This review introduces the statistical issues relating to multiple comparisons, describes the theoretical basis behind analysis of variance (ANOVA), discusses the essential differences between ANOVA and multiple t-tests, and provides an example of the computations and computer programming used in performing ANOVA.
These OLPs were weakly associated with HLA-B*57 expression, but these associations did not survive multiple test comparisons (described below, Table S2).
Global P values (i.e., corrected for multiple comparisons as described by Sawyer 1999) were considered significant when the value was below 0.05.
In bold, item-residuals > 2.5 or item χ2 fit with p <.05 corrected for Bonferroni Multiple Comparisons Results described in Table 4 show that 6 items (9, 14, 15, 19, 21 and 22) presented high residuals and/or item χ2 scores significantly different from the expected.
Because heteroscedasticity resilient to transformation precluded the use of analysis of variance, a total of 22 t-tests were performed and P-value thresholds were established to correct for multiple comparisons as described by Benjamini and Hochberg [ 31] using the script published by IBM for SPSS software [ 32].
Multiple species alignments were constructed using the PRANK alignment algorithm (Löytynoja and Goldman 2005) and tested using the three codeml model comparisons described above.
We note that these correlations could well be spurious, given that we did not control for multiple comparisons; we simply describe the obtained effect sizes.
In Section 3, we briefly introduce the problem of multiple comparisons; we then describe the proposed hierarchical Bayesian models for disease mapping and how to estimate posterior classification probabilities.
The false discovery rate among multiple comparisons was controlled as described above.
A stricter correction level (p < 0.005, FDR corrected) was used for the FC statistical maps because we obtained too many clusters if we used the same multiple comparison correction methods described above.
