsforR.html -
@Sebastian Raschka, Thanks a lot for providing the link to GUI for R. I will definitely download the R Studio and R Commander which looks very interesting.
-
You are welcome. The choice of GUI also depends on you operating system. I once tried a GUI on Linux, and it was really easy to set up. The name was Rkward if I remember correctly. Although, GUIs are always more or less limited in comparison to writing the code yourself, Rkward was really well done and you could do a lot of the simpler statistical analyses in a very quick and intuitive manner.
-
Thanks to all -
I completely agree, R with Rstudio -
I agree with Luz.....R is best as of now.
-
http://www.freestatistics.info/stat.php
This site has many software details to choose from. -
Thank you, Arvind. That site is very helpful.
-
Ya sir, Its a useful package for me and all emerging research group. -
R also has an Excel version called R thru Excel. This installs R Commander and also Excel add ins. You can perform most of your analysis just by pointing and clicking. You can also simultaneously program in R Commander and run the analysis. The added advantage with this is that the script for all the analysis you run is given in the R Commander window and you can also modify it. This is also a way to learn programming in R.
-
@Raveendhara, Can you lease tell me how to configure server because when i install its showing that no server found. so the addins remain hidden.
-
@Edwin, You need to chose a CRAN server nearest to you based on your location.
-
Tanagra is a free software of data mining and statistical analysis with an interface like Statistica software.
it was developed in a French university but language is english.
http://eric.univ-lyon2.fr/~ricco/tanagra/fr/tanagra.html -
As every one I use R, but if you don't (and want to use something simpler)
The Spread sheet "GNUmerics" comes with a statistical menu interesting including parametric and non parametric test... I havent used it for a while, so I don't know how stats has been add in.... Anyway it may be usefull and deserve attention.... and if it doesn't do anything you wanted.... at least it is a good spreadsheet, quicker than Exell :p -
In my teaching, i advocate MyStat, the free student version of SYSTAT. It is very good for most statistics, even logistic regression. The only limit is max 100 variables. -
I think one has really to distinguish between studying statistics and employing statistics for research. As a student whose goal it is to understand statistical concepts simple software is just fine. I liked Minitab for example, I got a 1 year license for students for ~20 Dollar. But if you are really employing statistical software for your research work you probably end up doing certain calculations over and over again where it might be worthwhile to use a commandline tool that is scriptable for the sake of automation.
-
Free and commercial software yield different factor solutions. This a matter of personal interest. -
R is a good option but if you want a easy GUI software PSPP is a good option
-
There are a lot of GUIs for R out there, too. So If you are scared by R's command line you can check those out (http://www.linuxlinks.com/article/20110306113701179/GUIsforR.html) to get started.
-
Let me add another vote for MYSTAT. It is pretty much the previous release (version 12) of SYSTAT, with a limitation of 100 varialbes and some of the high-end statistics deleted, but still more stats than come in the SPSS base. You can find it at www.systat.com -
Erik, MicrOsiris is much smaller, faster than MYSTAT (not really relevant unless using many cases, though) and is unlimited in number of variables and cases. It is fairly complete in statistics. (www.microsiris.com).
As an aside, the free Statistical Decision Tree is useful for finding statistics appropriate to the data and where to fine them in a few of the common stat packages. -
Erik, I should add that MicroSiris gives you access to IVEware, which is a great (free) package for multiple imputation of missing data. Bob
-
I have been using R for the last 11 years. It's powerful, flexible and adaptable to most any statistical questions. Its learning curve is not as easy to climb, but is very worthwhile. On the web are available numerous manuals and tutorials. The number of books with examples in R is also quite large and mailing lists are very helpful. I use R Studio as GUI. -
I agree with Antonio. Using R Studio helps a great deal. Also, if you want to get into R, there are a number of interest groups which meet regularly and can help with learning. There are also a number of free webinars about it and a number of on-line courses. Yes, it has a step learning curve, but persistence pays.
-
I use R and I am an R supporter. It is hard at the beginning but you can manage it efficiently little by little. -
I would add, as others have, the using R via RStudio is very helpful. It greatly expedites the processes.
One book that I have found very useful is Phil Spector's "Data Manipulation with R." It provides excellent advice about how to best use data with the R language. Although it's wording can sometimes be a bit murky, it is an excellent source for using data with R. -
Yes I use R and it is an excellent software. -
R is the best! No doubts about it! Some years ago, I had some other softwares in my cpu just to export data to txt format and use it in R, because clients usually don't even know R. But even this is not necessary now a days, because R have packages to import data from almost all software (like package "foreign").
R is all that you need! -
I second Sebastian Raschka...The book "R in Action" for beginners. Also I use Gnumeric, GNU PSPP, R studio and Weka
-
For OLS regression and a few other fancy models, Gretl is a good, and fairly robust programme that I've used for free. -
If you want free statistical software R is the best option. However, I found it easier to learn and use EPI Info, SPSS and also STATA. But once you master R, it is a very powerful software.
-
I would recommend SPSS (It is simple to learn) AND SAS....
Popular Answers