Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Unequal group sizes in randomised trials: guarding against guessing.

Kenneth F Schulz1, David A Grimes

  • 1Family Health International, PO Box 13950Research Triangle Park, NC 27709, USA. KSchulz@fhi.org

Lancet (London, England)
|March 29, 2002
PubMed
Summary

Randomised trials do not require equal group sizes; mild disparities reflect true randomness and prevent bias. Simple randomisation is preferred over methods that force balance, especially in non-double-blinded studies.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Smartphone-derived digital motor measures to monitor progression in idiopathic REM sleep behavior disorder.

NPJ Parkinson's disease·2026
Same author

SPIRIT 2025 statement: updated guideline for protocols of randomised trials.

Lancet (London, England)·2026
Same author

Biomarkers.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

Cognitive profiles at presentation and subsequent cognitive decline and quality of life in Parkinson's disease.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

Aligning SPIRIT and SAGER guidelines for equity in clinical trials - Authors' reply.

Lancet (London, England)·2025
Same author

The relation of eye movements to the occurrence of freezing of gait in Parkinson's disease.

Brain communications·2025

Area of Science:

  • Clinical Trials Methodology
  • Biostatistics
  • Research Integrity

Background:

  • The assumption that randomised controlled trials (RCTs) require equal sample sizes is a common misconception.
  • This misconception can lead to biased trial design and execution, particularly when investigators enforce equality through non-scientific means.

Purpose of the Study:

  • To challenge the notion of mandatory equal group sizes in RCTs.
  • To highlight the scientific benefits of accepting random variation in group sizes.
  • To advocate for the appropriate use of randomisation methods that preserve unpredictability and minimize bias.

Main Methods:

  • Discussion of simple (unrestricted) randomisation versus restricted randomisation techniques.
  • Analysis of the impact of permuted-block randomisation and its effect on unpredictability.

Related Experiment Videos

  • Introduction of alternative methods like urn randomisation and mixed randomisation for enhanced unpredictability.

    • Main Results:

    • Simple randomisation naturally produces unequal group sizes, reflecting true randomness.
    • Forcing equal group sizes, especially with permuted-block randomisation in non-double-blinded trials, can diminish unpredictability and introduce bias.
    • Simple randomisation is underused, while fixed-block randomisation is overused.

    Conclusions:

    • Investigators should embrace mild disparities in group sizes in simple randomisation for non-double-blinded trials over 200 participants.
    • Urn randomisation or mixed randomisation offer better unpredictability for smaller trials or subgroups.
    • Prioritising unpredictability over perfect balance is crucial for reducing selection bias in RCTs.