[R] correlation and causality examples

[Thomas Lumley]

On Sat, 15 Nov 2003, Jean lobry wrote:

> Dear All,
>
> I'am looking for examples showing that correlation does not imply
> causality, the targeted audience consists of undergraduate students
> (their first year at the university but in the BioMathStat track).
> All practicals are under R.
>

There's a nice example we use (data at
http://courses.washington.edu/b517/datasets/fev.txt
documentation at
http://courses.washington.edu/b517/datasets/fevdoc.txt
)

These are lung function (FEV1) data on children, taken at routine
checkups, and we tell people to do
a t.test comparing smokers and non-smokers.

There is a large and statistically significant difference -- the smokers
have *higher* FEV1, because they are older.

In a statistics course for graduate students in public health there are
always a few people who see the difference and forget to check the
direction...


	-thomas