Plot of predicted probabilities of interaction term with sjPlot - CI level does not change

Question

I am trying to create a plot of predicted probabilities from a generalised multilevel linear model (random intercept) using sjPlot package. I want to plot the effect of an interaction term. I am attempting to change the C.I. level, which I assume is 95% by default (the documentation of the package is not very clear about this...). However, the C.I.s do not seem to change. I reproduced what I am trying to do using the iris dataset, to show what I mean. Below you can find the code:

library(patchwork)
library(lme4)
library(sjPlot)

data(iris)

iris$dummy <- rbinom(150, 1, 0.5)
iris$group <- round(runif(150, 1, 4), 0)

iris$Petal.Width <- as.factor(iris$Petal.Width)
iris$dummy <- as.factor(iris$dummy)
iris$group <- as.factor(iris$group)

model <- glmer(dummy ~ (1 | group) + Sepal.Width*Species, family = binomial, data = iris)

summary(model)

plot1 <- plot_model(model,
                    type = "pred",
                    terms = c("Sepal.Width[all]", "Species"))

plot2 <- plot_model(model,
                   type = "pred",
                   terms = c("Sepal.Width[all]", "Species"),
                   ci.lvl = 0.8)

plot1 + plot2

The two plots, plot1 and plot2 look exactly the same to me. How can I make this work? Thanks!

Answer 1

You want to pass the argument down to the ggeffects package, where the argument is ci_level.

library(sjPlot)
library(patchwork)
library(lme4)

plot1 <- plot_model(model,
                    type = "pred",
                    terms = c("Sepal.Width[all]", "Species"))

plot2 <- plot_model(model,
                    type = "pred",
                    terms = c("Sepal.Width[all]", "Species"),
                    ci_level = 0.8)

plot1 + plot2

^{Created on 2024-07-22 with reprex v2.1.0}