Multiple comparisons

Multiple comparisons or multiple testing is a term used in statistics to address the problem of simultaneously testing a series of hypotheses. The issue arises when one considers that, as the number of hypotheses tested increases, the likelihood of identifying at least one significant result due to random chance also increases.

Overview

The problem of multiple comparisons arises when a number of inferential statistics tests are being performed simultaneously. This is a common occurrence in data analysis, where multiple variables or outcomes are being compared. The issue is that as the number of comparisons increases, the probability of observing a statistically significant result by chance (a Type I error) also increases.

Solutions

Several solutions have been proposed to control the familywise error rate (FWER) or the false discovery rate (FDR). These include the Bonferroni correction, the Šidák correction, the Holm-Bonferroni method, and the Benjamini–Hochberg procedure.

See also

• Statistical hypothesis testing
• Type I and type II errors
• Familywise error rate
• False discovery rate
• Bonferroni correction
• Šidák correction
• Holm-Bonferroni method
• Benjamini–Hochberg procedure

References

Multiple comparisons Resources
Latest articles Most recent articles Ongoing trials	Wikipedia