Re-Label Axis and data points in Johnson-Neyman Plot

[jacob-long]

Re-Label Axis and data points in Johnson-Neyman Plot

[jacob-long]

Hi! I've transferred this issue to the new interactions package, where this function has moved.

As for your questions, it is possible to accomplish this using ggplot2, which is the package this function relies on to create the plot. Many people who are new (and often not so new) to R run into issues with ggplot2, so don't feel bad. I made the package because I know it can be hard to learn all this stuff!

I'll use some example code to show how to accomplish this. First, the axis titles should be fairly simple, using the ggplot2 functions xlab() and ylab().

library(interactions)
library(ggplot2)
states <- as.data.frame(state.x77)
states$HSGrad <- states$`HS Grad`
fit <- lm(Income ~ HSGrad + Murder * Illiteracy, data = states)
object <- johnson_neyman(model = fit, pred = Murder, modx = Illiteracy)
# the plot is stored in object$plot
object$plot + xlab("My new x-axis title") + ylab("My new y-axis title")

Relabeling the points along the x-axis is a little more complex, but doable. It will require you to know the mean of the variable on the x-axis because we're going to add that number to the existing labels. For demonstration purposes, we'll say the mean is 5. We make a function that adds the mean to its input and then give that to ggplot2's scale_x_continuous() function to the labels argument, telling it to apply that function to all the values on the x-axis.

# Creating a function that will edit the labels
new_labeller <- function(x) x + 5 # adds 5 
# still using the object from earlier
object$plot + scale_x_continuous(labels = new_labeller)

Putting those pieces together, then...

library(interactions)
library(ggplot2)
states <- as.data.frame(state.x77)
states$HSGrad <- states$`HS Grad`
fit <- lm(Income ~ HSGrad + Murder * Illiteracy, data = states)
object <- johnson_neyman(model = fit, pred = Murder, modx = Illiteracy)
# Creating a function that will edit the labels
new_labeller <- function(x) x + 5 # adds 5 
# the plot is stored in object$plot
object$plot + xlab("My new x-axis title") + ylab("My new y-axis title") +
  scale_x_continuous(labels = new_labeller)

[Mababyak]

Thanks so much for this, Jacob! Been trying to figure this out for some time! Maybe you can include this info on the vignettes page?

Really appreciate the package!

[zaslove]

Thanks! I too have been trying to figure this out. But this works!