calEffSizes: Calculate Effect Sizes using lavaan Models

In metaSEM: Meta-Analysis using Structural Equation Modeling

Calculate Effect Sizes using lavaan Models

Description

It calculates effect sizes with Delta Method by formulating the effect sizes as functions of SEM in lavaan.

Usage

calEffSizes(model, data=NULL, n, Cov, Mean=NULL, group=NULL, lavaan.output=FALSE, ...)

Arguments

model	A lavaan model. Effect sizes are defined as functions of SEM parameters with :=.
data	A data frame of the observed variables. If it is NULL, summary statistics are required.
n	Sample sizes
Cov	A covariance matrix or a list of covariance matrices.
Mean	Optional sample means.
group	A character of the variable name in the data frame defining the groups in a multiple group analysis.
lavaan.output	If TRUE, it returns the fitted object instead of the effect sizes and their sampling covariance matrix.
...	Further arguments passed to sem.

Value

Effect sizes and their sampling covariance matrix or a lavaan fitted object.

Note

The input matrices are treated as covariance matrices unless there are explicit constraints in the model.

Author(s)

Mike W.-L. Cheung <[e-mail]>

References

Cheung, M. W.-L. (2015). Meta-analysis: A structural equation modeling approach. Chichester, West Sussex: John Wiley & Sons, Inc.

Cheung, M. W.-L. (2018). Computing multivariate effect sizes and their sampling covariance matrices with structural equation modeling: Theory, examples, and computer simulations. Frontiers in Psychology, 9(1387). https://doi.org/10.3389/fpsyg.2018.01387

See Also

smdMES, smdMTS

Examples

## Not run:  
## Select ATT, Bi, and BEH 
obs.vars <- c("BEH", "BI", "ATT")

## Select one study from Cooke16 for illustration
my.cor <- Cooke16$data[[4]][obs.vars, obs.vars]
my.n  <- Cooke16$n[4]

## Effect sizes: indirect effect and direct effect
model <- "BEH ~ c*ATT + b*BI
          BI ~ a*ATT
          ## Indirect effect
          Ind := a*b
          Dir := c"

calEffSizes(model=model, n=my.n, Cov=my.cor, lavaan.output=FALSE)

## Return the lavaan fitted model
fit <- calEffSizes(model=model, n=my.n, Cov=my.cor, lavaan.output=TRUE)
lavaan::summary(fit)
    
lavaan::parameterestimates(fit)  

## End(Not run)

metaSEM documentation built on Aug. 10, 2023, 1:09 a.m.