Workshop SEM-PLS

"SEM-PLS" stands for Structural Equation Modeling using Partial Least Squares. It's a statistical technique commonly used in social sciences, marketing, management, and information systems when: 1. The research model is complex 2. The sample size is small 3. The data does not follow a normal distribution 4. The focus is on prediction rather than confirmation Key Concepts of SEM-PLS: - Structural Model (Inner Model): - Shows relationships between latent constructs (e.g., customer satisfaction → loyalty). Measurement Model (Outer Model): Specifies how observed indicators (survey items, for example) relate to their respective latent variables. Latent Variables: Variables that are not directly observed but are inferred (e.g., "trust", "brand image"). PLS Algorithm: Uses an iterative procedure to estimate weights and path coefficients. Prioritizes explained variance (R²) and predictive accuracy.