How to show that unicorn milk is a chronobiotic: The regression-to-the-mean statistical artifact

Greg Atkinson, Jim Waterhouse, Thomas Reilly, Ben Edwards

    Research output: Contribution to journalArticle

    13 Citations (Scopus)

    Abstract

    Few chronobiologists may be aware of the regression-to-the-mean (RTM) statistical artifact, even though it may have far-reaching influences on chronobiological data. With the aid of simulated measurements of the circadian rhythm phase of body temperature and a completely bogus stimulus (unicorn milk), we explain what RTM is and provide examples relevant to chronobiology. We show how RTM may lead to erroneous conclusions regarding individual differences in phase responses to rhythm disturbances and how it may appear as though unicorn milk has phase-shifting effects and can successfully treat some circadian rhythm disorders. Guidelines are provided to ensure RTM effects are minimized in chronobiological investigations.

    Original languageEnglish
    Pages (from-to)1041-1053
    Number of pages13
    JournalChronobiology International
    Volume18
    Issue number6
    DOIs
    Publication statusPublished - 1 Dec 2001

    Fingerprint Dive into the research topics of 'How to show that unicorn milk is a chronobiotic: The regression-to-the-mean statistical artifact'. Together they form a unique fingerprint.

  • Cite this