Sunday, October 02, 2005

Fun with statistics

What remains of the Pearsonian revolution is the idea that the "things" of science are not the observables but the mathematical distribution functions that describe the probabilities associated with observations. Today, medical investigations use subtle mathematical models of distributions to determine the possible effects of treatments on long-term survival. Sociologists and economists use mathematical distributions to describe the behavior of human society. In the form of quantum mechanics, physicists use mathematical distributions to describe subatomic particles. No aspect of science has escaped the revolution. Some scientists claim that the use of probability distributions is a temporary stopgap and that, eventually, we will be able to find a way to return to the determinism of nineteenth-century science. Einstein's famous dictum that he did not believe that the Almighty plays dice with the universe is an example of that view. Others believe that nature is fundamentally random and that the only reality lies in distribution functions. Regardless of one's underlying philosophy, the fact remains that Pearson's ideas about distribution functions and parameters [moments] came to dominate twentieth-century science and stand triumphant on the threshold of the twenty-first century.
David Salsburg. The Lady Tasting Tea: How Statistics Revolutionized Science In The Twentieth Century, pg. 24.

0 Comments:

Post a Comment

<< Home