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

Simple procedures for blinded sample size adjustment that do not affect the type I error rate.

Meinhard Kieser1, Tim Friede

  • 1Department of Biometry, Dr Willmar Schwabe Pharmaceuticals, Willmar-Schwabe-Str 4, D-76227 Karlsruhe, Germany. meinhard.kieser@schwabe.de

Statistics in Medicine
|December 4, 2003
PubMed
Summary

Sample size adjustment in statistical tests is crucial. This study shows that using blind variance estimators maintains the type I error rate and ensures desired statistical power, even with initial variance misestimations.

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

Variable selection for clinical prediction models in low-dimensional data - a simulation study comparing traditional regression and machine learning methods.

BMC medical research methodology·2026
Same author

Estimation, testing and sample size calculation within the responder-stratified exponential survival model.

Journal of biopharmaceutical statistics·2026
Same author

Parental mental illness and child brain structure: A diffusion MRI study of emotion regulation related pathways.

NeuroImage·2026
Same author

Multimodal assessment of emotion regulation in children of parents with a mental illness.

European child & adolescent psychiatry·2026
Same author

Association of hemodynamic variability during endovascular stroke treatment with functional outcome and parenchymal hemorrhage.

Therapeutic advances in neurological disorders·2025
Same author

Multicenter upfront randomized phase II trial of quizartinib and high-dose cytarabine plus mitoxantrone in relapsed/refractory acute myeloid leukemia with FMSlike tyrosine kinase 3 internal tandem duplication.

Haematologica·2025

Area of Science:

  • Biostatistics
  • Statistical methodology
  • Clinical trial design

Background:

  • Accurate sample size determination is essential for statistical power in hypothesis testing, particularly for normally distributed data where variance is key.
  • Uncertainty in initial variance estimates necessitates flexible approaches like two-stage procedures for sample size adjustment.
  • Regulatory requirements emphasize maintaining study blindness and controlling the type I error rate during sample size recalculations.

Purpose of the Study:

  • To evaluate existing sample size adjustment procedures for the t-test, focusing on their ability to preserve blindness and control the type I error rate.
  • To demonstrate that sample size recalculation using blind variance estimators does not impact the type I error rate of the t-test.
  • To present a generalizable method for sample size adjustment using permutation tests that maintains the significance level.

Related Experiment Videos

Main Methods:

  • Analytical computations were performed to assess the impact of blind variance estimators on the type I error rate of the t-test.
  • The expected power of the proposed sample size adjustment procedures was evaluated under initial variance misspecification.
  • A permutation test-based approach was developed for broader applicability in sample size adjustment strategies.

Main Results:

  • The type I error rate of the t-test remains unaffected when blind variance estimators are employed for sample size recalculation.
  • The evaluated methods effectively achieve the desired statistical power, even when the initial variance is inaccurately specified.
  • The proposed permutation test method successfully maintains the significance level across various design situations and blind sample size recalculation strategies.

Conclusions:

  • Sample size adjustment in statistical testing can be effectively performed using blind variance estimators without compromising the type I error rate.
  • These methods provide robust control over statistical power, mitigating issues arising from initial variance estimation errors.
  • A flexible permutation test-based procedure offers a reliable approach to maintaining significance levels in diverse study designs with sample size adjustments.