Apply custom model labels in ggplot2

Question

This is likely trivial, but how can I:

Apply labels to each modelnum value on the y axis? Could just be named "Label1" for "a", "Label2" for "b" or something like that.
Reverse the order alphabetically within each facet, so that the letter higher in the alphabet is at the top of a given facet?

Code:

library(ggplot2)
ggplot(d, aes(y = modelnum, x = odds_ratio, color = sample,
               xmin = ci_lower, xmax = ci_upper)) +
  geom_errorbar(width = .1) +
  geom_point(size = 3) +
  geom_vline(xintercept = 1, 
             linetype = "dashed", 
             color = "black",
             linewidth = .3) +
  facet_wrap(~sample,
             scales = "free_y",
             nrow = 3) +
  scale_color_manual(values = c("Never" = "gray30",
                                "Pooled" = "red",
                                "Post" = "blue")) +
  labs(title = "Models",
       x = "Odds Ratio",
       y = "") +
  theme_bw() +
  theme(legend.position = "none"
        )

Data:

d <- structure(list(modelnum = c("a", "b", "c", "a", "b", "d", "e", 
"f"), odds_ratio = c(0.11, 0.151, 0.0629, 0.193, 0.052, 0.1065, 
0.2, 0.089), ci_lower = c(0.048, 0.054, 0.0049, 0.052, 0.004, 
0.0566, 0.067, 0.017), ci_upper = c(0.271, 0.425, 0.799, 0.646, 
0.72, 0.19, 0.5948, 0.461), sample = c("Pooled", "Pooled", "Pooled", 
"Never", "Never", "Never", "Post", 
"Post")), class = c("tbl_df", 
"tbl", "data.frame"), row.names = c(NA, -8L))

PBulls · Accepted Answer · 2023-11-08 14:38:26Z

1

You can do this very similar to how you mappped the colors, just add this somewhere in your ggplot:

+ scale_y_discrete(labels = c("a"="Label1", "b"="Label2"), limits=rev)

The limits=rev option will reverse a discrete axis. You could also use numeric values and map the breaks/labels to those for more fine-grained control over order.

answered Nov 8, 2023 at 14:38

PBulls

2,8291 gold badge11 silver badges24 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Friede · Accepted Answer · 2023-11-08 14:51:03Z

Just one way to achieve what you are after. Create a label variable and reorder y inside aes() accordingly, i.e., y = reorder(label, desc(modelnum)).

library(ggplot2); library(dplyr)
d |>
  mutate(label = paste0("Label", modelnum)) |>
  ggplot(aes(y = reorder(label, desc(modelnum)), x = odds_ratio, color = sample,
              xmin = ci_lower, xmax = ci_upper)) +
  geom_errorbar(width = .1) +
  geom_point(size = 3) +
  geom_vline(xintercept = 1, 
             linetype = "dashed", 
             color = "black",
             linewidth = .3) +
  facet_wrap(~ sample,
             scales = "free_y",
             nrow = 3) +
  scale_color_manual(values = c("Never" = "gray30",
                                "Pooled" = "red",
                                "Post" = "blue")) +
  labs(title = "Models",
       x = "Odds Ratio",
       y = "") +
  theme_bw() +
  theme(legend.position = "none")

Collectives™ on Stack Overflow

Apply custom model labels in ggplot2

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related