Re: Cohen's latest reward

[Kai Kaletsch <info@eic.nu>]

Don,



"inter county variability" in itself can never be a reason for anything, not

just Cohen's data. There has to be a systematic reason (mechanism) why inter

county variability would produce such a strong correlation. If there is

random inter county variability, it would bias the results toward the null.



If you have a graph of x vs. y which shows a statistically significant

structure, your explanation of the graph must, in some way, involve both x

and y. Otherwise it is not an explanation of the graph. This holds for all

sciences. Epidemiology is not exempt from the basic laws of logic.



Kai Kaletsch



P.S.: You can probably control the inter county variability (by stratifying

the data) better than you can control the inter person variability (by

choosing the right controls) in case control studies. Would you not be more

concerned about "intra" county variability? As long as you leave the data at

the county level, there is no ecological fallacy. You could consider each

county to be an organism [my version of the Gaia theory :) ]. You would

still be left with the question: What are the counties with high radon doing

right and what are the counties with low radon doing wrong?



----- Original Message -----

From: "Rad health" <healthrad@hotmail.com>

To: <info@eic.nu>; <radsafe@list.vanderbilt.edu>

Sent: Wednesday, January 16, 2002 8:18 PM

Subject: Re: Cohen's latest reward





> Kai,

>

> What emperical basis do you have to show that Cohen's inverse relationship

> is not due to inter county variability?

>

> Don Smith

>

>

> >From: Kai Kaletsch <info@eic.nu>

> >To: Rad health <healthrad@hotmail.com>, radsafe@list.vanderbilt.edu

> >Subject: Re: Cohen's latest reward

> >Date: Wed, 16 Jan 2002 19:39:17 -0600

> >

> >Explaining a statistically strong association with "cross level bias and

> >inter county variability" is like explaining the pollution in a lake with

> >"all the dead fish that keep washing up on shore". Both the cross level

> >bias

> >and the dead fish are symptoms of some more fundamental mechanism that

> >needs

> >to be identified.

> >

> >Kai Kaletsch

> >

> >----- Original Message -----

> >From: "Rad health" <healthrad@HOTMAIL.COM>

> >To: <radsafe@list.vanderbilt.edu>

> >Sent: Wednesday, January 16, 2002 4:48 PM

> >Subject: Cohen's latest reward

> >

> >

> > > Dr. Cohen,

> >

> > >....one of the major factors,

> > > among others, most likely causing your paradoxical findings is

> >cross-level

> > > bias or inter county variability....

> >

> >

> >

>

>

> _________________________________________________________________

> Send and receive Hotmail on your mobile device: http://mobile.msn.com

>



************************************************************************

You are currently subscribed to the Radsafe mailing list. To unsubscribe,

send an e-mail to Majordomo@list.vanderbilt.edu  Put the text "unsubscribe

radsafe" (no quote marks) in the body of the e-mail, with no subject line. You can view the Radsafe archives at http://www.vanderbilt.edu/radsafe/