This document discusses structural equation modeling (SEM) and its applications. SEM allows analyzing relationships between independent and dependent variables that can be continuous or discrete. It involves two main components: a measurement model using confirmatory factor analysis to represent unobserved latent variables, and a structural model to represent paths between variables. Model examination involves assessing fit indices, factor loadings, errors, and regression coefficients. Mediation can be examined using SEM by analyzing indirect paths from an independent variable to a dependent variable through a mediator variable. Bootstrapping provides a more accurate method than the Sobel test for estimating standard errors and confidence intervals in mediation models.