Hi

On Sun, 21 Nov 2004, Christopher D. Green wrote:
> The fact of the matter is that correlation tells one virtually nothing 
> about causation at all. It is certainly not sufficient (things can be 
> correlated without having a causal connection), and it probably isn't 
> even necesary (Imagine, e.g., that A causes B, but only in context C, 
> which is rare. Since C is rare, the correlation between A and B will be 
> very low across a variety of contexts, most of which are much more 
> common than C).

Chris should be stronger here ... correlation is neither
sufficient nor necessary for causation, and equally so for both
criteria.  The problem is essentially the same in both cases ...
an uncontrolled but correlated variable could be _producing_ or
_masking_ the effect of the target variable.

Best wishes
Jim

============================================================================
James M. Clark                          (204) 786-9757
Department of Psychology                (204) 774-4134 Fax
University of Winnipeg                  4L05D
Winnipeg, Manitoba  R3B 2E9             [EMAIL PROTECTED]
CANADA                                  http://www.uwinnipeg.ca/~clark
============================================================================


---
You are currently subscribed to tips as: [email protected]
To unsubscribe send a blank email to [EMAIL PROTECTED]

Reply via email to