Structural Equation Modeling

Description

Jointly estimates a measurement model of latent constructs and a structural model of paths between them, giving direct, indirect and total effects with fit indices.

SEM combines confirmatory factor analysis (latent constructs measured by indicators) with path models among constructs. In Vivek Gopinath's payments case (n = 520, robust ML), constructs are Trust, Fraud Risk, Slowness and Support with 3 Likert items each, plus observed Double-Checking and Adoption. Loadings range from 0.70 to 0.82 with CR 0.81–0.87 and AVE 0.57–0.69. Structural paths follow the DAG: FR → TR −0.45, SR → TR +0.31, TR → DC −0.50, DC → SL +0.42, SL → AD −0.28 and TR → AD +0.26, with indirect TR → AD ≈ +0.059 and total ≈ +0.32. Fit is reported with CFI 0.962, TLI 0.951, RMSEA 0.041 (90% CI .032–.050) and SRMR 0.037 against thresholds CFI/TLI ≥ 0.95, RMSEA ≤ 0.06, SRMR ≤ 0.08. Interpretation rules: |β| ≈ 0.10 small, 0.30 medium, 0.50 large; never interpret paths without qual context. The Nash protocol uses SEM in its Prove stage.

Attributes

Loss type
Latent
Effort
Weeks

Steps

  1. Specify the measurement model (each latent construct with its indicators).
  2. Screen missingness (< 10% per item, FIML), check α ≥ 0.70, and run CFA, dropping loadings < 0.60 unless theory demands.
  3. Add structural paths exactly as the DAG specifies.
  4. Estimate with robust ML (MLR) and report CFI, TLI, RMSEA and SRMR.
  5. Decompose direct, indirect and total effects.
  6. Test multi-group invariance, sensitivity to added confounders and out-of-sample stability.
  7. Report with the template: measurement reliabilities and fit, structural effects with significance, and prescriptions.

Sources

  • Vivek Gopinath (2025), DAGs, SEM & Mixed Methods in UX: Turning Insight into Evidence in Banking & Payments
  • Vivek Gopinath (2025), Equilibrium and the Design Object: How Nash Equilibrium Illuminates Why Designs Stabilise, Plateau, and Change
  • Principles and Practice of SEM
  • Statistical Analysis in JASP
  • Principles and Practice of Structural Equation Modeling (Kline)
  • Tenenbaum, G. & Eklund, R. (eds), Handbook of Sport Psychology, 3rd ed. (2007)
  • Veal, A. J. & Darcy, S., Research Methods in Sport Studies and Sport Management (2014)

Also known as

  • SEM
  • Covariance structure analysis
  • Latent variable path modelling
  • Covariance-based SEM
  • Path analysis with latent variables
  • Structural Equation Modelling (SEM)
  • Statistical path analysis
  • Wright's path analysis

Categories

  • Analytics & Behaviour
  • Statistics & Modelling