A method for testing the significance of geographical correlations with application to industrial lung cancer in France

Publisher Website
Google Scholar
Abstract
This paper discusses some of the problems encountered when testing significance in geographical epidemiology where variables typically exhibit some spatial autocorrelation. A test of partial correlation between spatially autocorrelated variables is presented. It is based on evaluation of an effective sample size which takes account of spatial structure. Performance is assessed by Monte Carlo simulation. The proposed method is used to study male lung cancer rates in specific industries in France.