Welcome to DSN

Mplus is a specialist statistical modeling platform developed by Muthén & Muthén for researchers who need to analyse complex relationships between observed and latent variables.

Unlike general-purpose statistical packages that concentrate mainly on standard statistical tests, Mplus provides an integrated framework for advanced methods such as Structural Equation Modeling (SEM), Confirmatory Factor Analysis (CFA), Exploratory Factor Analysis (EFA), latent class analysis, growth models, multilevel analysis, Bayesian modeling, survival analysis, and longitudinal modeling.

It can work with cross-sectional and longitudinal datasets, single-level and multilevel data, observed and latent heterogeneity, and datasets containing missing values. Supported outcome types include continuous, censored, binary, ordinal, nominal, and count variables.

This makes Mplus particularly useful for research in psychology, education, healthcare, epidemiology, sociology, behavioural sciences, economics, business research, public health, and other fields where complex statistical relationships must be modeled.

Why Choose Mplus?

Powerful Structural Equation Modeling

Build and estimate sophisticated SEM models incorporating observed variables, latent constructs, indirect effects, multiple groups, growth models, and complex relationships.

Factor Analysis

Perform both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) for scale development, construct validation, measurement models, and latent-variable research.

Mplus 9.1 also introduces second-order exploratory factor analysis methods, including SEFA and DSEFA.

Multilevel Modeling

Analyse hierarchical or clustered datasets involving individuals within schools, patients within hospitals, employees within organisations, repeated observations within individuals, and similar nested research designs.

Multilevel capabilities include regression, path analysis, factor analysis, SEM, growth modeling and survival modeling.

Mixture Modeling & Latent Class Analysis

Mplus provides extensive tools for identifying unobserved population subgroups through techniques such as:

• Latent Class Analysis (LCA)
• Latent Profile Analysis (LPA)
• Growth Mixture Modeling
• Latent Transition Analysis
• Longitudinal Mixture Modeling

Bayesian Statistical Analysis

Use Bayesian estimation for complex latent-variable and multilevel models where traditional estimation techniques may be difficult or computationally demanding.

Longitudinal & Time-Series Analysis

Analyse changes across time using growth modeling, longitudinal SEM and Dynamic Structural Equation Modeling (DSEM).

Version 9.1 includes improvements for short time-series estimation and additional DSEM3 capabilities.

Monte Carlo Simulation

Researchers can generate and analyse simulated datasets using models available within Mplus, making it useful for simulation studies, methodological research and statistical power investigations.

What’s New in Mplus 9.1?

Mplus Version 9.1 was released on May 19, 2026. Important additions include:

• Second-order Exploratory Factor Analysis (SEFA)
• Direct Effects Second-Order EFA (DSEFA)
• Improved ESEM output formatting
• Improved estimation for short time series
• Additional DSEM3 capabilities
• Random residual variances and covariances for DSEM3
• Enhancements for measurement-invariance testing

Version 9 also introduced or expanded capabilities including Penalized Structural Equation Modeling (PSEM), three-level DSEM, multistep mixture modeling, bootstrap support for two-level models, VIF output, Bayesian Effective Sample Size information and partial correlations.

Major Mplus Analysis Capabilities

Mplus can be used for:

Structural Equation Modeling

• SEM
• CFA
• EFA
• ESEM
• Path analysis
• Mediation analysis
• Measurement invariance
• Latent-variable interactions

Longitudinal Analysis

• Growth modeling
• Latent growth models
• Longitudinal mixture models
• Latent transition analysis
• Dynamic Structural Equation Modeling

Advanced Modeling

• Multilevel modeling
• Mixture modeling
• Bayesian analysis
• Item Response Theory
• Survival analysis
• Complex survey analysis
• Missing-data analysis
• Monte Carlo simulation
• Time-series modeling

Who Is Mplus For?

Mplus is particularly suitable for:

• University students
• PhD and postgraduate researchers
• Academic researchers
• Psychology researchers
• Social science researchers
• Education researchers
• Healthcare researchers
• Public health researchers
• Epidemiologists
• Economists
• Data analysts
• Quantitative researchers
• Research institutions
• Universities and laboratories

It is especially valuable when research goes beyond conventional regression and requires latent variables, hierarchical structures, longitudinal observations, categorical outcomes, mixtures or sophisticated SEM models.