effectmeasures: Total and Indirect Effects of Variables

In gesca: Generalized Structured Component Analysis (GSCA)

Description

In the user-specified structural equation model, this function calculates the total and indirect effects of (1) latent variables on other latent variables in the structural model and (2) latent variables on their indicators in the measurement model.

Usage

effectmeasures(object)

Arguments

object

An object of class. This can be created via the gesca.run function.

Details

The bootstrapped standard errors will be provided for all total and indirect effects. The user can specify the number of bootstrap samples via the gesca.run function. If the number of bootstrap samples is equal to 0, these standard errors will not be provided.

Value

Four matrices whose row and column names are the same as the names of indicators and latent variables.

References

Hwang, H., & Takane, Y. (2014). Generalized Structured Component Analysis: A Component-Based Approach to Structural Equation Modeling (pp.119-123). Boca Raton, FL: Chapman & Hall/CRC Press.

See Also

gesca.run

Examples

library(gesca)
data(gesca.rick2) # Organizational identification example of Bagozzi

# Model specification
myModel <- "
		# Measurement model 
		OP =~ cei1 + cei2 + cei3
		OI =~ ma1 + ma2 + ma3
		AC_J =~ orgcmt1 + orgcmt2 + orgcmt3
		AC_L =~ orgcmt5 + orgcmt6 + orgcmt8
		
		# Structural model 
		OI ~ OP
		AC_J ~ OI
		AC_L ~ OI
"

# Run a multiple-group GSCA with the grouping variable gender:
# GSCA.group <- gesca.run(myModel, gesca.rick2, group.name = "gender")
# summary(GSCA.group)		# default output
# effectmeasures(GSCA.group)	# total and indirect effects per group 
GSCA.group.nbt0 <- gesca.run(myModel, gesca.rick2, group.name = "gender", nbt = 0)
effectmeasures(GSCA.group.nbt0)	# no bootstrapped standard error of parameters

# Second-order latent model with formative indicators
SndForm <- "
		# Measurement model 
		OP(0) =~ cei1 + cei2 + cei3 + cei4 + cei5 + cei6 + cei7 + cei8
		OI(0) =~ ma1 + ma2 + ma3 + ma4 + ma5 + ma6
		AC_J(0) =~ orgcmt1 + orgcmt2 + orgcmt3 + orgcmt7
		AC_L(0) =~ orgcmt5 + orgcmt6 + orgcmt8

		# Second-order latents 
		AC(0) =: AC_J + AC_L

		# Structural model 
		OI ~ OP
		AC ~ OI
"

# SndFormFit <- gesca.run(SndForm, gesca.rick2, "gender")
# summary(SndFormFit)
# effectmeasures(SndFormFit)
# Total/Indirect effects of latent variables on indicators
# will not be provided and instead, be reported as "NULL".

gesca documentation built on May 2, 2019, 7:28 a.m.