Controlling false discovery rates in factorial experiments with between-subjects and within-subjects tests

BMC Research Notes

Marjan van Erk¹  Suzan Wopereis¹  Carina M Rubingh¹  Eric D Schoen² 

[1] TNO Earth, Environmental and Life Sciences, PO Box 360, 3700 AJ Zeist, Netherlands;Department of Environment, Technology and Technology Management, University of Antwerp, Prinsstraat 13, 2000 Antwerp, Belgium

关键词: Within-subjects effects;    False discovery rate;    Factorial experiment;    Between-subjects effects;    Analysis of variance;   

Others  :  1232501
DOI  :  10.1186/1756-0500-6-204

 received in 2012-11-02, accepted in 2013-03-21,  发布年份 2013

【 摘 要 】

Background

The False Discovery Rate (FDR) controls the expected number of false positives among the positive test results. It is not straightforward how to conduct a FDR controlling procedure in experiments with a factorial structure, while at the same time there are between-subjects and within-subjects factors. This is because there are P-values for different tests in one and the same response along with P-values for the same test and different responses.

Findings

We propose a procedure resulting in a single P-value per response, calculated over the tests of all the factorial effects. FDR control can then be based on the set of single P-values.

Conclusions

The proposed procedure is very easy to apply and is recommended for all designs with factors applied at different levels of the randomization, such as cross-over designs with added between-subjects factors.

Trial registration

NCT00959790

【 授权许可】

   
2013 Schoen et al.; licensee BioMed Central Ltd.

【 图 表 】

Figure 1.

Figure 2.

【 参考文献 】

• [1]Hochberg Y, Tamhane A: Multiple comparison procedures. New York: Wiley; 1987.
• [2]Benjamini Y, Hochberg Y: Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B 1995, 57:289-300.
• [3]Storey JD, Tibshirani R: Statistical significance for genomewide studies. Proceedings of the National Academy of Sciences USA 2003, 100:9440-9445.
• [4]Greenhouse SW, Geisser S: On methods in the analysis of profile data. Psychometrika 1959, 24:95-112.
• [5]McLean RA, Sanders WL, Stroup WW: A unified approach to mixed linear models. Am Statistician 1991, 45:54-64.
• [6]Bailey RA: Design of comparative experiments. Cambridge: Cambridge University Press; 2008.
• [7]Fisher RA: Statistical methods for research workers. London: Oliver and Boyd; 1935.
• [8]Fang Y, Wit E: Test the overall significance of p-values by using joint tail probability of ordered p-values as test statistic. In Advanced Data Mining and Applications Volume 5139 of Lecture Notes in Computer Science. Edited by Tang C, Ling C, Zhou X, Cercone N, Li X. Heidelberg: Springer Berlin; 2008:435-443.
• [9]Benjamini Y, Yekutieli D: The control of the false discovery rate in multiple testing under dependency. Ann Stat 2001, 29:1165-1188.