What's wrong with Bonferroni adjustments

In research, an effective treatment may be deemed no better than placebo. [...]contrary to what some researchers believe, Bonferroni adjustments do not guarantee a "prudent" interpretation of results. [...]adjustments are appropriate when the same test is repeated in many subsamples, such...

Full description

Saved in:
Bibliographic Details
Published in:BMJ 1998-04, Vol.316 (7139), p.1236-1238
Main Author: Perneger, Thomas V
Format: Article
Language:English
Subjects:
Analysis. Health state
Biological and medical sciences
Biomedical research
Confidence intervals
Data Interpretation, Statistical
Economic models
Education and Debate
Epidemiology
Error rates
False negative errors
General aspects
Heart attacks
Inference
Laboratories
Medical sciences
Null hypothesis
P values
Probability
Public health. Hygiene
Public health. Hygiene-occupational medicine
Researchers
Statistical analysis
Statistical inferences
Statistical significance
Statistical tests
Statistics as Topic - methods
Studies
Variables
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In research, an effective treatment may be deemed no better than placebo. [...]contrary to what some researchers believe, Bonferroni adjustments do not guarantee a "prudent" interpretation of results. [...]adjustments are appropriate when the same test is repeated in many subsamples, such as when stratified analyses (by age group, sex, income status, etc) are conducted without an a priori hypothesis that the primary association should differ between these subgroups.
ISSN:0959-8138
0959-8146
1468-5833
1756-1833
DOI:10.1136/bmj.316.7139.1236