1996
DOI: 10.2105/ajph.86.5.726
|View full text |Cite
|
Sign up to set email alerts
|

Adjusting for multiple testing when reporting research results: the Bonferroni vs Holm methods.

Abstract: Public health researchers are sometimes required to make adjustments for multiple testing in reporting their results, which reduces the apparent significance of effects and thus reduces statistical power. The Bonferroni procedure is the most widely recommended way of doing this, but another procedure, that of Holm, is uniformly better. Researchers may have neglected Holm's procedure because it has been framed in terms of hypothesis test rejection rather than in terms of P values. An adjustment to P values base… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

4
710
0
11

Year Published

2003
2003
2021
2021

Publication Types

Select...
10

Relationship

0
10

Authors

Journals

citations
Cited by 1,088 publications
(725 citation statements)
references
References 8 publications
4
710
0
11
Order By: Relevance
“…Multicomparison analyses between measurements have been done with HolmŠídák post hoc tests. 19,20 The differences between D (hypoventilation) and A (hyperventilation), C (normoventilation II) and A (hyperventilation) as well as B (normoventilation I) and D (hypoventilation) were calculated using t-tests for paired data or Welch test and nonparametric Wilcoxon signed-rank test, if indicated. 21 To provide an estimate of the effect of changes in PaCO 2 levels and their clinical meaningfulness, we calculated mean differences (MD) and their 95% confidence intervals (MD; 95% CI upper bound, lower bound; P value).…”
Section: Resultsmentioning
confidence: 99%
“…Multicomparison analyses between measurements have been done with HolmŠídák post hoc tests. 19,20 The differences between D (hypoventilation) and A (hyperventilation), C (normoventilation II) and A (hyperventilation) as well as B (normoventilation I) and D (hypoventilation) were calculated using t-tests for paired data or Welch test and nonparametric Wilcoxon signed-rank test, if indicated. 21 To provide an estimate of the effect of changes in PaCO 2 levels and their clinical meaningfulness, we calculated mean differences (MD) and their 95% confidence intervals (MD; 95% CI upper bound, lower bound; P value).…”
Section: Resultsmentioning
confidence: 99%
“…Type I error rate was controlled at 5% across the multiple tests of the NLMM analysis, using sequential Bonferroni adjustment (Aickin and Gensler, 1996). All analyses were conducted in version 9 of SAS (SAS Institute, Cary, NC).…”
Section: Resultsmentioning
confidence: 99%
“…Pairwise comparisons among stages were performed with a Wilcoxon rank-sum test with Holm-Sidak correction (Holm 1979). The Holm-Sidak correction method is considered to be more powerful than the Bonferroni correction for P-values while still controlling type I error (Aickin and Gensler 1996).…”
Section: Methodsmentioning
confidence: 99%