plspm.groups: Two Groups Comparison in PLS-PM

In plspm: Partial Least Squares Path Modeling (PLS-PM)

Two Groups Comparison in PLS-PM

Description

Performs a group comparison test for comparing path coefficients between two groups. The null and alternative hypotheses to be tested are: H0: path coefficients are not significantly different; H1: path coefficients are significantly different

Usage

  plspm.groups(pls, group, Y = NULL, method = "bootstrap",
    reps = NULL)

Arguments

pls	object of class "plspm"
group	factor with 2 levels indicating the groups to be compared
Y	optional dataset (matrix or data frame) used when argument dataset=NULL inside pls.
method	method to be used in the test. Possible values are "bootstrap" or "permutation"
reps	integer indicating the number of either bootstrap resamples or number of permutations. If NULL then reps=100

Details

plspm.groups performs a two groups comparison test in PLS-PM for comparing path coefficients between two groups. Only two methods are available: 1) bootstrap, and 2) permutation. The bootstrap test is an adapted t-test based on bootstrap standard errors. The permutation test is a randomization test which provides a non-parametric option.

When the object pls does not contain a data matrix (i.e. pls$data=NULL), the user must provide the data matrix or data frame in Y.

Value

An object of class "plspm.groups"

test	Table with the results of the applied test. Includes: path coefficients of the global model, path coeffs of group1, path coeffs of group2, (absolute) difference of path coeffs between groups, and the test results with the p-value.
global	List with inner model results for the global model
group1	List with inner model results for group1
group2	List with inner model results for group2

Author(s)

Gaston Sanchez

References

Chin, W.W. (2003) A permutation procedure for multi-group comparison of PLS models. In: Vilares M., Tenenhaus M., Coelho P., Esposito Vinzi V., Morineau A. (Eds.) PLS and Related Methods - Proceedings of the International Symposium PLS03. Decisia, pp. 33-43.

Chin, W.W. (2000) Frequently Asked Questions, Partial Least Squares PLS-Graph.

See Also

plspm

Examples

## Not run: 
 ## example with customer satisfaction analysis
 ## group comparison based on the segmentation variable "gender"

 # load data satisfaction
 data(satisfaction)

 # define inner model matrix
 IMAG = c(0,0,0,0,0,0)
 EXPE = c(1,0,0,0,0,0)
 QUAL = c(0,1,0,0,0,0)
 VAL = c(0,1,1,0,0,0)
 SAT = c(1,1,1,1,0,0)
 LOY = c(1,0,0,0,1,0)
 sat_path = rbind(IMAG, EXPE, QUAL, VAL, SAT, LOY)

 # define outer model list
 sat_blocks = list(1:5, 6:10, 11:15, 16:19, 20:23, 24:27)

 # define vector of reflective modes
 sat_mod = rep("A", 6)

 # apply plspm
 satpls = plspm(satisfaction, sat_path, sat_blocks,
                modes = sat_mod, scaled = FALSE)

 # permutation test with 100 permutations
 group_perm = plspm.groups(satpls, satisfaction$gender,
                           method="permutation", reps=100)
 group_perm
 
## End(Not run)

plspm documentation built on June 22, 2024, 11:29 a.m.