Multiple Testing Method for Output Test Data Set

This drop-down menu enables you to specify a method for filtering the Output Test Data Set to only the most significant correlations.

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.

Leave this field blank to specify a -log10(p-value) cutoff directly with the parameter below. A value of 0 for this parameter includes all p-values. That is, no filtering is performed.

Multiple Testing Method	Definition
None	• Select this option to enable the -log10(p-Value) Cutoff field/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)1Hochberg, Y. and Benjamini, Y. (1990) More Powerful Procedures for Multiple Significance Testing. Statistics in Medicine 9: 811–818. 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.
Hochberg	• Assumes that p-values are independent and uniformly distributed under their respective null hypotheses, Hochberg (1988)2Hochberg, Y. (1988). A Sharper Bonferroni Procedure for Multiple Significance Testing. Biometrika 75: 800–803. 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)3Hommel, G. (1988) A Comparison of Two Modified Bonferroni Procedures. Biometrika 75: 383–386..
Holm	• See StepBon, below
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)4Benjamini, Y. and Hochberg, Y. (2000). On the Adaptive Control of the False Discovery Rate in Multiple Testing with Independent Statistics. Journal of Educational and Behavioral Statistics 25: 60–83. adaptive linear step-up method.
Dependent FDR	• Requests adjusted p-values by using the method of Benjamini and Yekateuli (2001)5Benjamini, Y. and Yekateuli, D. (2001). The Control of the False Discovery Rate in Multiple Testing under Dependency. Annals of Statistics 29: 1165–1188..
FDR	• Requests adjusted p-values by using the linear step-up method of Benjamini and Hochberg (1995)6Benjamini, Y. and Hochberg, Y. (1995). Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society, B 57: 289–300.. • These p-values do not control the familywise error rate, but they do control the false discovery rate.
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 estimating the number of true null hypotheses, and results are less conservative than FDR.

To Specify a Multiple Testing Method:

8

Make a selection using the drop-down menu.

Refer to the SAS PROC MULTTEST documentation for details and caveats on these methods.