Re: RE: Re: Source of cancer data

[BERNARD L COHEN <blc+@PITT.EDU>]

On Wed, 19 Dec 2001, John Williams wrote:

>

> Because the QUALITY of data is much better.  As Smith and others have

> shown.  If you want to use ecologic or aggregate data, at least pick

> a state that has the most aggregates in the nation like Iowa.  There

> is 1% of the population there with 99 counties.  It has a SEER

> Registry and the highest radon in the U.S.  This is quality

> information.  The higher the quality information, the less you see

> Cohen's inverse association.



	--Smith et al give no numbers in their analysis; they show one

figure which has huge error bars, covering both positive and negative

slopes. Their data cover a 5 year period. The most convenient data I have

is for the 10 year period 1970-1979 during which 45 of the 99 counties had

fewer than 10 lung cancer death among females (13 with fewer than 5). Even

among males, only 17 of the 99 counties had as many as 100 deaths (i.e.

10% statistical accuracy for the 10 year period). Is this such "high

quality information"? If it is, they should quote numerical results with

standard deviations. These cannot be deduced from their one figure.





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

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.