Skip to main content
. Author manuscript; available in PMC: 2011 Nov 1.
Published in final edited form as: Genet Epidemiol. 2010 Nov;34(7):725–738. doi: 10.1002/gepi.20536

Table 5.

Change in per-disease significance values when the correlation (ρ = 0.5) is ignored

Dimension (k) True P-value
0.05 0.01 1E-3 1E-4 1E-5
3 0.055 0.0106 1.03E-3 1.01E-4 1.01E-5
10 0.064 0.0116 1.11E-3 1.03E-4 1.04E-5
50 0.082 0.0137 1.12E-3 1.07E-4 1.05E-5
100 0.093 0.0150 1.23E-3 1.11E-4 1.06E-5
1000 0.150 0.0216 1.55E-3 1.30E-4 1.17E-5
10000 0.253 0.0335 2.11E-3 1.59E-4 1.33E-5

Entries in the table show increase in the P-value adjusted with the Bonferroni correction, compared to the “true” P that accounts for correlation between P-values for k diseases. For example, the true P=0.05 becomes 0.055 when there are three diseases.