Rr. Sokal et al., TESTING FOR REGIONAL DIFFERENCES IN MEANS - DISTINGUISHING INHERENT FROM SPURIOUS SPATIAL AUTOCORRELATION BY RESTRICTED RANDOMIZATION, Geographical analysis, 25(3), 1993, pp. 199-210
Tests for differences among regional means are typically carried out b
y analysis Of variance (ANOVA). When such data are spatially autocorre
lated (SA), the assumptions of ANOVA are not met, giving rise to exces
sive type I error rates. Two spatially adjusted ANOVA methods, Griffit
h's and COCOPAN, have been proposed to overcome this problem. In this
study we show, by means of extensive simulations, the magnitude of the
error rates introduced by SA induced in isolation-by-distance models
typical of those used in population genetics. For data suspected of ex
hibiting such SA, we propose a strategy for distinguishing between inh
erent SA, generated within the data by a contagious process, and spuri
ous SA, introduced by regional differences in means. The approach adop
ted is that of restricted randomization Of distance matrices. We also
furnish error rates and power estimates for both Griffith's method and
COCOPAN. In addition to the simulated data, the methods are applied t
o an actual example from plant population biology.