[R] R & learning curves. Was RE: R routines vs. MATLAB/SPSS Routines

Charles C. Berry cberry at tajo.ucsd.edu
Fri Oct 26 20:51:13 CEST 2007


On Fri, 26 Oct 2007, Ravi Varadhan wrote:

> Please pardon my non-R related response, but I couldn't resist this!
>
> I have always felt that the phrase "steep learning curve" is incorrectly
> used.  If one plots "learning" on Y-axis and effort (or time) on the X-axis,
> then the (instantaneous) slope of the learning curve for R should be
> shallower (not steeper) than that of SPSS, which simply means that it takes
> a greater effort or longer time to learn in R.  Of course, the curve would
> be steeper if the axes were to be reversed, but this is not very meaningful,
> and further it will not be a "learning" curve.  If one were to assume that
> the rate of learning is the same for R and SPSS (for an individual), but the
> amount of learning necessary to accomplish a given set of tasks is greater
> in R, then the learning curve is not steeper, but simply higher.
>
> Ravi.


 	Steep is the way to mastery. Often nothing keeps the pupil on the move but
 	his faith in his teacher, whose mastery is now beginning to dawn on him.
 	He is a living example of the inner work, and he convinces by his mere
 	presence.
 	--Eugen Herrigel
 	Zen in the Art of Archery

 	Steep is the way to mastery. Often nothing keeps the programmer
 	on the move but his faith in his language and environment for
 	statistical computing...
 	--Anonymous Manuscript
 	Zen in the Art of Statistical Software

>
> ----------------------------------------------------------------------------
> -------
>
> Ravi Varadhan, Ph.D.
>
> Assistant Professor, The Center on Aging and Health
>
> Division of Geriatric Medicine and Gerontology
>
> Johns Hopkins University
>
> Ph: (410) 502-2619
>
> Fax: (410) 614-9625
>
> Email: rvaradhan at jhmi.edu
>
> Webpage:  http://www.jhsph.edu/agingandhealth/People/Faculty/Varadhan.html
>
>
>
> ----------------------------------------------------------------------------
> --------
>
>
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
> Behalf Of Frank Thomas
> Sent: Friday, October 26, 2007 11:54 AM
> To: r-help at r-project.org
> Cc: matt.dubins at utoronto.ca
> Subject: Re: [R] R routines vs. MATLAB/SPSS Routines
>
> Some major differences between R and SPSS:
> 1/ The learning curve of R is steep and the one of SPSS is largely flat.
> A difference any student will rapidly understand.
> 2/ The user interface in R is underdeveloped, in comparison to SPSS.
> 3/ In R without loving to spend time in programming you get nothing.
> With SPSS your students will concentrate on content, not on technology.
> 4/ SPSS is so easy to use that the statistical conditions for using
> specific procedures get easily forgotten. R is more close to the
> programming side so no way to forget the foundations.
> 5/ The economic price of SPSS is really steep, you pay more than 30
> years of development. R is free, but the real price for a student is his
> or her time to learn, which can also be steep.
>
> I think, how to evaluate the differences is in part a question of the
> mindset and the work environment of the future user. If your students
> are more mathematicians, program developers, engineers, science people,
> etc. and need to tweak a procedure to single case applications you will
> have an easy public with R. If they are more of economic, social
> sciences, service industry people, and routine applications or large
> data sets will be their job SPSS, SAS, SPAD are more adapted.
>
> But this may be ground for discussion.
>
> BTW: Contrary to some ideas both R  & SPSS can be programmed and the
> algorithms for both have been published. So, no matter whether open
> source or private property you know what you do (if you want).
>
> Hope this helps,
> F. Thomas
>
>
>
> Matthew Dubins wrote:
>> Hi all,
>>
>> I've become quite enamored of R lately, and have decided to try to teach
>> some of its basics (reading in data, manipulation and classical stats
>> analyses) to my fellow grad students at the University of Toronto.  I
>> sent out a mass email and have already received some positive
>> responses.  One student, however, wanted to know what differentiates the
>> routines that R uses, from those that MATLAB and SPSS use.  In other
>> words, in what respects do R routines work faster/more efficiently/more
>> accurately than those of MATLAB/SPSS.
>>
>> I thank you in advance for any answer you can give me (or rather, the
>> inquiring student).
>>
>> Cheers,
>> Matthew Dubins
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reroducible code.
>>
>>
>>
>
>
> -- 
> ..........................................
> Dr. Frank Thomas
> FTR Internet Research
> 93110 Rosny-sous-Bois
> France
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Charles C. Berry                            (858) 534-2098
                                             Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	            UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego 92093-0901



More information about the R-help mailing list