# 29. Higher-order models

## Purpose

Can the joy and love dimensions form a reflective-reflective Affective Commitment construct, and how do organizational prestige and identification predict it?

## Practice study

- File: [organizational-identification-model-comparison.csv](../samples/organizational-identification-model-comparison.csv)
- Prepared project: [organizational-identification-higher-order.qpls](../projects/organizational-identification-higher-order.qpls)
- Observations: 305 complete cases
- Data type: packaged academic teaching data
- Complexity: four lower-order constructs, 21 indicators, one reflective-reflective HOC, and a disjoint two-stage structural model

## Exact procedure

1. Start QuickPLS and choose **Open Project**. Select [organizational-identification-higher-order.qpls](../projects/organizational-identification-higher-order.qpls). This qualified project contains the exact dataset, higher-order definition, packaged result, bindings, and saved arranged layout used below.

2. Open **Data** and confirm the Organizational Identification dataset contains 305 complete observations. Return to **Model**. The linked CSV remains available separately for inspection and re-import practice.

   ![Higher-order models sample data](../screenshots/higher-order-models/01-data-ready.png)

3. Confirm these lower-order blocks: Organizational Prestige (org_pre1–org_pre8), Organizational Identification (org_ident1–org_ident6), Affective Commitment (Joy) (ac_joy1–ac_joy4), and Affective Commitment (Love) (ac_love1–ac_love3). Confirm that Joy and Love form the reflective-reflective **Affective Commitment** higher-order construct using the disjoint two-stage approach.

   Confirm these saved paths: Organizational Prestige → Organizational Identification; Organizational Prestige → Affective Commitment; Organizational Identification → Affective Commitment.

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.

   ![Higher-order models prepared model or variable roles](../screenshots/higher-order-models/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 Algorithm**. The calculation setup should identify the saved higher-order construct and the disjoint two-stage approach.

   ![Higher-order models calculation setup](../screenshots/higher-order-models/03-calculation-setup.png)

7. Complete the method-specific settings:

- Confirm the reflective-reflective disjoint two-stage HOC definition
- Run point estimation first; request bootstrap inference only when inferential intervals are needed

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 **Higher-order models** is selected and inspect these outputs in order:

1. **Component and structural estimates** for the higher-order loadings and structural paths
2. **Outer weights and loadings** for the stage-one measurement model
3. **Estimation stages** under **Advanced** when the exact stage receipts are needed
4. Inference availability before interpreting standard errors, confidence intervals, or p values

   ![Higher-order models primary result](../screenshots/higher-order-models/04-results-primary.png)

   ![Higher-order models secondary result](../screenshots/higher-order-models/05-results-secondary.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.
- The packaged result is the qualified point-estimation result. A separate bootstrap run is required before interpreting standard errors, confidence intervals, or p values.