在ggplot中重叠两个具有不同y轴的图

Question

I want to overlap two plots with different y-axis scales. I use stat_count() and geom_line . However, geom_line doesn't appear on the plot.

I use the following code.

library(ggplot2)

ggplot(X1, aes(x = Week)) + 
  stat_count() +
  scale_x_continuous(breaks = seq(from = 0, to = 21, by = 1))+
  scale_y_continuous(
    name = expression("Count"), 
    limits = c(0, 20),
    sec.axis = sec_axis(~ . * 15000 / 20, name = "Views"))+
  geom_line(aes(y = Views), inherit.aes = T)

Here is the reproducible example of my data frame X1 .

structure(list(Views = c(1749, 241, 309, 326, 237, 276, 2281, 
1573, 10790, 1089, 1732, 3263, 2601, 2638, 2929, 3767, 2947, 
65, 161), Week = c(1, 2, 2, 2, 3, 3, 4, 5, 5, 5, 6, 8, 8, 8, 
8, 9, 10, 10, 10)), row.names = c(NA, -19L), class = c("tbl_df", 
"tbl", "data.frame"))

Could you help me to put geom_line on the plot, please?

Answer 1

You also have to adjust the y values so that they fit inside the limits of the primary y-axis, ie apply the transfomation used for the secondary y-axis also inside geom_line . Try this:

X1 <- structure(list(Views = c(1749, 241, 309, 326, 237, 276, 2281, 
                         1573, 10790, 1089, 1732, 3263, 2601, 2638, 2929, 3767, 2947, 
                         65, 161), Week = c(1, 2, 2, 2, 3, 3, 4, 5, 5, 5, 6, 8, 8, 8, 
                                            8, 9, 10, 10, 10)), row.names = c(NA, -19L), class = c("tbl_df", 
                                                                                                   "tbl", "data.frame"))

library(ggplot2)

ggplot(X1, aes(x = Week)) + 
  stat_count() +
  scale_x_continuous(breaks = seq(from = 0, to = 21, by = 1))+
  scale_y_continuous(
    name = expression("Count"), 
    limits = c(0, 20),
    sec.axis = sec_axis(~ . * 15000 / 20, name = "Views"))+
  geom_line(aes(y = Views / 15000 * 20), inherit.aes = T)

^{Created on 2020-05-21 by the reprex package (v0.3.0)}

Answer 2

I also summarised the dataframe to improve the interpretation of the week 5 spike and plotted separate layers

x1 <- structure(list(Views = c(1749, 241, 309, 326, 237, 276, 2281, 
                               1573, 10790, 1089, 1732, 3263, 2601, 2638, 2929, 3767, 2947, 
                               65, 161), Week = c(1, 2, 2, 2, 3, 3, 4, 5, 5, 5, 6, 8, 8, 8, 
                                                  8, 9, 10, 10, 10)), row.names = c(NA, -19L), class = c("tbl_df", 
                                                                                                         "tbl", "data.frame"))

x2 <- x1 %>% 
  group_by(Week) %>% 
  summarise(Views = sum(Views))
library(ggplot2)

ggplot() + 
  geom_line(data = x2, mapping = aes(x = Week, y = Views/15000 * 20))+
  geom_bar(data = x1, mapping = aes(x = Week), stat = 'count')+
  scale_x_continuous(breaks = seq(from = 0, to = 21, by = 1))+
  scale_y_continuous( name = expression("Count"), 
                     ylim.prim <- c(0, 20),
                     ylim.sec <- c(0, 15000),
                     sec.axis = sec_axis(~ . * 15000 / 20, name = "Views"))