R help - Feb 2004 - R Included with Open Infrastructure for Outcomes (OIO) system

Hi all,

I came across this article on LinuxMedNews (http://www.linuxmednews.com)
this morning:

http://www.linuxmednews.com/linuxmednews/1076524250/index_html

This refers to an integrated data management and analysis system (OIO),
which includes R and utilizes the RSessionDA package (Greg Warnes).

More information is available here for those interested:

http://zope.org/Members/aho/Open_Infrastructure_for_Outcomes/LiveOIO/liveoio/LiveOIO_1_0_8

There is one statement pertaining to R on the above page that I might
challenge:

"Unfortunately, it is a bit intimidating to use except by expert
statisticians."

If I can use it, anybody can...

;-)

Regards,

Marc Schwartz

Marc Schwartz wrote:
>[snip]
>
>There is one statement pertaining to R on the above page that I might
>challenge:
>
>"Unfortunately, it is a bit intimidating to use except by expert
>statisticians."
>
>If I can use it, anybody can...
>
>;-)
>
>Regards,
>
>Marc Schwartz
>  
>
I think this perception is our biggest challenge in spreading R. 
 Spreadsheets and
word processors also were thought to be intimidating.  Indeed if we 
needed to
know all of what R does to use it, then anyone would be more than a bit
intimidated, I'm sure.

There are two counter-attacks that I see:

1)  Make it clear that you can do useful work in R without knowing 
everything.
This can be facilitated by clear instructions on how to do some 
(subject-specific)
tasks.  The clarity of the instructions can be called into question, but an
example of what I have in mind is:
http://www.burns-stat.com/pages/Techan/r_for_tac.html

2)  Highlight what is central, and make that as simple and intuitive as 
possible.
I think the new division into a small "base" is a good step in that 
direction.

Patrick Burns

Burns Statistics
patrick at burns-stat.com
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")
>  
>

Marc Schwartz <MSchwartz at medanalytics.com> writes:
> Hi all,
>
> I came across this article on LinuxMedNews (http://www.linuxmednews.com)
> this morning:
>
> http://www.linuxmednews.com/linuxmednews/1076524250/index_html
>
> This refers to an integrated data management and analysis system (OIO),
> which includes R and utilizes the RSessionDA package (Greg Warnes).
>
> More information is available here for those interested:
>
>
http://zope.org/Members/aho/Open_Infrastructure_for_Outcomes/LiveOIO/liveoio/LiveOIO_1_0_8
>
> There is one statement pertaining to R on the above page that I might
> challenge:
>
> "Unfortunately, it is a bit intimidating to use except by expert
> statisticians."
>
> If I can use it, anybody can...
Consider the application area of OIO, though, which is the collection
of CRF's (patient-based case report forms for hospital as well as
possibly clinical-trials related activities).  This is a fair caution,
I think, for that general population of researchers and hospital IT
folks (and clinician/IT types).  

Issues include use of R as well as application of medium to advanced
statistical methods.

On the other hand, it's a nice approach, and one which we are working
on to leverage for similar projects.

best,
-tony

-- 
rossini at u.washington.edu            http://www.analytics.washington.edu/ 
Biomedical and Health Informatics   University of Washington
Biostatistics, SCHARP/HVTN          Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC  (M/W): 206-667-7025 FAX=206-667-4812 | use Email

CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}}