Summary: Supplementary materials for this article are available online. Please click the TAS link at http://pubs.amstat.org.
When Can One Test an Explanation? Compare and Contrast
Benford's Law and the Fuzzy CLT
David ALDOUS and Tung PHAN
Testing a proposed explanation of a statistical phenomenon
is conceptually difficult. This class segment is intended to spot-
light the issues. This article has supplementary material online.
KEY WORDS: Benford's law; Fuzzy CLT; Normal distribu-
tion for data; Teaching.
The material in this article was developed for a future class
segment in an upper-division course (Aldous 2009a) which ex-
amines critically the real-world uses of probability theory. This
article is addressed to potential instructors of such a course.
Because such a course is unusual--perhaps unique--this in-
troduction seeks to outline carefully the style and goals of the
course. The style is to treat probability as somewhat analogous
to physics--making general predictions about the real world,
so analogous to a lab course in physics, the course is a kind