From: hrubin@odds.stat.purdue.edu (Herman Rubin) Subject: Re: Would you consider statistics as science? Date: 5 Nov 2000 08:58:17 -0500 Newsgroups: sci.math Summary: [missing] In article <3A04D249.47086D12@brunel.ac.uk>, Johannes H Andersen wrote: >Yes, I would consider statistics as a science. Ronald A. Fisher did >pioneering work on statistical methods. He was an controversial >figure and not always liked, but nevertheless he did a huge amount >of theoretical work and introduced hypothesis testing. Statistics >as a science should not be mixed up with the different applications >by people who uses statistical recipes rather than having a deep >understanding of the subject. I agree with your last sentence, but I do not believe that you have all the rest correct. Hypothesis testing goes back at least to the 18th century, long before Fisher was born. Fisher developed SOME statistical methods, but these were of the type of recipes, rather than fundamentally new ones. He did derive some distributions, and introduced the idea of sufficiency, but got it wrong; Neyman corrected it. However, he could not handle the key idea of decision theory, which is that all consequences need to be balanced, and this balance throws out many of his dicta. This still does not answer the question. Statistics should be considered as the theory of decision making under uncertainty about the state of nature. As such, it underlies science, but is not science, any more than mathematics is. >Johannes >jsedsg@my-deja.com wrote: >> What is you view? >> Sent via Deja.com http://www.deja.com/ >> Before you buy. -- This address is for information only. I do not claim that these views are those of the Statistics Department or of Purdue University. Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399 hrubin@stat.purdue.edu Phone: (765)494-6054 FAX: (765)494-0558