plotting an average line in ggplot

Question

Plotting an average line in ggplot.

I have the following data;

structure(list(Region.in.country = c("Andalucia", "Aragon", "Asturias", 
"Canary Islands", "Cantabria", "Castilla-La Mancha", "Castilla y Leon", 
"Cataluna", "Comunidad Valenciana", "Extremadura", "Galicia", 
"Islas Baleares", "La Rioja", "Madrid", "Murcia", "Navarra", 
"Pais Vasco"), count = c(540L, 117L, 74L, 362L, 36L, 150L, 299L, 
952L, 797L, 72L, 283L, 353L, 39L, 1370L, 302L, 46L, 255L)), .Names = c("Region.in.country", 
"count"), row.names = c(NA, -17L), class = c("tbl_df", "tbl", 
"data.frame"), na.action = structure(18L, .Names = "18", class = "omit"))

I am trying to add an average line across the bar plot in ggplot 2. The average line is the avergae of the count column over the 17 regions.

sum(region$count) / 17

 ggplot(data = region, aes(x = Region.in.country, y = count)) +
   geom_bar(stat="identity") +
   geom_line(data = region, aes(355.7059)) +
   coord_flip()

The above code returns an error

EDIT:

try `geom_hline(aes(yintercept = 355.7059))` instead of `geom_line(...)`. (alas, this task is not as quixotic as it appears (one can't help but notice La Mancha in the `Region.in.county` column).) — bouncyball, Dec 28 '17 at 20:02
really... I have edited the original post and included my graphic — user113156, Dec 28 '17 at 20:03
Great! thanks bouncyball - it worked perfect! One additional question, is it possible to put the equation in the geom_hline. E.G. `geom_hline(aes(yintercept = sum(region$count) / nrow(region)))` — user113156, Dec 28 '17 at 20:06
Since you specified `data` in your `ggplot` call, you can succintly type: `geom_hline(aes(yintercept = mean(count)))` — bouncyball, Dec 28 '17 at 20:07
Exactly what I wanted, thanks! I tried the same thing but with `geom_line` which gave me the errors — user113156, Dec 28 '17 at 20:08

InfiniteFlash · Accepted Answer · 2018-05-11T14:11:35.940

4

This should do the job. Credit to bouncyball for suggesting aes(yintercept = mean(count)) instead of yintercept = 355.7059

  ggplot(region, aes(x= reorder(Region.in.country, count), count))+
    geom_bar(stat ="identity")+
    coord_flip()+
    xlab("Region")+
    ylab("Counts")+
    geom_hline(aes(yintercept = mean(count)))

If you want to create an ordered bar plot (by the numeric value), always remember to use reorder() on the column beforehand. It'll be unsorted otherwise even if you use arrange() or sort() to sort the data before plotting it. If you don't use reorder() on it, it'll be sorted by the corresponding id variable, Region.in.country in alphabetical order (as shown in the other answer posted after this one).

edited May 11 '18 at 14:11

answered Dec 28 '17 at 20:07

InfiniteFlash

1,038
1
10
22

1

Yes this works perfect also. Thank you. I like bouncyball solution of adding `geom_hline(aes(yintercept = mean(count)))` instead of the number 355.7059 – user113156 Dec 28 '17 at 20:11
Let me edit my post to accommodate for your adjustment (will give credit at top). Nice. – InfiniteFlash Dec 28 '17 at 20:11
Another interesting answer would be to check the `stat_summary()` function. As in `ggplot2::stat_summary(fun.y = mean, ggplot2::aes(x = 1, yintercept = ..y..), geom = "hline")`. More details can be found [on this SO question](https://stackoverflow.com/questions/52330348/plot-hline-at-mean-with-geom-bar-and-stat-identity) – xav Jun 14 '19 at 10:49

score 0 · Answer 2 · edited Dec 29 '17 at 02:08

0

Use geom_hline as follows:

avg <- mean(region$count)

region %>%
  ggplot(aes(x = Region.in.country, y = count)) +
  geom_bar(stat="identity") +
  geom_hline(aes(yintercept = avg)) +
  coord_flip()

edited Dec 29 '17 at 02:08

Tamás Sengel

55,884
29
169
223

answered Dec 28 '17 at 20:22

AlphaDrivers

136
4

plotting an average line in ggplot

2 Answers2