analyze_path: Compute the moderated paths coefficients for analayses of...

In NicolasJBM/modestim: Functions and interfaces to build, estimate, and report complex models

Description

Compute the moderated paths coefficients for analayses of moderated mediations.

Usage

analyze_path(coefficients, moderations, mediations)

Arguments

coefficients	Tibble. Output of the function compute_coefficients.
moderations	Tibble. Output of the function retrieve_modval with option "allval" FALSE.
mediations	List. List of vectors specifying the paths, from the first independent variable to the last dependent variable through the moderators in order.

Value

A tibble gathering all (direct, indirect, and total) the moderated path coefficients.

References

Edwards, J. R., & Lambert, L. S. (2007). Methods for integrating moderation and mediation: a general analytical framework using moderated path analysis. Psychological Methods, 12(1), 1–22.

Examples

# Prepare the specifications for subsequent analyses

library(dplyr)
library(tibble)

sample <- modestim::mm_edwards2007 %>%
  mutate(
    fbkc.x.cenc = fbkc * cenc,
    satc.x.cenc = satc * cenc,
    fbkc.x.gen = fbkc * gen,
    satc.x.gen = satc * gen
  )

moderations <-  modestim::retrieve_modval(
  sample,
  moderators = c("gen","cenc"),
  types = c("discrete","continuous"),
  allval = FALSE
)

mediations <- list(
  c("fbkc","satc","comc")
)

# Replicate Edwards and Lambert's 2007 analysis of moderated mediation with gender


specgen <- tibble::tibble(
  sample = rep(list(sample),2),
  formula = list(
    as.formula("satc ~ fbkc + gen + fbkc.x.gen"),
    as.formula("comc ~ satc + fbkc + gen + satc.x.gen + fbkc.x.gen")
  ),
  method = "glm",
  family = NA,
  md_imputation = "deletion",
  md_method = "deletion",
  draws = 10,
  size = NA,
  cores = 2
)

coeffgen <- modestim::compute_coefficients(specgen)

pathanagen <- modestim::analyze_path(coeffgen, moderations, mediations)


# Replicate Edwards and Lambert's 2007 analysis of moderated mediation with centrality

speccenc <- tibble::tibble(
  sample = rep(list(sample),2),
  formula = list(
    as.formula("satc ~ fbkc + cenc + fbkc.x.cenc"),
    as.formula("comc ~ satc + fbkc + cenc + satc.x.cenc + fbkc.x.cenc")
  ),
  method = "glm",
  family = NA,
  md_imputation = "deletion",
  md_method = "deletion",
  draws = 10,
  size = NA,
  cores = 2
)
 
 coeffcenc <- modestim::compute_coefficients(speccenc)
 
 pathanacenc <- modestim::analyze_path(coeffcenc, moderations, mediations)
 

NicolasJBM/modestim documentation built on Aug. 26, 2019, 5:29 a.m.