# 6. PLS-SEM Bootstrapping

## Purpose

Which structural and measurement-model estimates have stable bootstrap inference?

## Practice study

- File: [brand-choice-bootstrap.csv](../samples/brand-choice-bootstrap.csv)
- Prepared project: [brand-choice-bootstrap.qpls](../projects/brand-choice-bootstrap.qpls)
- Observations: 420
- Data type: deterministic synthetic instructional data
- Complexity: a realistic multi-construct model with multi-item measurement blocks

**Prerequisites**

- Run PLS-SEM Algorithm successfully on the same saved model first

## Exact procedure

1. Start QuickPLS and choose **Open Project**. Select [brand-choice-bootstrap.qpls](../projects/brand-choice-bootstrap.qpls). This prepared teaching project contains the exact dataset, analysis-ready base model, bindings, and a saved layout. Complete any method-specific term, group, or higher-order instruction stated below before calculation.

2. Open **Data** and confirm **brand-choice-bootstrap.csv** contains 420 observations. Return to **Model**. The linked CSV remains available separately for inspection and re-import practice.

   ![PLS-SEM Bootstrapping sample data](../screenshots/pls-sem-bootstrapping/01-data-ready.png)

3. Confirm that the prepared model contains these constructs and measurement blocks: Promotion Intensity as a formative composite using promotion_reach, promotion_frequency, and discount_depth; Quality (quality_1–quality_4), Trust (trust_1–trust_3), Satisfaction (satisfaction_1–satisfaction_4), and Loyalty (loyalty_1–loyalty_4) as reflective constructs.

   Confirm these saved structural paths: Promotion Intensity, Quality, Trust, and Satisfaction → Loyalty. Keep these four paths direct so this tutorial demonstrates ordinary PLS bootstrapping rather than the separate mediation workflow.

4. The prepared project already contains a saved, neatly arranged model layout. Do not choose **Arrange** or otherwise change the canvas before this tutorial calculation. If you later edit the model, arrange it once and save the edited project before calculating.

   ![PLS-SEM Bootstrapping prepared model or variable roles](../screenshots/pls-sem-bootstrapping/02-arranged-model.png)

5. Choose **Validate** and confirm there are no blockers. Do not save the unchanged prepared project; save only after you intentionally edit it.

6. Use **Calculate → PLS-SEM Bootstrapping**.

   ![PLS-SEM Bootstrapping calculation setup](../screenshots/pls-sem-bootstrapping/03-calculation-setup.png)

7. Complete the method-specific settings:

- Use 1,000 bootstrap samples for the tutorial run (increase for final research reporting as appropriate)
- Use a fixed seed and two-tailed 95% confidence intervals
- Leave Open Results when finished selected

8. Read **Readiness**, leave **Open Results when finished** selected, and choose **Start calculation** once. Wait for **Completed**; do not close the project while the result is being saved.

9. In **Results**, confirm **PLS-SEM Bootstrapping** is selected and inspect these outputs in order:

1. Inference → Path coefficients
2. Inference → Outer loadings
3. Inference → Outer weights
4. Inference → R-square
5. Direct effects when applicable
6. Graphical Output with Inner model = t value and Outer model = t value

   ![PLS-SEM Bootstrapping primary result](../screenshots/pls-sem-bootstrapping/04-results-primary.png)

   ![PLS-SEM Bootstrapping secondary result](../screenshots/pls-sem-bootstrapping/05-results-secondary.png)

   ![PLS-SEM Bootstrapping outer-model statistics](../screenshots/pls-sem-bootstrapping/06-outer-model-bootstrap.png)

   ![PLS-SEM Bootstrapping t values on the graphical model](../screenshots/pls-sem-bootstrapping/07-t-values-diagram.png)

10. Choose **Save Report** to preserve the exact result. Use **Export** for the needed table/report format. Close and reopen the project once and confirm the saved result remains selectable.

## Reading guidance

- The data are deterministic synthetic teaching data. They are suitable for reproducing the workflow, not for substantive publication claims.
- QuickPLS 2.62.8 exposes the ordinary bootstrap families used in this tutorial through the unchanged Results interface
- Original, sample mean, standard deviation, |t|, p value, and confidence bounds come from the saved bootstrap result; QuickPLS does not fabricate absent historical statistics
- This model deliberately contains both reflective and formative measurement, but no indirect chain, so the ordinary bootstrap tables are not replaced by the separate mediation result workflow
- The calculation-settings screenshot shows the recommended 1,000-sample tutorial setting. The automated screenshot-evidence run uses the valid 100-sample minimum to remain bounded, so do not compare its illustrative numerical values with your 1,000-sample result