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

Testing for adverse impact when sample size is small.

Michael W Collins1, Scott B Morris

  • 1Institute of Psychology, Illinois Institute of Technology, IL, USA. michael.w.collins@nasa.gov

The Journal of Applied Psychology
|March 26, 2008
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

The development and mixed-method stakeholder evaluation of a sport-related concussion symptom and treatment monitoring system in community rugby.

Irish journal of medical science·2026
Same author

Connecting the Dots: Network Analysis of the Concussion Clinical Profiles Screening among Adolescents.

The Journal of pediatrics·2026
Same author

Targeted Multidomain Treatment for Mild Traumatic Brain Injury: A Randomized Clinical Trial.

JAMA network open·2026
Same author

A brief, clinically feasible alternative to evaluate response to exercise following sport-related concussion: initial validation of the 10-M in male healthy controls.

The Journal of sports medicine and physical fitness·2026
Same author

Effect of co-morbid anxiety and depression history on multidomain concussion clinical outcomes and recovery time.

Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology·2026
Same author

Effect of Dose of Targeted Rehabilitation Exercises for Vestibular (T-REV) Impairments following mTBI on Dizziness Handicap: A Randomized Clinical Trial.

Journal of neurologic physical therapy : JNPT·2026
Same journal

Grateful leaders and attentive followers: How grateful attention transmits the cascade of gratitude expressions to strengthen coworker relationships.

The Journal of applied psychology·2026
Same journal

Scoring employment interviews with large language models: Evaluation design components, validity investigations, and best practice recommendations.

The Journal of applied psychology·2026
Same journal

A profile analysis of leader interpersonal emotion management strategies.

The Journal of applied psychology·2026
Same journal

A choice architecture intervention to increase diversity: Diverse defaults can counteract hiring discrimination.

The Journal of applied psychology·2026
Same journal

Transformational leadership in context: A meta-analysis of 40 years of research.

The Journal of applied psychology·2026
Same journal

Organizations espousing an authenticity ideology repel stigmatized job seekers.

The Journal of applied psychology·2026
See all related articles

For adverse impact evaluations, significance testing has limited value due to low statistical power. The Z-test is a reasonable choice for comparing selection rates, especially in larger samples.

Area of Science:

  • Statistics
  • Social Sciences
  • Legal Studies

Background:

  • Adverse impact evaluations require statistical tests to assess disparities in selection rates between groups.
  • Choosing the appropriate statistical test is crucial for accurate and reliable results.

Purpose of the Study:

  • To compare the effectiveness of various statistical test statistics for adverse impact evaluations.
  • To evaluate Type I error rates and statistical power, particularly in small sample scenarios.

Main Methods:

  • Comparison of several alternate test statistics.
  • Focus on Type I error rates and statistical power in small sample sizes.
  • Evaluation of the Z-test, Fisher Exact Test, Yates's chi-square test, and a test by G. J. G. Upton.

Related Experiment Videos

Main Results:

  • Significance testing demonstrated limited value across all tested methods due to low statistical power.
  • The Z-test performed well, except in extremely small sample sizes.
  • Upton's test offered marginal Type I error control but yielded similar results to the Z-test.
  • Fisher Exact Test and Yates's chi-square test are not recommended due to conservative Type I errors and lower power.

Conclusions:

  • The Z-test is a practical choice for assessing selection rate disparities, with considerations for very small sample sizes.
  • Fisher Exact Test and Yates's chi-square test are less suitable for adverse impact analysis compared to the Z-test.