#path-analysis

Test complex theoretical models using SEM. Model specification: 1. Draw path diagram showing hypothesized relationships. 2. Identify endogenous (dependent) and exogenous (independent) variables. 3. Specify direct and indirect paths. 4. Include error terms for endogenous variables. Analysis in R lavaan or SPSS AMOS: 1. Measurement model: confirmatory factor analysis for latent constructs. 2. Structural model: test path relationships. 3. Model identification: degrees of freedom ≥ 0 for identified model. Sample size: minimum 200 observations, 10-20 per parameter. Model fit assessment: 1. Chi-square test (non-significant preferred but sensitive to sample size). 2. Comparative Fit Index (CFI > 0.95). 3. Root Mean Square Error of Approximation (RMSEA < 0.08). 4. Standardized Root Mean Square Residual (SRMR < 0.08). Modification indices suggest model improvements, but use theory-driven changes only.