semPLS: plsmUtils – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

plsmUtils

Utility methods for plsm objects.

Description

By means of the utility methods pls model descrpitions inheriting from class plsm can easily be altered or investigated.

Usage

plsmEdit(model, ...)
## S3 method for class 'plsm'
plsmEdit(model, data, ...)

addLV(model, ...)
## S3 method for class 'plsm'
addLV(model, data, LV, mode, MVs, pred, succ, ...)

invertLVs(model, ...)
## S3 method for class 'plsm'
invertLVs(model, LVs, ...)

removeLVs(model, ...)
## S3 method for class 'plsm'
removeLVs(model, which, ...)

addMVs(model, ...)
## S3 method for class 'plsm'
addMVs(model, data, LV, MVs, ...)

removeMVs(model, ...)
## S3 method for class 'plsm'
removeMVs(model, MVs, ...)

addPath(model, ...)
## S3 method for class 'plsm'
addPath(model, from, to, ...)

removePath(model, ...)
## S3 method for class 'plsm'
removePath(model, from, to, ...)

exogenous(model)
endogenous(model)
reflective(model)
formative(model)
indicators(model, LV)
predecessors(model)
successors(model)
connected(model)
acyclic(model)

Arguments

`model`	An object inheriting from class `plsm` as returned from `plsm` or `read.splsm`.
`data`	A `data.frame` containing the observed variables (MVs). The storage mode for all the MVs included in the model must be `numeric`.
`LV`	A `character` value naming an LV to add or the LV to associate with newly created MVs.
`mode`	A `character` value, specifying the measurement mode of the LV to add: `"A"` for a reflective LV, `"B"` for a formative LV.
`MVs`	A `character` vector naming MVs, which are either to be added, deleted or associated with a newly created LV. Note: the names of the MVs must exist as `numeric` variables in `data`.
`pred`	A `character` vector naming the predecessors of an LV. If `pred` is `NULL` the LV is exogenous.
`succ`	A `character` vector naming the successors of an LV.
`LVs`	A `character` vector naming LVs to invert. Invert means to invert the direction of pathes of associated measurement models. This means changing reflective measurement to formative and vice versa.
`which`	A `character` vector naming LVs to delete.
`from`	A `character` vector naming predecessors of LVs specified in `to`.
`to`	A `character` vector naming successors of LVs specified in `from`.
`...`	Currently unused.

Details

exogenous returns the exogenous latent variables contained in a path model.

endogenous returns the endogenous latent variables contained in a path model.

reflective returns the reflectively measured latent variables contained in a path model.

formative returns the formatively measured latent variables contained in a path model.

indicators returns the index or observed variables related to the given latent variable.

predecessors returns a list with the names of predecessors for each latent variable.

successors returns a list with the names of successors for each latent variable.

connected returns TRUE if the inner model is a connected graph.

acyclic returns TRUE if the inner model is recursive, thus represented by an acyclic graph.

Value

All utility methods return an object of class plsm.

Examples

data(mobi)
data(ECSIsm)
data(ECSImm)

ECSI <- plsm(data=mobi, strucmod=ECSIsm, measuremod=ECSImm)


### Some sense free examples

# Print the block of MVs associated with "Expectation".
ECSI[["blocks"]]["Expectation"]

# Change measurement model for "Expectation" from reflective to
# formative and print its block MVs
invertLVs(model=ECSI, LVs=c("Expectation"))[["blocks"]]["Expectation"]

# Print the adjacency matrix for the inner model.
ECSI[["D"]]

# Add an additional path from "Quality" to "Loyalty" and
# print the resulting adjacency matrix.
addPath(model=ECSI, from="Quality", to="Loyalty")[["D"]]

# Remove the path previously added.
removePath(model=ECSI, from="Image", to=c("Satisfaction", "Loyalty"))[["D"]]

# Print all MVs used in the model
ECSI$manifest

# Remove some MVs and print the MVs used in the model
removeMVs(model=ECSI, MVs=c("IMAG3", "CUEX1", "PERQ7"))$manifest


### some handy functions

exogenous(ECSI)
endogenous(ECSI)
reflective(ECSI)
formative(ECSI)
indicators(ECSI, "Image")
predecessors(ECSI)

semPLS

Structural Equation Modeling Using Partial Least Squares

v1.0-10

GPL-2

Authors

Armin Monecke <armin.monecke@stat.uni-muenchen.de>

Initial release

2013-01-21