We explained the "Higher-Order Construct Model" with an example. How to fit this type of model with SEM-PLS is very simple like other models. We describe the example of the “Higher-Order Construct Model” article step by step:

1- Defining the number of variables in the model (in this example it is 4).

2- We define the names of the variables in the model through the "Name of Latent Variable" button (in this example "Ability", "Verbal", "Spatial", "Numerical").

3- We define the "Measurement variables" corresponding to each of the Latent variables. Note that in the "Higher-Order Construct Model", latent variables can be measured once for the HOC variable and once for any of the LOC variables (see https://statisme.com/Home/BlogSingle/21.). In this example, we define the following variables:

Ability: “M1_1”, “M2_4”, “M3_6”, “M4_9”, “M5_13”, “M6_17”

Verbal: “M1_1”, “M2_4”

Spatial: “M5_13”, “M6_17”

Numerical:“M3_6”, “M4_9”

4- We define the "Higher-Order Construct Model" type through the “Relation Between Latent Variables” box (see the type of "Higher-Order Model" at https://statisme.com/Home/BlogSingle/21.).