Multi-row x-axis labels with facet_grid where each facet has unique labels

Question

I am trying to create grouping x-axis labels to put above existing x-axis labels; recent browsing on StackOverflow recommends facet_grid for this purpose. However, I am unable to modify individual x-axis labels for each facet. Reprex below:

library(tidyverse)

# make data
scale <- c("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o")
score <- rep(10, 15)
input <- data.frame(scale, score)
input <- t(input)
names(input) <- input[, 1]
input <- data.frame(janitor::row_to_names(input, row_number = 1))

# put it in a dataframe
df <- data.frame(rbind(
  input$a, input$b, input$c, input$d,
  input$e, input$f, input$g, input$h, input$i,
  input$j, input$k, input$l,
  input$m, input$n, input$o)) %>%
  tibble::rownames_to_column(var = "scale") %>%
  rename(scale_score = 2) %>%
  mutate(scale_group = c("SCALE1", "SCALE1", "SCALE1", "SCALE1",
                         "SCALE2", "SCALE2", "SCALE2", "SCALE2", "SCALE2", 
                         "SCALE3", "SCALE3", "SCALE3", 
                         "SCALE4", "SCALE4", "SCALE4"))

df$scale_group = factor(df$scale_group, levels = c("SCALE1", "SCALE2", "SCALE3", "SCALE4"))

df$gp <- c(1,1,1,1,2,2,2,2,2,3,3,3,4,4,4)

# make ggplot labels
labels = c("A", "B", "C", "(D)",
                    "E", "F", "G", "(H)", "(I)",
                    "J", "K", "(L)",
                    "M", "N", "(O)")

# make a ggplot with facet_grid and it looks weird - why do the sub-labels repeat themselves?
df %>%
  ggplot(aes(x=interaction(scale, factor(scale_group)), y=as.numeric(scale_score), 
             label = (scale_group),
             group=interaction(scale_group, gp))) +
  theme_bw() +
  geom_line(stat="identity") + 
  geom_point(size=2, color="blue") +
  geom_rect(aes(xmin = -Inf, xmax = Inf, ymin = 9.5, ymax = 10.5),
            alpha = 0.015, fill = "darkturquoise") +
  scale_y_continuous(breaks = (seq(0, 20, by = 1)), limits = c(0,20)) +
  scale_x_discrete(position = "top", labels = unique(labels)) +
  xlab("") + ylab("") +
  ggtitle("I want to see all the letters, not ABC repeating") + theme(plot.title = element_text(hjust = 0.5)) +
  facet_grid( ~ scale_group,
              scales = "free", space = "free")

I would also like SCALE1/SCALE2/etc. to be above A, B, C, etc., but that is likely a separate quesiton. I would appreciate any tips or insights.

Answer 1

The following works for me. First, make the labels vector with the name from your x-axis label. Second, remove unique from the labels = unique(labels) .

# make ggplot labels
labels <-  c("A", "B", "C", "(D)",
             "E", "F", "G", "(H)", "(I)",
             "J", "K", "(L)",
             "M", "N", "(O)")
names(labels) <- interaction(df$scale, factor(df$scale_group))

df %>%
  ggplot(aes(x=interaction(scale, factor(scale_group)), y=as.numeric(scale_score), 
             label = (scale_group),
             group=interaction(scale_group, gp))) +
  theme_bw() +
  geom_line(stat="identity") + 
  geom_point(size=2, color="blue") +
  geom_rect(aes(xmin = -Inf, xmax = Inf, ymin = 9.5, ymax = 10.5),
            alpha = 0.015, fill = "darkturquoise") +
  scale_y_continuous(breaks = (seq(0, 20, by = 1)), limits = c(0,20)) +
  scale_x_discrete(position = "top", labels = labels) +
  xlab("") + ylab("") +
  ggtitle("I want to see all the letters, not ABC repeating") + theme(plot.title = element_text(hjust = 0.5)) +
  facet_grid( ~ scale_group,
              scales = "free", space = "free")