Plot the datapoints in a plot_model function

Question

I'm trying to add the datapoints using geom_point() to the plot_model() that is showing the linear relationship between the age and left.LGN variables.

plot <- plot_model(m1, terms = c("age"), type = "pred") +
    geom_point(data = data, aes(x = age, y = Left.LGN)) +
    labs(x = "Age (standardized)", y = "Adj. Left-LGN (standardized volume)", size = 20)

The script works until I add the third variable ("sex"):

plot <- plot_model(m1, terms = c("age", "sex"), type = "pred") +
   geom_point(data = data, aes(x = age, y = Left.LGN, color = sex, group = sex)) +
   labs(x = "Age (standardized)", y = "Adj. Left-LGN (standardized volume)", size = 20)

in which I got this error:

Error in `geom_point()`:
! Problem while computing aesthetics.
ℹ Error occurred in the 3rd layer.
Caused by error in `FUN()`:
! object 'group_col' not found
Run `rlang::last_error()` to see where the error occurred.

Has anybody had the same issue?

I expect to have a plot showing the relationship between age and Left-LGN volume related to males and females, adding the datapoints from the data.

Allan Cameron · Accepted Answer · 2023-09-14 10:02:12Z

We don't have your data or model, but we can reproduce your problem with the following set-up:

library(sjPlot)
library(ggplot2)

set.seed(1)

data <- data.frame(age = rep(31:70, 2),
                   Left.LGN = c(31:70, 41:80)/6 + 100 + rnorm(80),
                   sex = rep(c('Male', 'Female'), each = 40))

m1 <- lm(Left.LGN ~ age + sex, data)

Now there is no problem using your own plotting code to generate a plot with just age:

plot_model(m1, terms = c("age"), type = "pred") +
  geom_point(data = data, aes(x = age, y = Left.LGN)) +
  labs(x = "Age (standardized)", 
       y = "Adj. Left-LGN (standardized volume)", 
       size = 20)

But if we try to add in the sex variable, we get the same error that you report:

plot_model(m1, terms = c("age", "sex"), type = "pred") +
  geom_point(data = data, 
             aes(x = age, y = Left.LGN, color = sex, group = sex)) +
  labs(x = "Age (standardized)", 
       y = "Adj. Left-LGN (standardized volume)", 
       size = 20)
#> Error in `geom_point()`:
#> ! Problem while computing aesthetics.
#> i Error occurred in the 3rd layer.
#> Caused by error in `FUN()`:
#> ! object 'group_col' not found

The reason for the error is that plot_model generates a ggplot object, which has aesthetic mappings to variables present in the data frame that plot_model creates from your model. One of these variables is called group_col, which of course is not present in your own data frame. To fix this, simply specify inherit.aes = FALSE in your geom_point layer:

plot_model(m1, terms = c("age", "sex"), type = "pred") +
  geom_point(data = data, 
             aes(x = age, y = Left.LGN, color = sex, group = sex),
             inherit.aes = FALSE) +
  labs(x = "Age (standardized)", 
       y = "Adj. Left-LGN (standardized volume)", 
       size = 20)

Note though that plot_model also includes the parameter show.data if you want to include the data points in your model, so instead of adding a geom_point layer, you can do:

plot_model(m1, terms = c("age", "sex"), type = "pred", show.data = TRUE) +
  labs(x = "Age (standardized)", 
       y = "Adj. Left-LGN (standardized volume)", 
       size = 20)

^{Created on 2023-09-14 with reprex v2.0.2}

Collectives™ on Stack Overflow

Plot the datapoints in a plot_model function

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related