How about pspp? It's an open source alternative to SPSS. I don't know how active the development is, but perhaps it is worth the try: http://www.gnu.org/software/pspp/

Take a look at the Wikipedia article about statistical softwares at:

http://en.wikipedia.org/wiki/List_of_statistical_packages

Frank Siegmund's suggestion PAST looks a good alternative

I think past is a good alternative.

PAST does seem to be a good alternative. Thank you Frank.

Try orange

http://orange.biolab.si/

For some people using R through Rstudio makes a lot of difference, have you tried?

http://rstudio.org/

For more technical users, Julia is one of the new kids on the block:

http://julialang.org/

You could have a look at pyvttbl which is coded in Python.

https://code.google.com/p/pyvttbl/

Another alternative (also in Python):

http://statsmodels.sourceforge.net/

There are plenty of other Python packages for data analysis, but to suggest the right one we would need to know which methods you would like to use.

R Studio is a great tool totally changing how you work with R language. But it takes time to get acquainted. While Orange is similar, the modules tou use are easier, but again con is there are limited module preset available, seeming like it is a general statistics software rather than bio oriented, similar to R Studio.

Matlab is a general programming language that works again in modular way and can be used for statistics and other porpuses. Modules may or may not be available for bio oriented statistics.

Have heard lot about SAS and SPSS. Also available are PSPP (alternative to SPSS) and WPS (Alternative to SAS). Both support programming and have been known for its bio-oriented use. But none of them have been tried yet. JMP is also a good software to give a try but it seems to be Bio-oriented and doesnot support any programming

Special mention might be Statistica but little less is known to me about it.

Limited functionality but special bio-oriented statistical softwares are MEV from Tiger and GraphPad. Both software are not programming supported hence have limited functionality and functions are predefined.

SYSTAT has a free student version called MYSTAT. It is based on the previous release of SYSTAT, version 12, and is limited to working with data sets of 50 or fewer variables and lacks the statistics that are considered advanced, but actually includes more than SPSS includes without buying a bunch of modules. You can get more information www.systat.com

What's bad about a steep learning curve? I suggest to use R with R-Studio as others suggested. The learn effect you can improve with intensivly use of the web-ressources which is for R very, very good.

I also agree with Fabio,

spend a month to practice on R, then you will benefit all rest of your carrier

... and I would avoid drag and drop or menu-tools. You'll never know what happend behind. Drag and drop / menu are useful at the beginning if you want to learn the syntax of the code but it is dangerous since it induces not to learn the programming language. I think basic knowledge of one programming language whatever it is should be in the toolbox of an academic (whatever kind humans or natural sciences) like basic English is for me. By the way: Psychology student at the university of Zurich changed from SPSS to R.

Hi all,

I totally agree with people suggesting the use of R. If you want, there is an internet course on coursera you may follow right now:

https://www.coursera.org/course/compdata

It is for beginners and it is free. It is 4 weeks long, it started one week ago, but with a little effort you can catch up quickly and I am sure you will love it !

Tito

If the issue is a steep learning curve, you may be interested in this free online training resource for statistical modelling – it starts fairly gently with single and multiple regression and the develops into discrete outcome modelling, multilevel modelling and more modules are planned on such things as missing data and longitudinal data analysis. It has been produced by the Centre for Multilevel Modelling at the University of Bristol. There are MLwiN, R and Stata versions of each practical; comparing across the practicals will help users to become familiar with the other package.

http://www.bristol.ac.uk/cmm/learning/course.html

As one participant said "A great course. Just what I needed. Explained everything in detail. The tests were particularly good as they highlighted aspects you thought you understood but hadn't really grasped”)

Try MicrOsiris. the interface is simple enough and has capabilities similar to SAS, MYSTAT, SPSS except for graphs. It takes up a lot less space and is very fast, too.

Hi, again, Arnab, I have been thinking a bit more about this, and I want to reinforce Fabio's comment. Drag and drop, cut and paste, are very powerful tools, but they are also worrisome. As an analyst, I try to have the data come to me full processed (it rarely is), and the front end work is sometimes done in Excel. That often frightens me, because I have seen over the years (and it's been many) duplicated data, data pasted off by one or more columns, lost data--you name it. Also, there is often stuff in the midst of the data, such as duplicated labels, where I can only accept and analyze columns and rows of numbers. Although, I am often using specialized packages, such as HLM (Scientific Software International), my main package for handling data is Systat v13. You can cut an paste data, work with drop downs, and write code. Using drop downs generates code that you save. Of those choices the one I hardly ever use is the cut and paste. I doubt I used it once in 2012. If for some reason, it might seem like the best solution, I'd make a notation in a program of exactly what I did. So my point is that although the drag an drop feature of a spreadsheet (the word we used before the word Excel became a synonym) may give a certain ease of use, to me, and I think the others who have suggested various packages, the ability in a database or a general purpose stats packages (such as SAS, SPSS, STATA) to document what you are doing is worth the cost in something of a learning curve, which to my way of thinking, for a program with a decent GUI, such as SPSS of SYSTAT (or free MYSTAT version) is worth it. Although I checked out the site of MicrOsiris and can verify it has excellent features and works with huge data sets (MYSTAT can only do 100 variables), I haven't tried it yet, but if MYSTAT can't do the job, I would definitely give it a try. I definitely will because it runs IVEware, which is a great program for doing multiple imputations of missing values.

Well... I've read some comments and it's amazing to find new statistic software. On the other hand... in reference with my experience, I recommend you to use R, it's a amazing tool, which you can develop whatever function that you need, without depending if the software has a specifically function... and... you can also try Statistica... in my opinion is a perfect software, although some functions are necessary to add.

There are many free statistics software package. Most of them are listed here

http://gsociology.icaap.org/methods/soft.html

or

http://statpages.org/javasta2.html

I reviewed some of them, listing my review on the gsociology page. I also like Neal van Eck's Microsiris, possibly because it did a good job of handling my strange data.

R also has many graphical interfaces, as many mentioned, and they are listed on my (gsociology) page, along with some sites that review the various GUIs.

I found R with the R-commander interface (Rcmdr) is fairly user-friendly with a pull-down menu system like SPSS (and wiIl even read SPSS data files).

JASP is a free software for data analysis (frequentist and bayesian) with a graphical user interface.

https://jasp-stats.org/

Microsiris