Skip to main content
. Author manuscript; available in PMC: 2011 Mar 1.
Published in final edited form as: Health Place. 2009 Nov 10;16(2):321. doi: 10.1016/j.healthplace.2009.10.017

Table 2.

Impact of the randomization procedure (random swapping versus spatially ordered swapping of indicators of late-stage diagnosis) on the results of the test of hypothesis. The number of incidence rates in the contingency table of Figure 2A that are declared significantly larger or smaller than the tri-county average of 18% are reported for α=0.05 and three types of multiple testing correction: No correction, False discovery rate (FDR) correction, and simulation-based procedure (JEF).

Multiple testing correction
Swapping None FDR JEF
Random 20/15 3/2 9/4
Spatially ordered 19/9 4/0 8/3