TESTS OF GEOGRAPHICAL CORRELATION WITH ADJUSTMENT FOR EXPLANATORY VARIABLES - AN APPLICATION TO DYSPNEA IN THE ELDERLY

We propose a test of correlation of the residuals in generalized linea r models which is a generalization of the spatial autocorrelation test based on Moran's I. It allows adjustment for sizes of geographical ar eas and for explanatory variables. A formula is given to compute the w eights according to the alternative hypothesis. We compare inference u sing the distribution in the model and using the permutation distribut ion. A simulation study showed that the model-based test may be very c onservative and this leads to a loss of power compared to the permutat ion test or to the model-based test with correction for estimated para meters. As this latter is intractable for very large samples when the model includes explanatory variables, we recommend the use of the perm utation test. The permutation test is used to study geographical corre lation of dyspnoea in the elderly. (C) 1997 by John Wiley & Sons, Ltd.