More powerful procedures for multiple significance testing

Yosef Hochberg*, Yoav Benjamini

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


The problem of multiple comparisons is discussed in the context of medical research. The need for more powerful procedures than classical multiple comparison procedures is indicated. To this end some new, general and simple procedures are discussed and demonstrated by two examples from the medical literature: the neuropsychologic effects of unidentified childhood exposure to lead, and the sleep patterns of sober chronic alcoholics.

Original languageEnglish
Pages (from-to)811-818
Number of pages8
JournalStatistics in Medicine
Issue number7
StatePublished - Jul 1990


Dive into the research topics of 'More powerful procedures for multiple significance testing'. Together they form a unique fingerprint.

Cite this