Covariance-based & Partial Least squares SEM modelling: Outline & applications
summary
Comparison shows concrete choices: R code uses lavaan and semPLS in R to fit CB‑SEM and PLS‑SEM, reads an SPSS file, computes Cronbach’s alpha >0.7 for scales, and plots models with semPlot; CB‑SEM expects large samples (~200+) while PLS runs on small samples (demo uses 444 respondents for a 3‑construct football sponsorship study). Significance for PLS comes from bootstrapping 500 resamples. Examples run end‑to‑end and produce fit indices, standardized coefficients and diagnostics.
AI-written summary · about
chapters (10)
- 0:00Covariance-based & PLS-SEM overview
- 1:48What is SEM?
- 15:14SEM model graphs
- 27:29Path analysis in SEM
- 38:40Exploratory vs confirmatory FA
- 43:38CB-SEM vs PLS-SEM
- 51:35SEM validation metrics
- 1:03:53SEM modeling issues
- 1:16:29SEM in R
- 1:37:41Summary & Q&A
related talks
15mInterpretable Machine Learning with R
02h 10mLasso and Ridge regression models in R: Introduction
01h 17mLightning Talks
27mI have no clue what my model is doing...
22mLeah
40mMachine Learning over Small Data: generating new variables with predictive power
24mStochastic Block Models with R: Statistically rigorous clustering
45mThe Statistical Fairness-Accuracy Frontier