mlm_plot: Publication-ready interaction plot for multilevel models

In mlmoderator: Probing, Plotting, and Interpreting Multilevel Interaction Effects

Publication-ready interaction plot for multilevel models

Description

Creates a ggplot2-based interaction plot showing predicted values of the outcome across levels of pred, with separate lines for each selected value of modx. Confidence bands and raw data overlay are optional.

Usage

mlm_plot(
  model,
  pred,
  modx,
  modx.values = c("mean-sd", "quartiles", "tertiles", "custom"),
  at = NULL,
  interval = TRUE,
  conf.level = 0.95,
  points = FALSE,
  point_alpha = 0.3,
  colors = NULL,
  line_size = 1,
  x_label = NULL,
  y_label = NULL,
  legend_title = NULL
)

Arguments

model	An lmerMod object with a two-way interaction between pred and modx in the fixed-effects.
pred	Character scalar. Focal predictor (x-axis).
modx	Character scalar. Moderator (separate lines).
modx.values	Strategy for moderator values. Same options as mlm_probe(): "mean-sd", "quartiles", "tertiles", "custom".
at	Numeric vector of custom moderator values (used when modx.values = "custom").
interval	Logical. Draw confidence bands? Default TRUE.
conf.level	Confidence level for bands. Default 0.95.
points	Logical. Overlay raw data points? Default FALSE.
point_alpha	Transparency for raw data points. Default 0.3.
colors	Character vector of colours for moderator lines. If NULL, uses a accessible default palette.
line_size	Line width for predicted lines. Default 1.
x_label	Label for x-axis. Defaults to pred.
y_label	Label for y-axis. Defaults to response variable name.
legend_title	Label for the legend. Defaults to modx.

Value

A ggplot object.

Examples

set.seed(1)
dat <- data.frame(
  y   = rnorm(200),
  x   = rnorm(200),
  m   = rep(rnorm(20), each = 10),
  grp = factor(rep(1:20, each = 10))
)
dat$y <- dat$y + dat$x * dat$m
mod <- lme4::lmer(y ~ x * m + (1 | grp), data = dat,
                  control = lme4::lmerControl(optimizer = "bobyqa"))
mlm_plot(mod, pred = "x", modx = "m")
mlm_plot(mod, pred = "x", modx = "m", modx.values = "quartiles")

mlmoderator documentation built on April 4, 2026, 1:07 a.m.