PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
January 1, 20039,046 citationsOpen Access

Evaluating Goodness-of-Fit Measures in Structural Equation Modeling

Evaluating the Fit of Structural Equation Models: Tests of Significance and Descriptive Goodness-of-Fit Measures

View Full Paper
Ask AI
Bookmark
Share

Authors

KSKarin Schermelleh-EngelHMHelfried MoosbruggerHMH. G. Müller

Discussion

Loading...

Member takes

Overview

Methodological review demonstrates strategies for evaluating structural equation model fit using simulated data, highlighting guidelines for navigating conflicting goodness-of-fit indices.

Key Points

  • To provide applied researchers with practical guidelines for evaluating model adequacy and interpreting conflicting goodness-of-fit indices in structural equation modeling.
  • Reviewed parameter estimation methods, specifically maximum likelihood (ML) and weighted least squares (WLS), in relation to fit assessment.
  • Analyzed the characteristics, behavior, and recommendations for common descriptive goodness-of-fit indices and significance tests.
  • Generated an artificial dataset from a known baseline model to evaluate fit across two correctly specified and two misspecified models.
  • Goodness-of-fit indices frequently yield conflicting evaluations regarding how well a structural equation model matches empirical data.
  • Fit index performance and interpretation depend directly on the chosen parameter estimation framework, such as ML or WLS.
  • Simulations of correctly specified and misspecified models provide practical benchmarks for distinguishing between poor, adequate, and good model fit.

Cite This Study

Schermelleh-Engel et al. (2003) studied this question.

synapsesocial.com/papers/69d6fa2099397875bbaa7f79https://doi.org/10.23668/psycharchives.12784
View Full Paper
Ask AI
Bookmark
Share