Software & Tools

Open-source R packages for structural equation modeling and business analytics.

SEMinR

CRAN Co-Creator

A Domain-Specific Language for Building and Estimating Structural Equation Models in R. SEMinR provides an intuitive, natural syntax that lets researchers express PLS and CBSEM models using familiar terminology.

130,000+ downloads Version 2.5.0

model.R

library(seminr)

# Define measurement model
measurements <- constructs(
  composite("Satisfaction",
    multi_items("sat", 1:3)),
  composite("Loyalty",
    multi_items("loy", 1:4)),
  composite("Quality",
    multi_items("qual", 1:5))
)

# Define structural model
structure <- relationships(
  paths("Quality",
    to = "Satisfaction"),
  paths("Satisfaction",
    to = "Loyalty")
)

# Estimate and bootstrap
model <- estimate_pls(
  data        = survey_data,
  measurement = measurements,
  structure   = structure
)

boot <- bootstrap_model(model,
  nboot = 1000)

summary(boot)

Key Features

Natural Syntax
Express models using familiar SEM constructs like composites, paths, and relationships
PLS & CBSEM Estimation
Full PLS path modeling with bootstrapping, plus CBSEM support
Predictive Assessment
Out-of-sample prediction with cross-validation and PLSpredict
Model Visualization
Generate path diagrams and result plots directly from estimated models

CRAN GitHub

seminrExtras

CRAN Creator v1.0.3

Extensions and extra features for SEMinR, including additional model assessment tools, visualization functions, and advanced reporting capabilities for PLS-SEM analysis.

CRAN GitHub

SEMinR in the browser

Web App Runs on your computer No upload

Paste the SEMinR model you would run in R and your indicator data, and get the complete PLS-SEM analysis: reliability and validity, bootstrapped path coefficients, mediation, PLSpredict and CVPAT, the bootstrapped congruence test (Franke, Sarstedt & Danks, 2021), a threshold assessment following PLS-SEM Using R, the path diagram, a downloadable report, and the R script that reproduces it. No R and no install.

Nothing you paste is ever transmitted anywhere. The model is estimated and bootstrapped inside your browser, on your own machine. It runs the open-source TypeScript ports of SEMinR and seminrExtras; point estimates, summary tables and PLSpredict reproduce R exactly.

Launch App seminr-ts