Package semMediation

Package semMediation is an extension of package lavaan. The main functions of package semMediation are as follows:

makeEquation()

Consider a classical mediation setup with four variables: Y is the dependent variable, X is the predictor, and M1 and M2 are mediators. For illustration, we create a toy dataset containing these four variables, and fit a path analysis model that includes the direct effect of X on Y and the indirect effect of X on Y via M1 or M2 and second indirect effect via M1 and M2.

set.seed(1234)
X <- rnorm(100)
M1 <- 0.5*X + rnorm(100)
M2 <-0.6*X + rnorm(100)
Y <- 0.3*M1 + 0.4*M2 + rnorm(100)
data <- data.frame(X = X, Y = Y, M1 = M1, M2 = M2)
str(data)

'data.frame':   100 obs. of  4 variables:
 $ X : num  -1.207 0.277 1.084 -2.346 0.429 ...
 $ Y : num  -0.7323 -0.7088 0.3375 0.2245 0.0679 ...
 $ M1: num  -0.189 -0.336 0.608 -1.675 -0.611 ...
 $ M2: num  -0.239 0.863 0.836 -0.707 0.569 ...

To make a mediatino equation, you can use makeEquation() function.

require(lavaan)
require(semPlot)
require(semMediation)
model=makeEquation(X="X",M=c("M1","M2"),Y="Y")
cat(model)


 # Mediation Effect
Y~b1*M1+b2*M2+c1*X
M1~a1*X
M2~a2*X+d1*M1
ind1:=a1*b1
ind2:=a2*b2
secondInd1:=a1*d1*b2
total1:=c1+a1*b1+a2*b2+a1*d1*b2

fit=sem(model,data=data)
summary(fit)

lavaan (0.5-20) converged normally after  15 iterations

  Number of observations                           100

  Estimator                                         ML
  Minimum Function Test Statistic                0.000
  Degrees of freedom                                 0
  Minimum Function Value               0.0000000000000

Parameter Estimates:

  Information                                 Expected
  Standard Errors                             Standard

Regressions:
                   Estimate  Std.Err  Z-value  P(>|z|)
  Y ~                                                 
    M1        (b1)    0.449    0.100    4.483    0.000
    M2        (b2)    0.423    0.108    3.917    0.000
    X         (c1)    0.062    0.132    0.472    0.637
  M1 ~                                                
    X         (a1)    0.474    0.103    4.613    0.000
  M2 ~                                                
    X         (a2)    0.636    0.104    6.093    0.000
    M1        (d1)    0.088    0.092    0.957    0.339

Variances:
                   Estimate  Std.Err  Z-value  P(>|z|)
    Y                 1.046    0.148    7.071    0.000
    M1                1.054    0.149    7.071    0.000
    M2                0.898    0.127    7.071    0.000

Defined Parameters:
                   Estimate  Std.Err  Z-value  P(>|z|)
    ind1              0.213    0.066    3.215    0.001
    ind2              0.269    0.082    3.295    0.001
    secondInd1        0.018    0.019    0.911    0.362
    total1            0.562    0.121    4.652    0.000

To visualize model, you can use semPaths() function from semPlot package. But, semPaths() function is in appropriate for models with two or more mediators.

semPaths(fit)

mediationPlot() can visualize models with multiple mediators, multiple independent variables and/or multiple dependent variables.

mediationPlot(fit)

You can show standardized parameter estimates(default) or parameter estimates(est), or names(name) by settting the parameter “whatLabels”.

mediationPlot(fit,whatLabels = "name")

mediationPlot(fit,whatLabels = "est")

You can visualize indirect effects of this model.

mediationPlot(fit,regression=FALSE,correlation=FALSE,indirect=TRUE)

You can also visualize seconary indirect effect of this model.

mediationPlot(fit,regression=FALSE,correlation=FALSE,secondIndirect=TRUE)

Package semMediation

Keon-Woong Moon

2016-08-07

Installation

makeEquation()