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

Solutions to the Behrens-Fisher problem.

James F Reed1

  • 1Research Institute, St. Lukes Hospital & Health Network, 801 Ostrum Street, Bethlehem, PA 18015, USA. reedj@slhn.org

Computer Methods and Programs in Biomedicine
|February 13, 2003
PubMed
Summary

This study introduces new statistical methods for comparing means of two independent groups when standard assumptions are violated. The developed FORTRAN code offers robust alternatives for unequal variances and non-normal distributions.

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

Long-term outcome in elderly patients after operation for traumatic intracranial hemorrhage.

Delaware medical journal·2014
Same author

Training mothers in infant cardiopulmonary resuscitation with an instructional DVD and manikin.

The Journal of the American Osteopathic Association·2013
Same author

Weight status and perception of colorectal cancer risk.

Journal of the American Board of Family Medicine : JABFM·2012
Same author

Components of the metabolic syndrome differ between young and old adults in the US population.

Journal of clinical hypertension (Greenwich, Conn.)·2012
Same author

Twitter use during emergency medicine conferences.

The American journal of emergency medicine·2012
Same author

The relationship of peripheral arterial disease and metabolic syndrome prevalence in asymptomatic US adults 40 years and older: results from the National Health and Nutrition Examination Survey (1999-2004).

Journal of clinical hypertension (Greenwich, Conn.)·2012

Area of Science:

  • Statistics
  • Biostatistics
  • Statistical Computing

Background:

  • Classical Student's t-test assumes equal variances and normal distributions.
  • Welch's t-statistic and Satterthwaite's F test address unequal variances but can be non-robust.
  • Standard statistical tests have strict assumptions often unmet in real-world data.

Purpose of the Study:

  • To provide robust statistical methods for comparing means of two independent populations.
  • To address limitations of traditional t-tests when normality and homogeneity of variance assumptions are violated.
  • To offer alternative statistical procedures for non-normal distributions and unequal variances.

Main Methods:

  • Development of executable FORTRAN code.
  • Implementation of statistical methods by Cressie and Whitford, Yuen and Dixon, and Yuen.
  • Focus on robust estimation for two-sample mean comparisons.

Main Results:

  • Executable FORTRAN code is available for advanced statistical analyses.
  • The code facilitates the application of robust statistical methods.
  • Provides alternatives to standard t-tests when assumptions are not met.

Conclusions:

  • The developed FORTRAN code offers practical solutions for complex statistical scenarios.
  • Researchers can utilize these methods when standard parametric test assumptions are indefensible.
  • Enhances the reliability of statistical inference in diverse data distributions.

Related Experiment Videos