GaB
GaB

Reputation: 1132

ggplot does not show me the legend

I am trying to add a legend on the plot but it is not working. Here is my example dataset:

structure(list(hour = c("0", "1", "2", "3", "4", "5", "6", "7", 
                    "8", "9", "10", "11", "12", "13", "14", "15", "16", 
                    "17", "18", "19", "20", "21", "22", "23"), 
           average = c(2.61835748792271, 2.11352657004831, 
                       1.71497584541063, 1.40338164251208, 
                       1.15700483091787, 1.86376811594203, 
                       1.83574879227053, 1.83478260869565, 
                       1.1256038647343,  1.7512077294686, 
                       2.4951690821256, 2.08695652173913, 
                       3.52898550724638,3.85990338164251, 
                       3.96376811594203, 4.00968523002421, 
                       3.9225181598063, 3.96610169491525, 
                       3.89588377723971, 3.95883777239709, 
                       3.81884057971014, 3.71497584541063, 4.5, 
                       3.08454106280193), 
          avg_arrivals = c(2.71428571428571,  1.91666666666667, 
                           1.30612244897959, 1.38, 1.85106382978723, 
                           1.79583333333333, 1.14285714285714, 
                           2.93877551020408, 3.33333333333333, 
                           4.82456140350877, 6.03448275862069, 
                           6.47368421052632, 6.53448275862069, 
                           6.48275862068965, 5.77586206896552, 
                           6.49122807017544, 6.37931034482759, 
                           5.89655172413793, 5.70689655172414, 
                           6.17241379310345, 5.77586206896552, 
                           4.27586206896552, 4.1551724137931, 
                           2.7719298245614)), 
          class = c("tbl_df", "tbl", "data.frame"), 
          row.names = c(NA, -24L))

Here is the R code for it and used ggplot2:

ggplot(data, aes(x = as.numeric(hour), y = percent_occ, 
                 group = hour)) +
  geom_bar(stat = "identity", alpha=0.7, width = 0.50, fill="bars" ) +
  geom_point(aes(y = percent_arrivals)) +
  geom_line(aes(y = percent_arrivals, group = 1, color = 
                  "line")) + # color = "line 
  theme(legend.position = "bottom", legend.box = "horizontal")

Upvotes: 0

Views: 3338

Answers (1)

Michael Harper
Michael Harper

Reputation: 15369

I am assuming you have misnamed the data in the provided example, with average being percent_occ and avg_arrivals being percent_arrivals. Running your example code results in the error:

Error in grDevices::col2rgb(colour, TRUE) : invalid color name 'bars'

This comes from the following:

geom_bar(stat = "identity", alpha=0.7, width = 0.50, fill="bars")

The problem is that ggplot is looking for the colour bars, which is not a valid colour name. If you instead want a colour to be assigned to the value based on a variable name, you need to specify it within the aes() argument.

Using this in your example:

ggplot(data, aes(x = as.numeric(hour), y = percent_occ, group = hour)) +
  geom_bar(stat = "identity", alpha=0.7, width = 0.50, aes(fill="bars")) +
  geom_point(aes(y = percent_arrivals)) +
  geom_line(aes(y = percent_arrivals, group = 1, color = "line"))

enter image description here

If you wish to define the colour of a variable which is mapped to an aesthetic using aes(), you will have to add a colour scale command to override the ggplot default pallete. For your example including these:

ggplot(data, aes(x = as.numeric(hour), y = percent_occ, group = hour)) +
  geom_bar(stat = "identity", alpha=0.7, width = 0.50, aes(fill="bars")) +
  geom_point(aes(y = percent_arrivals)) +
  geom_line(aes(y = percent_arrivals, group = 1, color = "line")) +
  scale_colour_manual(values = c("black")) +
  scale_fill_manual(values = c("yellow4"))

Upvotes: 1

Related Questions