This article provides an overview of statistical analysis with latent variables, extending traditional structural equation modeling (SEM) to include a wide range of statistical concepts. It argues for the integration of psychometric and statistical modeling ideas, presenting a general model that unifies various analysis types such as factor models, growth curve models, multilevel models, latent class models, and discrete-time survival models. The article discusses the use of continuous and categorical latent variables, highlighting their applications in random effects, missing data, hierarchical data, finite mixtures, latent classes, and clusters. It emphasizes the importance of latent variables in capturing unobserved heterogeneity and improving the integration of psychometric and statistical modeling. The article also includes examples and illustrations to demonstrate the practical application of these models, particularly using the Mplus software.This article provides an overview of statistical analysis with latent variables, extending traditional structural equation modeling (SEM) to include a wide range of statistical concepts. It argues for the integration of psychometric and statistical modeling ideas, presenting a general model that unifies various analysis types such as factor models, growth curve models, multilevel models, latent class models, and discrete-time survival models. The article discusses the use of continuous and categorical latent variables, highlighting their applications in random effects, missing data, hierarchical data, finite mixtures, latent classes, and clusters. It emphasizes the importance of latent variables in capturing unobserved heterogeneity and improving the integration of psychometric and statistical modeling. The article also includes examples and illustrations to demonstrate the practical application of these models, particularly using the Mplus software.