Of the commonly used statistical analysis software (SPSS, STATA, R, SAS, MiniTab), most researchers prefer one over the others. So form your experience, what is your favorite? And why?
R because it's free, highly developed and can be used for practically all tasks during research (read and shape data, analysis, graphical and text output). In addition, RStudio provides a
R because it is free, powerful, has more state of the art methods available or can be gotten from other researchers, by in large if you speak linear models you are in good shape(not like SPSS that has it's own terms for many common things), many procedures not in packages , have code in the literature, e.g.see Packt books as well as my all time favorite: https://www.amazon.com/Everyone-Advanced-Analytics-Graphics-Addison-Wesley/dp/013454692X/ref=sr_1_1?keywords=lander+jared+R+for+Everyone&qid=1563828327&s=books&sr=1-1 , GREAT graphics, many sources of help and on and on. David Booth
R because it's free, highly developed and can be used for practically all tasks during research (read and shape data, analysis, graphical and text output). In addition, RStudio provides a
While a US Govt employee, starting in 1977, I only had what was available at work. I used Fortran for a number of years, and say what you will, it was labor intensive but flexible and you might be more certain of what it was doing. Also I had SPSS long ago, then BMDP, then SAS. Perhaps they are more flexible now, but I did not care much for SPSS or BMDP because it felt too much like one size-fits-all to be used for a lot of the research I did. SAS has its PROCs, which can be useful, but it is also a programming language, and though you have to be careful that it is doing what you meant it to do, it is very flexible. The guy who implemented a lot of my work at a statistical agency used SAS to set up production of Official Statistics with a great many files and tables produced. I recall him saying that he looked into R, and I think he thought it was similar, but not able to take on the huge files and complexity needed, had we been able to use it. Then I recall a large meeting at the Joint Statistical Meetings (JSM) one year where I think it was the president of SAS Inc spoke, and afterward an audience member identifying himself as a student said he "... would NEVER use SAS, because I have R."
One year at the JSM I visited different vendor booths such as Stata, etc., to see what software might also implement my methodology, and I think I recall thinking that they all would.
At any rate, there are a lot of choices, and I expect their capabilities change sometimes rapidly and sometimes slowly, but I haven't heard a lot of bad things about R. The weakness would obviously be that there is no for profit company running it so it might be a little disjoint. But the strength is that there is no for profit company running it, so the users really can drive it's development. They develop it! And I have seen that there are manuals and some structure that helps keep it less disjoint. Perhaps someone would care to comment on the availability of manuals???
Hearing about experience some people have had with other systems would be interesting to me too.
Great points have been made so far, thank you all for joining the discussion! For those who haven't indicated "why?", please share with us why you prefer a certain software.
James R Knaub it's interesting to read about your transition from one software to another since the 1970's, your experience gives a good glimpse of how they've developed through the years. Thank you!