[R] Is R good for not-professional-statistician, un-mathematical clinical researchers?
Marc Schwartz
MSchwartz at MedAnalytics.com
Thu Aug 19 16:33:02 CEST 2004
On Thu, 2004-08-19 at 01:45, Jacob Wegelin wrote:
> Alternate title: How can I persuade my students that R is for them?
>
> Alternate title: Can R replace SAS, SPSS or Stata for clinicians?
>
> I am teaching introductory statistics to twelve physicians and two veterinarians
> who have enrolled in a Mentored Clinical Research Training Program. My course is the
> first in a sequence of three. We (the instructors of this sequence) chose to teach
> R rather than some other computing environment.
>
> My (highly motivated) students have never encountered anything like R. One frankly
> asked:
>
> "Do you feel (honestly) that a group of physicians (with two vets) clinicians will
> be able to effectively use and actually understand R? If so, I will happily call this
> bookstore and order this book [Venables and Ripley] tomorrow."
>
> I am heavily biased toward R/S because I have used it since the first applied statistics
> course I took. But I would love to give these students some kind of objective information
> about the usability of R by non-statisticians--not just my own bias.
>
> Could anyone suggest any such information? Or does anyone on this list use R who is
> a clinician and not really mathematically savvy? For instance, someone who doesn't
> remember any math beyond algebra and doesn't think in terms of P(A|B)?
>
> Or have we done a disservice to our students by choosing to make them
> learn R, rather than making ourselves learn SAS, Stata or SPSS?
>
> Thank you for any ideas
>
> Jake Wegelin
A couple of questions:
1. What is the intended goal of the series of classes?
2. What are the expectations of the clinicians for themselves and what
is their likely career path?
Possible answers to the questions:
1. Provide the clinicians a reasonable (and perhaps broad) foundation of
statistical knowledge.
2. To be able to have a reasonable comprehension of statistical concepts
and methods so that in the future, as they are busy with patients
(animals for the vets) in a clinical practice, they can intelligently
interact with formally trained statisticians when engaged in clinical
research in a multi-disciplinary team environment.
If the above is close to reality, then let me suggest that you consider
Peter's book "Introductory Statistics with R" rather than MASS, at least
for the first class in the series. I cannot think of a more gentle,
broad and competent way to introduce clinicians to both statistics and R
at the same time.
If these clinicians are likely to move on to busy clinical practices, in
my experience having come out of the clinical environment, they will not
have the time to sit at a computer and grind out analyses, much less
maintain their proficiency with a programming language (R, Stata or SAS)
or the broad range of statistical methodologies that they would likely
encounter over their careers.
They will however, need to be able to sit and interact with
statisticians, bringing the significant value of their clinical training
and knowledge, to the process of designing clinical research projects
and effectively comprehend the multitude of issues in that endeavor.
They will need to have an understanding of the complex processes by
which data are collected, managed, manipulated and analyzed in the
course of obtaining the resultant analyses.
In other words, it is important that they realize that it is more than
just a "point and click" process where voila, you have logistic
regression model. They need to appreciate both the subtleties and
complexities of dealing with real world research, incomplete data, etc.
Many clinicians do not and this results in mis-matched expectations in
the future as they deal with real world situations.
There are certainly physicians who have made the decision to focus their
careers on the statistical part of the research process, forsaking any
significant clinical patient care role. They are far and few between, to
my experience, though two or three immediately come to mind. They have
also generally made the commitment to formal graduate level education in
math/statistics securing advanced degrees.
Short of that, there is typically a future dependence upon trained
statisticians, either within an academic medical environment or via
contracted services.
The above is based upon my own experience, which is largely in
sub-specialty clinical areas. Others may and perhaps will differ, based
upon their own bias.
HTH,
Marc Schwartz
More information about the R-help
mailing list