Complete practical course for learning how to run, assess, interpret, and report regression-based analysis in SmartPLS 4.
Created by Mahbub Hassan for students, thesis researchers, and applied quantitative researchers who need a step-by-step SmartPLS workflow from data preparation to final reporting.
This course covers two practical SmartPLS 4 workflows:
-
Observed-variable regression in SmartPLS 4
Single and multiple linear regression using measured variables, including coefficients, standardized coefficients, significance testing, robust standard errors, diagnostics, and reporting. -
PLS-SEM structural path assessment
Regression-like path modeling with latent constructs, including measurement model assessment, structural model assessment, bootstrapping, R-square, effect size, predictive relevance, PLSpredict, and reporting.
The course is intentionally practical. Learners follow SmartPLS menu steps, export results, fill assessment sheets, and write results in a publishable format.
This course now includes a dedicated paper-based track built around the shared 2026 SmartPLS software tutorial:
- Paper-based learning track
- Simple explanation of the paper
- Six-stage regression workflow
- HBAT-style variable map
- SmartPLS click paths for linear and logistic regression
- Results assessment workbook
Use this track if your main goal is to learn SmartPLS 4 regression exactly in the style of the published tutorial paper.
By the end of the course, learners should be able to:
- Prepare a clean dataset for SmartPLS 4.
- Build a regression model in the SmartPLS graphical interface.
- Run SmartPLS linear regression and regression bootstrapping.
- Interpret unstandardized and standardized coefficients.
- Assess p-values, t-values, standard errors, and confidence intervals.
- Use HC3/HC4 robust standard errors when appropriate.
- Inspect QQ plots and regression diagnostic output.
- Assess PLS-SEM measurement models using loadings, reliability, AVE, Fornell-Larcker, and HTMT.
- Assess structural models using VIF, path coefficients, R-square, f-square, bootstrapping, and predictive assessment.
- Write professional result sections for thesis, journal paper, or report submission.
| Week | Module | Main Outcome |
|---|---|---|
| 1 | SmartPLS Regression Workflow | Understand regression options in SmartPLS 4 |
| 2 | Data Preparation | Prepare CSV/XLSX data and codebook |
| 3 | Building a Regression Model | Draw dependent, independent, and control variables |
| 4 | Running Linear Regression | Estimate single and multiple regression |
| 5 | Assessing Regression Results | Interpret coefficients, R-square, and diagnostics |
| 6 | Regression Bootstrapping | Test coefficient significance using bootstrapping |
| 7 | Reporting Regression Results | Write tables and result paragraphs |
| 8 | PLS-SEM Measurement Model | Assess reliability and validity |
| 9 | PLS-SEM Structural Model | Assess paths, VIF, R-square, f-square, and significance |
| 10 | Prediction and PLSpredict | Assess predictive performance |
| 11 | Common Reviewer Problems | Fix reporting and interpretation weaknesses |
| 12 | Capstone SmartPLS Report | Complete a full SmartPLS analysis report |
Optional paper-based extension:
| Module | Topic | Main Outcome |
|---|---|---|
| 13 | Logistic Regression in SmartPLS 4 | Run and report binary-outcome regression |
| Lab | Topic | File |
|---|---|---|
| 1 | Import dataset and create project | Lab 1 |
| 2 | Build and run a SmartPLS regression model | Lab 2 |
| 3 | Assess regression output | Lab 3 |
| 4 | Run regression bootstrapping | Lab 4 |
| 5 | Assess PLS-SEM measurement and structural results | Lab 5 |
| 6 | Write the final results section | Lab 6 |
- SmartPLS-ready synthetic dataset
- HBAT-style synthetic dataset based on the paper's variable structure
- Codebook
- Step-by-step labs
- Result assessment checklists
- Regression reporting templates
- Logistic regression reporting template
- PLS-SEM reporting templates
- Export-result tracking sheets
- Python benchmark script for checking regression estimates outside SmartPLS
- GitHub Pages course homepage
- Download or clone this repository.
- Open SmartPLS 4.
- Import
datasets/smartpls_regression_training_data.csv. - Start with Lab 1.
- Use the templates in
templates/while reporting your results.
For the paper-based track, import:
datasets/hbat_smartpls_regression_case.csv
Then follow:
Optional Python benchmark:
python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
python scripts/benchmark_regression.py
python scripts/benchmark_hbat_case.pySmartPLS is proprietary software. This repository provides teaching materials, datasets, templates, and workflow guidance. It does not include SmartPLS software or a license.
- SmartPLS Regression documentation
- SmartPLS Regression Bootstrapping documentation
- SmartPLS PLS-SEM Algorithm documentation
- SmartPLS PLSpredict documentation
- SmartPLS Algorithms and Techniques
When reporting analyses conducted with SmartPLS, cite SmartPLS as recommended by the software provider:
Ringle, C. M., Wende, S., & Becker, J.-M. (2024). SmartPLS 4. SmartPLS. https://www.smartpls.com
Course text and teaching materials are intended for open teaching and learning with attribution. Code examples are released under the MIT License.