Lavaan.survey output help: what am I looking at?

[H. Kberg]

Hi all,

Im new with mediation analysis. Currently Im trying to do a really simple one with weighted survey data, so Im using lavaan,survey.

I have one mediator.

What does this output exactly telling me?

Cheers

My code is:

#Define model

modellvm <- ' # direct effect

LVMass ~ c*METquantthird

# mediator

ATPIIImets ~ a*METquantthird

LVMass ~ b*ATPIIImets

# indirect effect (a*b)

ab := a*b

# total effect

total := c + (a*b)'

fitlvm <- lavaan(modellvm, data = finalNEOtrailMET, ordered=c("ATPIIImets"), auto.var = T, std.lv = T, meanstructure = T, int.ov.free = T)

fitlvm.surv <- lavaan.survey(lavaan.fit = fitlvm, survey.design=designweightsMET)

summary(fitlvm.surv)

with output:

lavaan 0.6-5 ended normally after 39 iterations

  Estimator                                         ML

  Optimization method                           NLMINB

  Number of free parameters                          7

                                                      

  Number of observations                           916

                                                      

Model Test User Model:

                                              Standard      Robust

  Test Statistic                                 0.000       0.000

  Degrees of freedom                                 0           0

Parameter Estimates:

  Information                                 Expected

  Information saturated (h1) model          Structured

  Standard errors                           Robust.sem

Regressions:

                   Estimate  Std.Err  z-value  P(>|z|)

  LVMass ~                                            

    METqntthrd (c)    5.858    1.390    4.214    0.000

  ATPIIImets ~                                        

    METqntthrd (a)   -0.059    0.020   -2.908    0.004

  LVMass ~                                            

    ATPIIImets (b)   11.258    2.322    4.849    0.000

Intercepts:

                   Estimate  Std.Err  z-value  P(>|z|)

   .LVMass           85.395    3.089   27.646    0.000

   .ATPIIImets        0.352    0.044    8.053    0.000

Variances:

                   Estimate  Std.Err  z-value  P(>|z|)

   .LVMass          633.419   40.489   15.644    0.000

   .ATPIIImets        0.177    0.009   20.296    0.000

Defined Parameters:

                   Estimate  Std.Err  z-value  P(>|z|)

    ab               -0.663    0.262   -2.532    0.011

    total             5.196    1.406    3.697    0.000

[H. Kberg]

Btw 

METquantthird = exogenous ordinal variable with 3 levels (1,2,3)
LVMass  = continuous variable

ATPIIImets = endogenous dichotome variable

Op zondag 5 januari 2020 18:43:53 UTC+1 schreef H. Kberg:

[Terrence Jorgensen]

METquantthird = exogenous ordinal variable with 3 levels (1,2,3)
LVMass  = continuous variable

ATPIIImets = endogenous dichotome variable

If you have a binary mediator, you can't use lavaan.survey.  Well, you "can" use it like you did, by pretending the mediator and predictor variables are continuous, but the survey package does not have the functionality needed to adequately account for the categorical nature of ordered/binary variables.

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam

http://www.uva.nl/profile/t.d.jorgensen