[ RadSafe ] lognormal distribution of indoor radon measurements

[Dukelow, James S Jr]

Catherine Organo and John Andrews wrote:
-----Original Message-----
From: radsafe-bounces at radlab.nl [mailto:radsafe-bounces at radlab.nl] On
Behalf Of John Andrews
Sent: Friday, May 20, 2005 10:08 AM
To: Catherine Organo; radsafe
Subject: Re: [ RadSafe ] lognormal distribution of indoor radon
measurements


Catherine Organo wrote:

>Would anybody on this list know or have a guess at the following 
>question:
>
>Assuming they are in place, would radon protective measures in newly 
>built homes (barrier and activated sump) have an influence on the 
>'usually' log-normal distribution of indoor measurements?
>
>Regards
>
>Catherine
>
>Catherine Organo
>Scientific Officer
>Natural Radioactivity Section
>Radiological Protection Institute of Ireland
>
>  
>
In my opinion, it would change the geometric mean, but not the 
underlying distribution.  Most natural phenomena have a log normal 
distribution.  Only those few that depend on binary random processes, 
like flipping a coin, have normal distributions.

John Andrews,
_______________________________________________

The situation is a bit more complicated than John describes.

If a real world property is accurately modeled as a product of several
other randomly varying quantities, then the Central Limit Theorem tells
us that the property's probability distribution will be approximately
lognormal.  If the real world property is more accurately modeled as a
sum of random variables (not necessarily binary random variables), then
the CLT tells us its probability distribution will be approximately
normal.  If modeled by a mixture of sums and products, then the
distribution will be somewhere in between normal and lognormal.

Best regards.

Jim Dukelow
Pacific Northwest National Laboratory
Richland, WA
jim.dukelow at pnl.gov

These comments are mine and have not been reviewed and/or approved by my
management or by the U.S. Department of Energy.