[R] Re: Thanks Frank, setting graph parameters, and why social scientists don't use R

[John Maindonald]

There are answers that could and should be applied in specific 
situations.  At least in academia and in substantial research teams, 
statisticians ought to have a prominent part in many of the research 
teams.  Senior statisticians should have a prominent role in deciding 
the teams to which this applies.  why should it be ok to do combine 
high levels of chemical expertise with truly appalling statistical 
misunderstandings, to the extent that the suppose chemical insights are 
not what they appear to be?

There should be a major focus on training application area students on 
training them to understand important ideas, to recognize when they are 
out of their depth, and to work with statisticians.

There should be much more use of statisticians in the refereeing of 
published papers.  Editors need to seek advice from experienced 
statisticians (some do) on what sorts of papers are candidates for 
statistical refereeing.

Publication in an archive of the data that have been used for a paper 
could be a huge help, so that others can check whether the data really 
do support the conclusion.  Even better, as Robert Gentleman has 
argued, would/will be papers that can be processed through Sweave or 
its equivalent.

Really enlightened people (in the statistical sense) in the applied 
communities will latch onto R, as some are doing, because the 
limitations inherent in much other software so often lead to crippled 
and/or misleading analyses.  Increasingly, we can hope that it will 
become difficult for statistics to in various applied area communities 
to proceed on its merry way, ignorant of or ignoring most of what has 
happened in the mainstream statistical community in the past 20 years.

The statistical community needs to be a lot more aggressive in 
demanding adequate standards of data analysis in applied areas, at the 
same time suggesting ways in which it can work with application area 
people to improve standards.

It is also fair to comment that the situation is very uneven.  There 
are some areas where the standards are pretty reasonable, at least for 
the types of problems that typically come up in those areas.
John Maindonald.

John Maindonald             email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.


On 18 Aug 2004, Bert Gunter wrote:
So we see fairly frequently indications
of misunderstanding and confusion in using R. But the problem isn't R 
-- it's that users don't know enough statistics.

. . . .
I wish I could say I had an answer for this, but I don't have a clue. I 
do not thing it's fair to expect a mechnical engineer or psychologist 
or biologist to have the numerous math and statistical courses and 
experience in their training that would provide the base they need. For 
one thing, they don't have the time in their studies for this; for 
another, they may not have the background or interest -- they are, 
after all, mechanical engineers or biologists, not statisticians. 
Unfortunately, they could do their jobs as engineers and scientists a 
lot better if they did know more
statistics.  To me, it's a fundamental conundrum, and no one is to 
blame. It's just the reality, but it is the source for all kinds of 
frustrations on both sides of the statistical divide, which both you 
and Roger expressed in your own ways.
. . . .