Statistics for Discovery

dc.contributor.author	Box, George
dc.date.accessioned	2014-06-09T16:44:05Z
dc.date.available	2014-06-09T16:44:05Z
dc.date.issued	2002-03
dc.identifier.uri	http://digital.library.wisc.edu/1793/69299
dc.description.abstract	In this paper the question is discussed why investigators in engineering and the physical sciences rarely use statistics. It is argued that statistics has been overly influenced by mathematical methods rather than the scientific method and consequently the subject has been greatly skewed towards testing rather than discovery.	en
dc.title	Statistics for Discovery	en
dc.type	Technical Report	en