Table 4

Average proportion of false positives for two-tailed tests (unadjusted and corrected for multiple testing). Results for the six test statistics and two significance levels are averaged over all classes of disparity magnitude. Significance levels are either unadjusted or reduced using the Bonferroni correction, the Holm's procedure or the False Discovery Rate (FDR) approach.

Test statistic
α = 0.05
α = 0.10




Unadj.
Bonf.
Holm
FDR
Unadj.
Bonf.
Holm
FDR

Prostate








DispI
0.048
0.0
0.0
0.002
0.096
0.001
0.001
0.005
DispII
0.093
0.025
0.025
0.039
0.139
0.027
0.027
0.047
DispIII
0.032
0.0
0.0
0.001
0.065
0.0
0.0
0.002
DispIV
0.063
0.012
0.012
0.020
0.098
0.013
0.013
0.024
DispV
0.091
0.016
0.016
0.031
0.139
0.018
0.018
0.039
DispVI
0.042
0.0
0.0
0.001
0.087
0.0
0.0
0.003
Lung








DispI
0.049
0.0
0.0
0.004
0.098
0.0
0.0
0.009
DispII
0.068
0.016
0.016
0.022
0.115
0.016
0.016
0.028
DispIII
0.039
0.0
0.0
0.003
0.079
0.0
0.0
0.006
DispIV
0.054
0.009
0.009
0.014
0.095
0.009
0.009
0.018
DispV
0.068
0.012
0.012
0.021
0.116
0.013
0.013
0.027
DispVI
0.047
0.0
0.0
0.003
0.095
0.0
0.0
0.008

Goovaerts et al. International Journal of Health Geographics 2007 6:32   doi:10.1186/1476-072X-6-32

Open Data