Skip to main content

Advertisement

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.

From: A comparative analysis of aspatial statistics for detecting racial disparities in cancer mortality rates

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