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

Skew and internal consistency.

Tammy Greer1, William P Dunlap, Samuel T Hunter

  • 1Department of Psychology, The University of Southern Mississippi, Hattiesburg, MS, USA. tammy.greer@usm.edu

The Journal of Applied Psychology
|November 15, 2006
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

Optimizing Implementation of Food is Medicine Interventions in a Community-Based Healthcare Setting: Nutrition360 Pilot Intervention Results.

American journal of health promotion : AJHP·2026
Same author

Extreme ("20") self-shocks during the Self-Aggression Paradigm and lifetime suicide attempt history: Potential use in suicide research.

Journal of psychiatric research·2025
Same author

Using machine learning models to predict vaccine hesitancy: a showcase of COVID-19 vaccine hesitancy in rural populations during the pandemic.

Vaccine·2025
Same author

Are Parental Internalizing Problems and Emerging Adult Emotion Regulation Moderated by Gender?

Psychological reports·2025
Same author

Creativity training needs assessment for homeland security enterprise: a case for creative thinking.

Journal of policing, intelligence and counter terrorism·2025
Same author

Exploring bias due to below-limit-of-detection values in influenza vaccine antibody modeling: A case study and instructional guide for the CIVIC study.

Vaccine·2025
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

Skewed data can slightly decrease reliability estimates like standardized item alpha. This effect is more pronounced with greater skew, fewer items, and low correlations between items.

Area of Science:

  • Psychometrics
  • Statistical Modeling
  • Survey Design

Background:

  • Standardized item alpha is a widely used reliability coefficient.
  • Its accuracy may be influenced by the distributional properties of the data.
  • Skewness is a common characteristic of real-world data that can affect statistical assumptions.

Purpose of the Study:

  • To investigate the impact of data skewness on standardized item alpha.
  • To compare alpha values derived from normal versus skewed distributions.
  • To examine how skewness interacts with other factors like interitem correlation and the number of items.

Main Methods:

  • Monte Carlo simulation techniques were employed.
  • Comparisons were made between alpha coefficients calculated from normal, lognormal, ranked, and skewed Likert-type variables.

Related Experiment Videos

  • Variations included the degree and direction of skew, population interitem correlation (rho), number of items, and number of Likert categories.

    • Main Results:

    • Data skewness was found to decrease the average interitem correlation.
    • Skewness produced small decreases in standardized item alpha.
    • The largest decreases in alpha occurred under conditions of high skew, low rho, opposite skew directions across items, and a small number of items.

    Conclusions:

    • Skewed data can attenuate standardized item alpha, potentially underestimating scale reliability.
    • Researchers should consider data distribution when interpreting alpha coefficients, especially in non-normally distributed datasets.
    • The findings highlight the importance of assessing distributional assumptions in psychometric analyses.