Confidence Intervals of Indirect Effects in an 'indirect_list' Object

Return the confidence intervals of the indirect effects stored in the output of many_indirect_effects().

Usage

# S3 method for class 'indirect_list'
confint(object, parm = NULL, level = NULL, ...)

Arguments

object

The output of many_indirect_effects().

parm

Ignored for now.

level

If set to NULL, the default, then the level of confidence used to generate object is used. If set to a value, this value will be used to recompute the confidence intervals. If the confidence interval is to be computed from the standard error, and so level is not set in object, then the default value is .95. (This new behavior applies to 0.3.6.15 and later version.)

...

Additional arguments. To be passed to confint.indirect(). (This new behavior applies to 0.3.6.15 and later version.)

Value

A two-column data frame. The columns are the limits of the confidence intervals.

Details

It extracts and returns the stored confidence interval if available.

The type of confidence intervals depends on the call used to compute the effects. This function merely retrieves the stored estimates, which could be generated by nonparametric bootstrapping, Monte Carlo simulation, or other methods to be supported in the future, and uses them to form the percentile confidence interval.

See also

many_indirect_effects()

Examples

library(lavaan)
data(data_serial_parallel)
mod <-
"
m11 ~ x + c1 + c2
m12 ~ m11 + x + c1 + c2
m2 ~ x + c1 + c2
y ~ m12 + m2 + m11 + x + c1 + c2
"
fit <- sem(mod, data_serial_parallel,
           fixed.x = FALSE)
# All indirect paths from x to y
paths <- all_indirect_paths(fit,
                           x = "x",
                           y = "y")
paths
#> Call: 
#> all_indirect_paths(fit = fit, x = "x", y = "y")
#> Path(s): 
#>   path                
#> 1 x -> m11 -> m12 -> y
#> 2 x -> m11 -> y       
#> 3 x -> m12 -> y       
#> 4 x -> m2 -> y        
# Indirect effect estimates
# R should be 2000 or even 5000 in real research
# parallel should be used in real research.
fit_boot <- do_boot(fit, R = 45, seed = 8974,
                    parallel = FALSE,
                    progress = FALSE)
out <- many_indirect_effects(paths,
                             fit = fit,
                             boot_ci = TRUE,
                             boot_out = fit_boot)
out
#> 
#> == Indirect Effect(s) ==
#> 
#>                        ind  CI.lo CI.hi Sig
#> x -> m11 -> m12 -> y 0.193  0.029 0.550 Sig
#> x -> m11 -> y        0.163 -0.346 0.570    
#> x -> m12 -> y        0.059 -0.156 0.208    
#> x -> m2 -> y         0.364  0.130 0.889 Sig
#> 
#>  - [CI.lo to CI.hi] are 95.0% percentile confidence intervals by
#>    nonparametric bootstrapping with 45 samples.
#>  - The 'ind' column shows the indirect effect(s).
#>  
confint(out)
#>                      Percentile: 2.5 % Percentile: 97.5 %
#> x -> m11 -> m12 -> y        0.02866626          0.5501760
#> x -> m11 -> y              -0.34617725          0.5702562
#> x -> m12 -> y              -0.15615492          0.2081817
#> x -> m2 -> y                0.12961514          0.8892807