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

Maximal reliability and power in covariance structure models.

Spiridon Penev1, Tenko Raykov

  • 1Department of Statistics, University of New South Wales, Sydney, Australia. S.Penev@unsw.edu.au

The British Journal of Mathematical and Statistical Psychology
|May 20, 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

On Estimating the Precision of Individual Test Scores: A Note on Pfadt, Molenaar, Hurks, and Sijtsma.

Psychometrika·2026
Same author

Proportion Explained Component Variance in Second-Order Scales: A Note on a Latent Variable Modeling Approach.

Educational and psychological measurement·2025
Same author

The One-Parameter Logistic Model Can Be True With Zero Probability for a Unidimensional Measuring Instrument: How One Could Go Wrong Removing Items Not Satisfying the Model.

Educational and psychological measurement·2025
Same author

Evaluating Change in Adjusted <i>R</i>-Square and <i>R</i>-Square Indices: A Latent Variable Method Application.

Educational and psychological measurement·2025
Same author

Fatigue in hematological cancer changes across chemotherapy trajectory within the context of IL-6, not hemoglobin level: evidence from growth curve modeling.

Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer·2025
Same author

An Omega-Hierarchical Extension Index for Second-Order Constructs With Hierarchical Measuring Instruments.

Educational and psychological measurement·2025

This study examines statistical power in latent linear regression models. It finds that item parcelling can reduce power, but optimal weighting strategies can mitigate this loss.

Area of Science:

  • Psychometrics
  • Statistical Modeling

Background:

  • Covariance structure modeling commonly uses the non-centrality parameter to indicate statistical power.
  • Latent linear regression models are crucial for understanding complex relationships between unobserved variables.

Purpose of the Study:

  • To investigate the contribution of maximal reliability coefficients to statistical power in latent linear regression.
  • To explicate the relationship between reliability and power in congeneric measures.
  • To assess the impact of item parcelling on the power of latent regression tests.

Main Methods:

  • Examining the maximal reliability coefficient's contribution to power.
  • Analyzing the relationship in the context of congeneric measures.
  • Evaluating the effect of item parcelling on hypothesis test power.

Related Experiment Videos

Main Results:

  • Maximal reliability coefficients are linked to increased statistical power in latent variable indicator analysis.
  • Item parcelling can decrease the power of tests for latent regression parameters.
  • Specific weighting recommendations are provided to avoid power reduction.

Conclusions:

  • Reliability of latent variable indicators directly influences statistical power in regression models.
  • Careful consideration of item parcelling and weighting is necessary to maintain adequate power.
  • Optimal linear composites with maximal reliability offer effective strategies for parcelling weights.