Skip to main content

Advertisement

Table 2 Average proportion of false positives committed when detecting a significant disparity (two-tailed tests). Calculations are conducted for the six test statistics and four classes of increasing magnitude for the disparity. Numbers into parentheses give the percentage of simulations where the particular statistic yields the smallest proportion of false positives.

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

  DispI DispII DispIII DispIV DispV DispVI
Prostate       
Magnitude 1 0.384 (21%) 0.408 (1%) 0.385 (32%) 0.405 (1%) 0.407 (0%) 0.382 (45%)
Magnitude 2 0.297 (28%) 0.330 (0%) 0.301 (25%) 0.327 (0%) 0.329 (0%) 0.297 (47%)
Magnitude 3 0.234 (33%) 0.272 (0%) 0.239 (22%) 0.269 (0%) 0.269 (0%) 0.234 (45%)
Magnitude 4 0.186 (40%) 0.225 (0%) 0.191 (19%) 0.222 (0%) 0.222 (0%) 0.186 (41%)
Lung       
Magnitude 1 0.256 (48%) 0.277 (0%) 0.263 (6%) 0.279 (0%) 0.276 (0%) 0.257 (46%)
Magnitude 2 0.159 (55%) 0.182 (0%) 0.167 (4%) 0.185 (0%) 0.181 (0%) 0.160 (41%)
Magnitude 3 0.111 (57%) 0.135 (0%) 0.119 (4%) 0.137 (0%) 0.133 (0%) 0.113 (39%)
Magnitude 4 0.081 (56%) 0.105 (0%) 0.088 (4%) 0.106 (0%) 0.102 (0%) 0.083 (40%)