|
Activity Number:
|
442
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #301558 |
|
Title:
|
Developing New Methodologies To Estimate the Proportion of Homes in a Region with Indoor Radon Levels Exceeding a Reference Threshold
|
|
Author(s):
|
Patrick E. Murphy*+ and Catherine Organo
|
|
Companies:
|
University College Dublin and Radiological Protection Institute of Ireland
|
|
Address:
|
School of Mathematical Sciences, UCD, Dublin, International, 0, Ireland
|
|
Keywords:
|
Radon ; Mapping ; Beta ; Modelling
|
|
Abstract:
|
The World Health Organization estimates that between 6% and 15% of all lung cancer deaths per year are caused by exposure to indoor radon. The statistical task involved in producing radon maps is to estimate the proportion of homes, P, in a particular region which exceed a reference level of 200 Bqm3 and consequently require remediation. Early work conducted in this field established that radon levels could be modeled using a Log-Normal distribution. This paper will describe the development of alternative modeling approaches using the Log-Normal, Gamma and Beta distributions to produce estimates for P. We will then report on the results of recent work which investigates the effect of outliers on predictions of P. We also provide for the first time accurate determinations of appropriate sample sizes to be used in surveys for the construction of radon maps.
|
