Learn R Programming

metaSEM (version 2.0)

issp89: A Dataset from Cheung and Chan (2005; 2009)

Description

Eleven covariance matrices on work-related attitudes were extracted from the Inter-University Consortium for Political and Social Research (1989). Nine variables were selected by Cheung and Chan (2005; 2009) for demonstration purposes. They were grouped into three constructs: Job Prospects measured by job security (JP1), income (JP2), and advancement opportunity (JP3); Job Nature measured by interesting job (JN1), independent work (JN2), help other people (JN3), and useful to society (JN4); and Time Demand measured by flexible working hours (TD1) and lots of leisure time (TD2).

Arguments

Author

Mike W.-L. Cheung <[email protected]>

Details

A list of data with the following structure:

data

A list of 11 studies of covariance matrices

n

A vector of sample sizes

References

Cheung, M. W.-L., & Chan, W. (2005). Meta-analytic structural equation modeling: A two-stage approach. Psychological Methods, 10, 40-64.

Cheung, M. W.-L., & Chan, W. (2009). A two-stage approach to synthesizing covariance matrices in meta-analytic structural equation modeling. Structural Equation Modeling, 16, 28-53.

See Also

issp05

Examples

Run this code

# \donttest{
data(issp89)

#### Analysis of correlation structure in Cheung and Chan (2005)
#### Fixed-effects model: Stage 1 analysis
cor1 <- tssem1(issp89$data, issp89$n, method="FEM", cor.analysis=TRUE)
summary(cor1)

## Three-factor CFA model in lavaan syntax
## f1: Job Prospects (JP1-JP3); f2: Job Nature (JN1-JN4); f3: Time Demand (TD1-TD2)
model <- "## Factor loadings
          f1 =~ f1JP1*JP1 + f1JP2*JP2 + f1JP3*JP3
          f2 =~ f2JN1*JN1 + f2JN2*JN2 + f2JN3*JN3 + f2JN4*JN4
          f3 =~ f3TD1*TD1 + f3TD2*TD2
          ## Factor correlations
          f1 ~~ corf2f1*f2
          f1 ~~ corf3f1*f3
          f2 ~~ corf3f2*f3"

RAM1 <- lavaan2RAM(model, obs.variables=c("JP1","JP2","JP3",
                                          "JN1","JN2","JN3","JN4",
                                          "TD1","TD2"))
RAM1

## ## Equivalent RAM specification using create.mxMatrix()
## Phi <- create.mxMatrix( c("0.3*corf2f1","0.3*corf3f1","0.3*corf3f2"),
##                         type="Stand", as.mxMatrix=FALSE )
## Psi <- create.mxMatrix( paste("0.2*e", 1:9, sep=""), type="Diag",
##                         as.mxMatrix=FALSE )
## S1 <- bdiagMat(list(Psi, Phi))
## S1 <- as.mxMatrix(S1)
##
## Lambda <- create.mxMatrix( c(".3*f1x1",".3*f1x2",".3*f1x3",rep(0,9),
##                              ".3*f2x4",".3*f2x5",".3*f2x6",".3*f2x7",
##                              rep(0,9),".3*f3x8",".3*f3x9"), type="Full",
##                              ncol=3, nrow=9, as.mxMatrix=FALSE )
## Zero1 <- matrix(0, nrow=9, ncol=9)
## Zero2 <- matrix(0, nrow=3, ncol=12)
## A1 <- rbind( cbind(Zero1, Lambda), Zero2 )
## A1 <- as.mxMatrix(A1)
##
## F1 <- create.Fmatrix(c(rep(1,9), rep(0,3)))

#### Fixed-effects model: Stage 2 analysis
cor2 <- tssem2(cor1, RAM=RAM1, intervals.type="LB")
summary(cor2)

## Display the model with the parameter estimates
plot(cor2, nDigits=1)

#### Analysis of covariance structure in Cheung and Chan (2009)
#### Fixed-effects model: Stage 1 analysis
cov1 <- tssem1(issp89$data, issp89$n, method="FEM", cor.analysis=FALSE)
summary(cov1)

#### Fixed-effects model: Stage 2 analysis
cov2 <- tssem2(cov1, RAM=RAM1)
summary(cov2)

## Display the model with the parameter estimates
plot(cov2, nDigits=1)
# }

Run the code above in your browser using DataLab