Others may differ, but I teach regression to the mean as being a different 
phenomenon than what you might refer to as capitalizing on the cyclical nature 
of an event. For example, many people may not go to the doctor until their cold 
or flu is at its worst which means that, by the time they get a prescription, 
it is impossible to distinguish the effect of the drug from the natural cycle 
of the illness.

I explain regression to the mean as being due to the inherent error in any test 
based on X = T + e (or the score on a test (X) is the sum of the true score (T) 
plus or minus error (e)).

This means that some extreme high scores are based on the true score plus 
error. In those cases, the obtained score will regress to the mean on second 
testing because traditional testing theory assumes that the error is as likely 
to decrease as to increase a score. Therefore, if you choose the extremely high 
scores, the score on second testing is likely to regress toward the mean (be 
lower) because the error is unlikely to make such a positive contribution on 
second testing.

At the other extreme, some extreme low scores are based on the true score minus 
error. In those cases, the obtained score will regress to the mean on second 
testing (get higher) because the error pulling the score down is not likely to 
be as strong on second testing. So, if you sample just the extremely low 
scores, the score on second testing is likely to regress toward the mean (be 
higher) because the error is unlikely to make such a negative contribution to 
the total score on second testing.

Do others make this distinction when discussing these effects?

Rick

Dr. Rick Froman, Chair
Division of Humanities and Social Sciences Box 3055
x7295
[email protected]
http://tinyurl.com/DrFroman

Proverbs 14:15 "A simple man believes anything, but a prudent man gives thought 
to his steps."


-----Original Message-----
From: Frantz, Sue [mailto:[email protected]]
Sent: Monday, February 09, 2009 11:15 AM
To: Teaching in the Psychological Sciences (TIPS)
Subject: RE: [tips] Bogus treatments


For regression to the mean, how about vitamin C to cure colds?  
(http://www.mayoclinic.com/health/cold-remedies/ID00036)  Everyone starts out 
with a cold, a day or two into it they start drinking lots of orange juice, and 
a few days later they feel better... which likely would have been the case 
anyway, but they credit the orange juice.


---
To make changes to your subscription contact:

Bill Southerly ([email protected])

Reply via email to