如何调整 effect_plot function 中的 y 轴范围

Question

I'm trying to plot raw outcome values against variable insulin , with a regression line on top. I followed the instruction here https://cran.r-project.org/web/packages/jtools/vignettes/effect_plot.html but I have a question about adjusting y-axis.

Do you know how may I customize my y-axis scales? For example, how can I set y-axis scale ranged from -10 to 0, with -2 increment? Sorry for this silly question, I'm new to R and most code I googled did not seem to work with the effect_plot() function. Below is my code:

pct_fat <- lm(pct_fat_change ~ DXA_age + female + race + hispanic + insulin, data = example)


effect_plot(example, pred = insulin, interval = TRUE, plot.points = TRUE, colors = "red")

Many thanks in advance!

Answer 1

Here's a toy example to change the y_axis limits.

effect_plot returns a ggplot object, so assign it to a variable.

library(ggplot2)
require(jtools)

data(mpg)

fit <- lm(cty ~ displ + year + cyl + class + fl, data = mpg[mpg$fl != "c",])

g <- effect_plot(fit, pred = displ, interval = TRUE, plot.points = TRUE, colors = "red")

After that, you can work on it as a ggplot object, here below 2 distinct way to achieve your result. (Suppose you want to rescale the y_axis from 20 to 32)

g +
  scale_y_continuous(limits=c(20, 32))

g+
  ylim(20, 32)