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

Some power considerations when deciding to use transformations

A Kingman1, G Zion

  • 1Epidemiology & Oral Diseases Prevention Program, National Institute of Dental Research, Bethesda, MD 20892.

Statistics in Medicine
|March 15, 1994
PubMed
Summary
This summary is machine-generated.

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

Erratum.

Journal of periodontology·2018
Same author

Oral health care services utilisation in the adult US population: Medical Expenditure Panel Survey 2006.

Community dental health·2013
Same author

Transmission filter for the extreme ultraviolet spectral region composed of a thin Saran (C(2)H(2)Cl(2)) foil.

Applied optics·2010
Same author

Peptide-based ELISAs are not sensitive and specific enough to detect muscarinic receptor type 3 autoantibodies in serum from patients with Sjogren's syndrome.

Annals of the rheumatic diseases·2010
Same author

This Is No Myth.

Public health reports (Washington, D.C. : 1974)·2009
Same author

Analysis of clinical trials involving non-cavitated caries lesions.

Journal of dental research·2004
Same journal

A Mixture of Distributed Lag Non-Linear Models to Account for Spatially Heterogeneous Exposure-Lag-Response Associations.

Statistics in medicine·2026
Same journal

Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi-Experimental Studies With Panel Data.

Statistics in medicine·2026
Same journal

Covariate Adjustment for Wilcoxon Two Sample Statistic and Test.

Statistics in medicine·2026
Same journal

Beyond Fixed Thresholds: Optimizing Summaries of Wearable Device Data via Piecewise Linearization of Quantile Functions.

Statistics in medicine·2026
Same journal

A Causal Framework for Evaluating the Total Effect of Strategies Aiming to Expand Screening and to Improve Outcomes.

Statistics in medicine·2026
Same journal

Causal Effects on Nonterminal Event Time With Application to Antibiotic Usage and Future Resistance.

Statistics in medicine·2026
See all related articles

For small, skewed datasets, simulation studies show the UMP test offers the highest statistical power. Transformations like log or square root can be effective, but power losses vary by method and sample size.

Area of Science:

  • Biostatistics
  • Statistical modeling
  • Data analysis

Background:

  • Skewed data presents challenges for standard statistical tests.
  • Transformations (log, square root) or distribution-free methods are common approaches.
  • Understanding their impact on statistical power and Type I error is crucial.

Purpose of the Study:

  • To compare the performance of various statistical tests for skewed data.
  • To evaluate the effects of data transformations on detecting group differences.
  • To assess Type I error and statistical power under different distributional assumptions.

Main Methods:

  • Simulation studies using gamma and log-normal distributions with matched moments.
  • Comparison of six statistical tests for gamma distributions.

Related Experiment Videos

  • Comparison of five statistical tests for log-normal distributions.
  • Analysis of Type I error and statistical power across varying sample sizes and skewness.

    • Main Results:

    • All tested procedures maintained robust Type I error rates.
    • The UMP (Uniformly Most Powerful) test generally yielded the highest statistical power.
    • Power losses for randomization and t-tests on square root or original scales were minimal.
    • Log transformations showed greater power loss compared to other methods.

    Conclusions:

    • The UMP test is recommended for skewed data when appropriate.
    • Square root transformations offer a good balance of power and simplicity.
    • Careful consideration of transformations is necessary to avoid power reduction.