Multiple Testing Method for Adjusting p-Values

Select a method for adjusting for multiple testing across all categories. Adjusted p-values are computed for the method that you select, and a corresponding -log10(p-value) cutoff is computed for Volcano Plots.

The AdaptiveHolm, AdaptiveHochberg, Bonferroni, Holm, Hommel, Sidak, StepBon, and StepSid methods all control for the familywise error rate. The methods based on FDR all control for false discovery rate.

Note: Bootstrap and permutation methods are not available for Parametric Analysis of Gene Set Enrichment (PAGE) tests.

Multiple Testing Method

Definition

None (blank entry)

Select this option to enable the {link}{Emphasis}-log10(p-Value) Cutoff field and slider. This selection enables you to specify a -log10(p-value) cutoff directly.

AdaptiveHolm

Requests adjusted p-values by using the Hochberg and Benjamini (1990)1 adaptive step-down Bonferroni method.

AdaptiveHochberg

Requests adjusted p-values by using the Hochberg and Benjamini (1990)1 adaptive step-up Bonferroni method.

Bonferroni

Specifies that the Bonferroni adjustments (number of tests p-value) be computed for each test.
Note: These adjustments can be extremely conservative and should be viewed with caution.

Bootstrap

Specifies adjusted p-values using the Bootstrap method of Westfall and Young (1993)2.

Hochberg

Assumes that p-values are independent and uniformly distributed under their respective null hypotheses, Hochberg (1988)3 demonstrates that Holm’s step-down adjustments control the familywise error rate even when calculated in step-up fashion. Since the adjusted p-values are uniformly smaller for Hochberg’s method than for Holm’s method, the Hochberg method is more powerful. However, this improved power comes at the cost of having to make the assumption of independence.

Hommel

Requests adjusted p-values by using the method of Hommel (1988)4.

Holm

See Stepbon, below.

Permutation

Adjusted p-values are identical to the Bootstrap method, except that the within-stratum resampling is performed without replacement instead of with replacement.

Sidak

Computes the Šidák adjustment for each test.
Note: These adjustments are slightly less conservative than the Bonferroni adjustments, but they still should be viewed with caution.

StepBon

Requests adjusted p-values by using the step-down Bonferroni method of Holm (1988).

StepSid

Requests adjusted p-values by using the Šidák method but in step-down fashion.

AdaptiveFDR

Requests adjusted p-values by using the Benjamini and Hochberg (2000)5 adaptive linear step-up method.

DependentFDR

Requests adjusted p-values by using the method of Benjamini and Yekateuli (2001)6.

FDR

Requests adjusted p-values by using the linear step-up method of Benjamini and Hochberg (1995)7.
These p-values do not control the familywise error rate, but they do control the false discovery rate in some cases

FDRBoot

Requests the bootstrap-resampling false discovery rate controlling method of Yekateuli and Benjamini (1999)8.

FDRPerm

Requests the permutation-resampling false discovery rate controlling method of Yekateuli and Benjamini (1999)8.

pFDR

Computes the "q-values" of Storey (2002) and Storey, Taylor, and Siegmund (2004). PROC MULTTEST treats these "q-values" as adjusted p-values.
The computations depend on selecting a parameter and an estimation method for the false discovery rate.

To Specify a Multiple Testing Method:

8 Make a selection using the drop-down menu.

For Additional Information

Refer to p-Value Adjustments for more details about each of these methods.