sem_effects: Structural Equation Model Effect Analysis and Visualization

Description

Provides standardized effect decomposition (direct, indirect, and total effects) for three major structural equation modeling frameworks: 'lavaan', 'piecewiseSEM', and 'plspm'. Automatically handles zero-effect variables, generates publication-ready 'ggplot2' visualizations, and returns both wide-format and long-format effect tables. Supports effect filtering, multi-model object inputs, and customizable visualization parameters.

Usage

sem_effects(
  object,
  target,
  plot = TRUE,
  delete_zero_effect = TRUE,
  total_only = FALSE,
  total_color = "skyblue",
  color_palette = c("darkgreen", "skyblue", "orange")
)

Value

A list containing three components:

effect_table: A data frame with variables and their standardized effect values (direct, indirect, total)
effect_long: A long-format version of effect_table
plot_object: A ggplot2 object (if plot=TRUE), NULL otherwise

Arguments

object: SEM object (lavaan/psem/plspm).
target: Character string specifying the target variable name for effect analysis.
plot: Logical indicating whether to generate effect visualization plots (default: TRUE).
delete_zero_effect: Logical indicating whether to removes rows where all specified effect columns contain only zeros (default: TRUE).
total_only: Logical controlling plot mode. If TRUE, shows only total effects with customizable colors; if FALSE, displays all effect types with palette coloring (default: FALSE).
total_color: Single color or vector of colors for total effect bars when total_only=TRUE (default: "skyblue").
color_palette: Character vector of 3 colors for direct/indirect/total effects when total_only=FALSE (default: c("darkgreen", "skyblue", "orange")).

Author

Weiping Mei

Examples

Run this code

# \donttest{

# Example 01: lavaan -------------------------------

library(lavaan)

model <- '
  # Measurement model
  ind60 =~ x1 + x2 + x3
  dem60 =~ y1 + y2 + y3 + y4
  dem65 =~ y5 + y6 + y7 + y8

  # Structural model
  dem60 ~ ind60
  dem65 ~ ind60 + dem60
'
fit <- sem(model, data = PoliticalDemocracy)

# Analyze effects for target variable "dem65"
results <- sem_effects(fit, target = "dem65")

print(results$effect_table)
print(results$effect_long)
print(results$plot_object)

# Customize plot appearance
results$plot_object +
  ggplot2::coord_flip()+
  ggplot2::theme_minimal() +
  ggplot2::ggtitle("Standardized effects for dem65")


# Example 02: piecewiseSEM --------------------------

library(piecewiseSEM)
pmod <- psem(
  lm(rich ~ cover, data = keeley),
  lm(cover ~ firesev, data = keeley),
  lm(firesev ~ age, data = keeley),
  data = keeley
  )

sem_effects(pmod, target = "rich",
        color_palette = c("darkgreen", "grey80", "purple"))


# Example 03: plspm ---------------------------------

library(plspm)
data(satisfaction)

# path matrix
IMAG = c(0,0,0,0,0,0)
EXPE = c(1,0,0,0,0,0)
QUAL = c(0,1,0,0,0,0)
VAL = c(0,1,1,0,0,0)
SAT = c(1,1,1,1,0,0)
LOY = c(1,0,0,0,1,0)
sat_path = rbind(IMAG, EXPE, QUAL, VAL, SAT, LOY)

# blocks of outer model
sat_blocks = list(1:5, 6:10, 11:15, 16:19, 20:23, 24:27)

# vector of modes (reflective indicators)
sat_mod = rep("A", 6)

# apply plspm
plsmodel = plspm(satisfaction, sat_path, sat_blocks, modes = sat_mod)

sem_effects(plsmodel, target = "LOY", plot = TRUE, delete_zero_effect = TRUE,
            total_only = TRUE,
            total_color = RColorBrewer::brewer.pal(5,"Set3"))

# }