> My favorite original example is the correlation between number of
> annual murders in a city and number of books in its libraries.
> Students have no trouble seeing that the two are going to have a
> fairly high correlation coefficient(*), but murders don't make
> people read and books don't make people kill.
There are many such examples. My favorites involve time series, for example,
hat size and shoe size (birth to adult); hair length and weight (birth to age
1); and those with a third factor, for example, temperature and electric bill.
Stu
Garfield High School
Los Angeles
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
