[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Stata and biology/biomedical sciences

From   "Martin Weiss" <>
To   <>
Subject   RE: st: Stata and biology/biomedical sciences
Date   Thu, 17 Jul 2008 08:45:42 +0200

Nick says with respect to R: " The fact that it is free works..." As an
economist/business administrator, let me put that into a little bit of
perspective: having spent countless hours with R, having bought countless
(pricey) books about R, I decided to go back to Stata and to keep the R
books on the shelf for very special cases where the latest advanced methods
are not (yet) implemented in Stata. So far, I have no incident of that kind
to report. 

The word "free" is, IMHO, not appropriate for R, as opportunity costs for
learning its intricacies eat into your budget. Whether money changes hands
or not is not the relevant question. For most of the research community that
I observe, Stata is as close to ideal as it gets. Others have (rightly)
pointed to those superior data handling capabilities. They are outstanding
and remarkable indeed. Although most of that can be done with the "free" R
as well, the (reasonable) price of Stata, the list and the books by Stata
press make learning Stata almost a breeze in comparison with trying to learn
the ropes in R. 

Add up those (additional) hours you spend learning R, multiply them by your
hourly wage, chip in the frustration that comes from incompatible packages
from the R community, and the resulting sum easily exceeds the price of a
copy of Stata (with all user guides and a Stata coffee mug included :-) )

Martin Weiss

Diplom-Kaufmann Martin Weiss
Mohlstrasse 36
Room 415
72074 Tuebingen

Fon: 0049-7071-2978184




-----Original Message-----
[] On Behalf Of Nick Cox
Sent: Wednesday, July 16, 2008 7:10 PM
Subject: RE: st: Stata and biology/biomedical sciences

Maarten may mean me as the "lonely geographer", although I think there
are other geographers on this list. 

I think there are many more biostatistically-minded people on this list
and in the user community than is sometimes apparent from the mix of

But some biological disciplines (e.g. ecology) seem thinly represented
in the Stata community. 

At present there are many people actively encouraging the use of R in
such disciplines. 
To what extent followers are springing into line I can't judge. It's one
thing to write books evangelising R and another thing to get many people
using it routinely. 

I admire R and I don't think it is antithetical to Stata in any real
sense, as I have said several times on the list. 

Recently, however, someone without a Stata axe to grind, and with very
wide knowledge of the statistical computing scene, commented privately
that R just isn't quite broad and strong enough to be able to meet the
extraordinary expectations that are being placed upon it in many
quarters. The fact that it is free works in various ways and conveys
major limitations as well as advantages. 


-----Original Message-----
[] On Behalf Of Maarten buis
Sent: 16 July 2008 17:53
Subject: Re: st: Stata and biology/biomedical sciences

--- David Airey <david.airey@Vanderbilt.Edu> wrote:
> Is it my mistaken impression or is Stata becoming mostly an  
> environment used by econometricians and social scientists, whereas as
> R is used more by biologists? Or is it just that Stata is very
> popular with these groups and healthy with others, etc. The user
> meeting has no talks from biologists who use quantitative methods. I
> am not saying that Stata's capabilities prevent use by biologists
> needing stats at all, just noting the majority of talks seem to be
> from econometricians or social scientists.

There was a very strong biomed precense at the last Nordic and Baltic
Stata Users Group meeting, particularly in the satelite meeting
preceding it, and there were also a couple at the last two London
meetings. So maybe the popularity of Stata differs not only by
discipline, but also by location. Maybe the lonely geographer on this
list could comment on the latter?

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index