The lavaan package is an R package for structural equation modeling (SEM), providing a free, open-source alternative to commercial software. It allows researchers to estimate latent variable models, including factor analysis, structural equation, longitudinal, multilevel, latent class, item response, and missing data models. The package is designed to be user-friendly, with a simple syntax that enables researchers to specify models using a text-based format. It supports non-normal continuous data and missing data, and includes options to mimic the output of commercial programs like Mplus and EQS. Lavaan was developed to address the limitations of existing SEM software, particularly the lack of open-source options. It is modular and allows for easy extension, such as adding new functions for computing standard errors. The package is particularly useful for applied researchers, educators, and statisticians who need a flexible and accessible tool for SEM analysis. The lavaan model syntax is a key feature, allowing users to specify models using a concise and intuitive format. It includes four types of formulas: regression formulas, latent variable formulas, residual variance and covariance formulas, and intercept formulas. This syntax is used to define models in various SEM applications, including confirmatory factor analysis (CFA) and structural equation modeling (SEM). In the CFA example, the Holzinger & Swineford 1939 dataset is used to illustrate the application of lavaan. The model includes three latent variables (factors) with three indicators each. The lavaan package automatically adds parameters such as residual variances and covariances, and allows users to specify equality constraints across groups. In the structural equation modeling example, the Industrialization and Political Democracy dataset is used to demonstrate the application of lavaan. The model includes three latent variables and specifies regressions among them. The package allows for the specification of equality constraints on parameters, and provides methods for examining model fit and parameter estimates. Lavaan also supports multiple-group analysis, allowing researchers to fit models across different groups and impose equality constraints on parameters across groups. The package includes various methods for examining model fit, including summary() and parameterEstimates(), as well as fitMeasures() for extracting specific fit indices. In addition, lavaan supports non-normal data through asymptotically distribution-free (ADF) estimation, scaled test statistics, and robust standard errors. These methods are particularly useful for handling non-normal data in SEM analysis. The package is continuously being developed and expanded, with a focus on providing a comprehensive and flexible tool for SEM research.The lavaan package is an R package for structural equation modeling (SEM), providing a free, open-source alternative to commercial software. It allows researchers to estimate latent variable models, including factor analysis, structural equation, longitudinal, multilevel, latent class, item response, and missing data models. The package is designed to be user-friendly, with a simple syntax that enables researchers to specify models using a text-based format. It supports non-normal continuous data and missing data, and includes options to mimic the output of commercial programs like Mplus and EQS. Lavaan was developed to address the limitations of existing SEM software, particularly the lack of open-source options. It is modular and allows for easy extension, such as adding new functions for computing standard errors. The package is particularly useful for applied researchers, educators, and statisticians who need a flexible and accessible tool for SEM analysis. The lavaan model syntax is a key feature, allowing users to specify models using a concise and intuitive format. It includes four types of formulas: regression formulas, latent variable formulas, residual variance and covariance formulas, and intercept formulas. This syntax is used to define models in various SEM applications, including confirmatory factor analysis (CFA) and structural equation modeling (SEM). In the CFA example, the Holzinger & Swineford 1939 dataset is used to illustrate the application of lavaan. The model includes three latent variables (factors) with three indicators each. The lavaan package automatically adds parameters such as residual variances and covariances, and allows users to specify equality constraints across groups. In the structural equation modeling example, the Industrialization and Political Democracy dataset is used to demonstrate the application of lavaan. The model includes three latent variables and specifies regressions among them. The package allows for the specification of equality constraints on parameters, and provides methods for examining model fit and parameter estimates. Lavaan also supports multiple-group analysis, allowing researchers to fit models across different groups and impose equality constraints on parameters across groups. The package includes various methods for examining model fit, including summary() and parameterEstimates(), as well as fitMeasures() for extracting specific fit indices. In addition, lavaan supports non-normal data through asymptotically distribution-free (ADF) estimation, scaled test statistics, and robust standard errors. These methods are particularly useful for handling non-normal data in SEM analysis. The package is continuously being developed and expanded, with a focus on providing a comprehensive and flexible tool for SEM research.