Re: Statistics

[Tuan V Nguyen <t.nguyen@garvan.unsw.edu.au>]

Hi Thang & friends,

>Hi ba'c Tua^'n, ba'c Huy and friends
>Du+o+'i dda^y la` tri'ch ddoa.n interview with Prof. H.Robbins, one of
>the world's leading mathematicians. He is best known for his book"
>What is mathematics ? which has been translated into several languages
>including Vietnamese.Hope you enjoy. 

	Yes, I know this gentleman. He is popular with his view 
on math in science.

>MP: In what directions is the field of statistics evolving?
>Robbins: Let"s take just the field, called biostatistics that deal
>with the application of statistics methods to human health and
>disease. The demand for trained biostatisticians is enormous, but
>there is absolutely no suplly. If I were given ten million dollars to
>spend for advancing science, I could spend it trying to produce one or
>two good biostatisticians. Statistical methods that are currently
>being used were mostly developed in England for analyzing such thing
>as agricultural experiments and industrial processes. Many of these
>techniques  are being blindly applied to situations for which they are
>not adapted. The methodology for handling important problems in
>biostatistics  does not exist. It is just begining now, its Newton or
>Einstein has yet to appear.

	Very true. As I said in one of my posts earlier, the 
challenge of applied statistics is to apply it wisely in 
biological situations and to make intelligent interpretation. 

	Some people tend to believe that when they have their 
PhD in statistics or mathematics, then they can do 
biostatistics (or anything easily); all they have to do is 
simply to insert "bio" in front of "statistics" and 
"mathematics". Unfortunately, this is too simplistic to be a 
biostatistician. A biostatistician has to know BOTH biology 
and statistics. Biological phenomena are very complex, and it 
would be foolish to say that all of them can be solved by 
math or stat. What we have been doing so far is to 
approximate them in a linear fashion.


>MP: How and where can one become trained as biostatistician?
>Robbins: A mathematically capable student who wants to become directly
>involved with problems of human welfare, should be doing
>biostatistics. Unfortunately, there are very little encouragement to
>do this, and there are very few places now to learn biostatistics. A
>mathematics department would never thing of advising anyone to study
>it. I would like to see a distinguished mathematics department in this
>country( U.S.)tell its student:" You are very capable and you could
>have a career in algebraic geometry, or whatever, but we would like 
>to encourage you to go into biostatistics.

	Also true. Pure math people look down on applied math 
people. Applied math people look down on engineers and 
scientists, which in turn look down on arts and literature 
workers. I take the view that when one is dedicated to learn 
any field from the root, then none of them can be considered 
to be easy.

	Career structure in math and stat people is another 
sticky point. Well, there is virtually none. You see, in 
medical profession, they have a very well structure, you can 
tell the seniority of a doctor by looking at his initials 
(e.g. fellowship, postgrad training etc.) In contrast, you 
can not tell the seniority of a statistical consultant. Their 
degrees do not, unfortunately, tell how suitable they are as 
a consultant.

	The situation in VN is even worse. As far as I know, 
there is no proper statistical department in ANY VNese 
university. What can we do to change this situation? Sending 
people to overseas to study is only short term solution. We 
need long term solution. Any ideas? 



>COMMENT: Va^'n dde^` nghie^n cu+'u a?nh hu+o+?ng cu?a cha^'t ddo^.c da
>cam ddo^'i vo+'i mo^i tru+o+`ng, ecology( sinh tha'i) va` su+'c khoe?
>cu?a da^n VN kho^ng the^? kho^ng co' su+. tham gia cu?a ca'c
>Vietnamese biostatisticians . How many are there such people in VN ?
>Cheers
>Thang

	Tro+`i o+i, you touch to a very sensitive subject of 
mine. As far as I know, there is very few, may be none, 
biostat people in VN. I do, however, know some VNese 
biostatisticians overseas, some of them have good track 
record in the field. 

	To me, the agent orange is a gold mine for us - 
doctors, epidemiologists, statisticians, scientists etc - to 
explore and tell the whole world about its effects and 
consequences. But, we have to do it properly, in a scientific 
manner. We can not do it in an ad hoc way as have been done 
right now. Otherwise, the data will again be owned by 
overseas scientists as it has been in the past 50 years or 
so.

	May be it is time for us to form a group and do 
something about this.

	Tuan