Re: Radon and Smoking

["Field, R. William" <bill-field@UIOWA.EDU>]

Dr. Cohen,



As I have stated before, validity of findings is not a matter of numbers. 

While I do not agree with all of NCRP No. 136, I do agree with the 

following quote from NCRP concerning the limitations of ecologic studies:



"Use of large numbers of geographic regions will not do away with 

biases.  More observations will generally increase precision but have 

little or no impact on validity, i.e., the degree of bias."  In other other 

words the observation can be precisely wrong.



What I find interesting is this quote in your 1991 paper:



"The results in Tables 2 and 3 clearly indicate that households with 

cigarette smokers have substantial lower Rn levels then others, but for 

some strange reason it seems that the difference decreases with increasing 

number of cigarette smoked.  It should be noted that from those tables that 

only 17% of all people who purchased Rn measurements have smoked in their 

households, whereas 33% of American adults are smokers."



Regards, Bill Field







>         --There is no reason why this ecological finding should equal the

>individual level finding, and I have never claimed that any ecological

>finding should equal an individual finding. Correlations come up for

>non-causal reasons in any data, and one would be hard pressed to explain

>the smoking vs radon correlation for U.S. counties as a causal

>relationship. The reason for the very small p-value is that there are so

>many data pooints. The actual correlation is not very large, something

>like 15%.

>         On the other hand, there is a disagreement between the radon vs

>smoking in the Iowa study and my individual level study in Health Physics

>60:631-642;1991- Tables 2 & 3. My study involves many tens of thousands of

>measurements, far more than the Iowa study



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

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/