Use a file in format comma separated value files (.csv). The file header with the names of the variable items is required. For example, the expected format of the headers for the items of latent variables A and B (with two items each) is A1, A2, B1, B2. Our wizard generates both the measurement model and the structural model. All variables are defined as reflectives (Mode A), and therefore consistent PLS is calculated. The last variable is used as the dependent variable and the others as independent variables. These definitions can be modified.
Use SEMinR functions to describe measurement model.
Use SEMinR functions to describe structural model.
Calculation time increases if a larger number of bootstrap resamples are used.



ShinyPLS V0.1 - 2020.09.22 Developed by Patricio Ramirez-Correa (patricio.ramirez@ucn.cl) based in SEMinR

Process


        

Structural Model

Model Path coefficients are displayed in the relationships.
Coefficient of Determination (R^2 Value) and Path Coefficients R^2 values of 0.75, 0.50 or 0.25 can be considered as substantial, moderate or weak, respectively. If a path coefficient is truly different from zero can be considered (see bootstrapping).

        Scatterplot Matrix of Construct Scores
        Scatterplot Matrix composes multiple score plots combined into a single panel.
        

Measurement Model

Reflective Measurement Models: Indicator Reliability (outer loading) If outer loading is > 0.70 then retain the reflective indicator.

        Formative Measurement Models: Significance and Relevance (outer weights)
        If outer weight is relatively high (> 0.50), the indicator should generally be retained (see bootstrapping).
        

        Consistent Reliability (rhoA reliability)
        Consistent Reliability is calculated for reflective constructs. rhoA values >=  0.7 are satisfactory.
        

        Composite Reliability (rhoC) and Convergent Validity (AVE, average Variance extracted)
        Composite Reliability is calculated for composite constructs. rhoC values >=  0.7 are satisfactory. AVE values >= 0.50 are acceptable.
        

        

Bootstrapping

Results Path coefficient values truly different from zero can be considered. For formative measurement models, outer weight values truly different from zero can be considered. For discriminant validity the HTMT values must be significantly smaller than one.