December 1st, 2005, 7:19 am
A recent survey conducted by Scientist magazine chose Weizmann Institute of Science as the top-ranking multidisciplinary research institution in the eastern hemisphere. In an interview following the poll (of over 2600 academics), several scientists at the institute not only express their satisfaction at work, from their environment but also reveal that intriguing proposals of positions and grants from commercial and prestigious institutions from the US in particular were rejected in favor of working in small labs within a mere 300-acers landscaped campus. Why? Much of this commitment attribute to the working environment and the conditions which are made accessible to the scientists, a surrounding that enable them to use their full potential for the benefit of humanity. By just glancing at one usual research project taken place in the institute, one would see staff from different disciplinary backgrounds join in team and working together. One could be from the faculty of biology, other from biochemistry, and another from mathematic and computer science. This sort of interrelation between scientists from different fields of science has its merit of better vision about what is being done – which in turn energies the participants and so stimulates actions. There is no pressure to put forward findings on published journals for the reason of mere honor as is customary in other institutions. Now the Collector asked:Can academic success stories be explained by pure randomness?? My point is academics typically try to explain successful traders versus unsuccessful traders as a random walk, if 10,000 monkeys select stocks randomly some of them will randomly come up with super returns simply due to randomness, these Monkeys (=Traders) will themselves of course think about themselves as geniuses, even if they simply are fooled by randomness. But what about academics, what if 10,000 chimpanzees (Professors and other academics) over some years publish hundreds of thousands of papers, then by pure randomness some of these papers will be better than others, and some could even by randomness be useful in practice. Are Super Chimpanzies (=Star Professors) simply a result of Fooled by Randomness as well?One of the most common errors in statistical procedures is measuring the wrong variables or using inappropriate or inefficient statistical methods. To my mind however, the most serious error lies in letting statistical procedure make decisions for the measurer himself. Methods of discrimination of indirect variable they as environment, teamwork, work pressure, and liberation of ideas remind behind the veil of ignorance, and hence the reason of why the cliché of pure randomness is praised. Yes, there are reasonable fears from falling into traps of Type I and Type II errors. But still, academic researchers will have to discover means to discriminate facts from numbers, because unlike particle or objects there are people behind the numbers. So it’s not surprising that over 50% of all medical and life science research papers are statistically incorrect, and, worse, misleading.
Last edited by
bashirf on November 30th, 2005, 11:00 pm, edited 1 time in total.