modsem {modsem}

Interaction between latent variables

Description

modsem is a function for estimating interaction effects between latent variables, in structural equation models (SEM's). Methods for estimating interaction effects in SEM's can basically be split into two frameworks: 1. Product Indicator based approaches ("dblcent", "rca", "uca", "ca", "pind"), and 2. Distributionally based approaches ("lms", "qml"). For the product indicator based approaces, modsem() is essentially a just a fancy wrapper for lavaan::sem() which generates the necessary syntax, and variables for the estimation of models with latent product indicators. The distributionally based approaches are implemented in seperately, and are are not estimated using lavaan::sem(), but rather using custom functions (largely) written in C++ for performance reasons.

Usage

modsem(
  modelSyntax = NULL,
  data = NULL,
  method = "dblcent",
  match = FALSE,
  standardizeData = FALSE,
  centerData = FALSE,
  firstLoadingFixed = TRUE,
  centerBefore = NULL,
  centerAfter = NULL,
  residualsProds = NULL,
  residualCovSyntax = NULL,
  constrainedProdMean = NULL,
  constrainedLoadings = NULL,
  constrainedVar = NULL,
  constrainedResCovMethod = NULL,
  auto.scale = "none",
  auto.center = "none",
  estimator = "ML",
  removeFromParTable = NULL,
  addToParTable = NULL,
  macros = NULL,
  run = TRUE,
  ...
)

Arguments

modelSyntax

lavaan syntax

data

dataframe

method

method to use: "rca" = residual centering approach (passed to lavaan), "uca" = unconstrained approach (passed to lavaan), "dblcent" = double centering approach (passed to lavaan), "pind" = prod ind approach, with no constraints or centering (passed to lavaan), "lms" = laten model structural equations (not passed to lavaan). "qml" = quasi maximum likelihood estimation of laten model structural equations (not passed to lavaan). "custom" = use parameters specified in the function call (passed to lavaan)

match

should the product indicators be created by using the match-strategy

standardizeData

should data be scaled before fitting model

centerData

should data be centered before fitting model

firstLoadingFixed

Sould the first factorloading in the latent prod be fixed to one?

centerBefore

should inds in prods be centered before computing prods (overwritten by method, if method != NULL)

centerAfter

should ind prods be centered after they have been computed?

residualsProds

should ind prods be centered using residuals (overwritten by method, if method != NULL)

residualCovSyntax

should syntax for residual covariances be produced (overwritten by method, if method != NULL)

constrainedProdMean

should syntax prod mean be produced (overwritten by method, if method != NULL)

constrainedLoadings

should syntax for constrained loadings be produced (overwritten by method, if method != NULL)

constrainedVar

should syntax for constrained variances be produced (overwritten by method, if method != NULL)

constrainedResCovMethod

method for constraining residual covariances

auto.scale

methods which should be scaled automatically (usually not useful)

auto.center

methods which should be centered automatically (usually not useful)

estimator

estimator to use in lavaan

removeFromParTable

rows to remove from partable before sending to lavaan (for advanced users)

addToParTable

rows to add to partable before sending to lavaan (for advanced users)

macros

macros to replace in the syntax before sending to lavaan

run

should the model be estimated

...

arguments passed to other functions, e.g,. lavaan

Value

ModSEM object

Examples

library(modsem)
# For more examples check README and/or GitHub.
# One interaction
m1 <- '
  # Outer Model
  X =~ x1 + x2 +x3
  Y =~ y1 + y2 + y3
  Z =~ z1 + z2 + z3
  
  # Inner model
  Y ~ X + Z + X:Z 
'

# Double centering approach
est1 <- modsem(m1, oneInt)
summary(est1)

## Not run: 
# The Constrained Approach 
est1Constrained <- modsem(m1, oneInt, method = "ca")
summary(est1Constrained)

# LMS approach
est1LMS <- modsem(m1, oneInt, method = "lms")
summary(est1LMS)

# QML approach
est1QML <- modsem(m1, oneInt, method = "qml")
summary(est1QML)


## End(Not run)

# Theory Of Planned Behavior
tpb <- ' 
# Outer Model (Based on Hagger et al., 2007)
  LATT =~ att1 + att2 + att3 + att4 + att5
  LSN =~ sn1 + sn2
  LPBC =~ pbc1 + pbc2 + pbc3
  LINT =~ int1 + int2 + int3
  LBEH =~ b1 + b2

# Inner Model (Based on Steinmetz et al., 2011)
  # Covariances
  LATT ~~ LSN + LPBC
  LPBC ~~ LSN 
  # Causal Relationsships
  LINT ~ LATT + LSN + LPBC
  LBEH ~ LINT + LPBC 
  LBEH ~ LINT:LPBC  
'

# double centering approach
estTpb <- modsem(tpb, data = TPB)
summary(estTpb)

## Not run: 
# The Constrained Approach 
estTpbConstrained <- modsem(tpb, data = TPB, method = "ca")
summary(estTpbConstrained)

# LMS approach
estTpbLMS <- modsem(tpb, data = TPB, method = "lms")
summary(estTpbLMS)

## End(Not run)