0

I need to plot a bar chart showing counts and a line chart showing the average. This all in one chart with multiple y axes.

Here is an example of the dataframe:

df <- data.frame("YearMonth" = c(20141, 20142, 20143, 20144), "Count" = c(435, 355, 360, 318), "Average" = c(107, 85, 86, 74))

How can this be done?

Thanks a lot.

SuGer
  • 11
  • 4
  • I would like to have two different y axes. With on the left the count and an secondary on the right, the average. – SuGer Mar 07 '19 at 15:08

1 Answers1

1

{ggplot2} intentionally does not support this kind of multiple y-axes because there’s a broad consensus that they are a bad idea because they invite misinterpretation of the data. See e.g. Why not to use two axes, and what to use instead:

We believe that charts with two different y-axes make it hard for most people to intuitively make right statements about two data series. We recommend two alternatives strongly: using two charts instead of one and using indexed charts.

The only kind of secondary y-axis that {ggplot2} supports is one that’s a scaling of the primary axis, see sec_axis. Such a secondary axis doesn’t suffer from the same problem, but it won’t work in your scenario: what you want is indeed one of the cases that {ggplot2} intentionally doesn’t support.

What you can do, though, is to duplicate a single y-axis, and overlay the bars and the averages (although it’s not clear what the bars represent in this case):

# fix the `YearMonth` column:
df$YearMonth = lubridate::ymd(paste(sub('(.)$', '0\\1', as.character(df$YearMonth)),'01'))

ggplot(df) +
    aes(YearMonth) +
    geom_col(aes(y = Count)) +
    geom_line(aes(y = Average), size = 2, color = 'lightblue') +
    scale_y_continuous(sec.axis = dup_axis(name = NULL))

dual y-axis

Konrad Rudolph
  • 530,221
  • 131
  • 937
  • 1,214