STATISTICS IN LINGUISTICS

[softrat]

On Tue, 2 Feb 1999 19:33:57 -0600, Patrick C. Ryan wrote:

> I said that the coincidence of A and B was 100%. That 100% defines
>causality.

Not true! For example if C causes both A and B and nothing else does,
the correlation of A and B will be 100%. This is a very simple
example. I am afraid that Mr. Ryan does not understand mathematical
statistics as well as he apparently thinks he does.

In general correlation is _not_ the same as causality, no matter what
the correlation coefficient.

George D. Freeman IV
the softrat
mailto:softrat at pobox.com
---
"I am a man of immense learning and no culture."

[ Moderator's note:
  This is the last round on this topic.  Take any resposnes to private e-mail.
  --rma ]