XLSTAT: Group comparison with PLS-PM

 

Background

Group comparison with PLSPM makes it possible to test whether the coefficients and parameters of a structural model differ significantly between several groups of observations. XLSTAT provides a parametric test based on bootstrap standard errors and a nonparametric permutation test to assess these differences.

 

Dataset for Example

250 customers of a mobile phone operator were surveyed to answer a large number of questions to model their loyalty to the operator. The PLS-PM model is based on the European Customer Satisfaction Index (ECSI).

This tutorial will show you how to compare groups in a PLS-PM context in Excel using XLSTAT.

 

Setting Up

• Select the expert display mode in the XLSTAT-PLSPM options.

• This dialog box appears:

• Select the expert display and save your settings.

• As in the general tutorial build the ECSI model using the PLSPMGraph sheet and the Path Modeling toolbar.

 

• Click on the button   to select the variable for defining groups.

 

• Click on the button  to run the PLS-PM analysis.

In the Multi tests tab:

• Select the multigroup t test and permutation test options.

Note: If the bootstrap confidence interval button in th Options tab, is not selected, the t-test cannot be applied.

 

 

Interpreting the Results

If groups and mutigroup tests are selected, some new outputs are available:

• Worksheet PLSPM(Group): For each group, complete results are displayed in separated worksheets.

• Worksheet PLSPM (Multigroup t test): For each path coefficient, results of the t test are summarized in a table. Each line represents a pair of groups.

• Worksheet PLSPM (Permutation test): For each type of parameter, results of the permutation test are summarized in a table.

The two first sheets can be analyzed independently using the general method described in the XLSTAT-PLSPM tutorial.

The groups have been selected using the question: How well do you think your mobile phone provider compares with your ideal mobile phone provider?

Results of the multigroup t test are the following:

Results of the multigroup permutation test are the following:

These two tests’ results show a significant difference between the groups for two path coefficients:

• The arrow between perceived value and satisfaction

• The arrow between satisfaction and loyalty

The second one has a strategic impact in marketing studies and experts should pay special attention to people answering negatively to the grouping question.