Correlation and contingency

Correlation = knowing A helps you guess B
o
Gender and height

Correlation does not imply causation
o
Parents – if you are like one of your parents (eye color), then you might
think that your father caused
your
blue eyes, but it was actually the gene
for blue eeyes that did it

Sometimes correlation
does
imply causation
o
Smokers were more like to get lung cancer
o
Amount of studying and the grades you get – there is a correlation

contingency – if you manipulate A and B and observe its affect on another
variable
o
This is how we test for causation, if modifying one modifies the other
o
Astronomy – you can’t find out what happens if you blow up a star
Contingency implies correlation, if A implies B
≠
if B implies A
Correlated, if A implies B, then B implies A
A perfect relationship has a correlation of 1
 height in inches and height in feet = 1 (since units don’t matter)
 height in inches, and distance from the ceiling in inches = 1
Illusory correlation

when people have theories about what the correlations ought to be, those theories
impose themselves on the actual events

make correlations when they are not there, and misperceive ones that are there
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '11
 Mellers
 Correlation and dependence, Correlation does not imply causation, Eye color, 43%, 58%

Click to edit the document details