Problem with color changing in R using scale_fill

I have a dataset called temperature4countries

Year	Spain	Cyprus	Iceland	Austria
1998	15	17	7	8
1999	20	21	7.5	8.5
2000	16	18	7	8
2001	17	20	8	8
2002	17.5	19	8	8
2003	20	21	7.5	8.5
2004	20	22	8	9
2005	20	21	8.5	9.5
2006	21	27	9	10
2007	22	23	9.5	10.5
2008	25	24	9	11

temperature <- data.frame(
  stringsAsFactors = FALSE,
  Year= c(1998,1999,2000,2001,2002,2003,2004,2005,2006,2007,2008),
  Spain = c(15,20,16,17,17.5,20,20,20,21,22,25),
  Cyprus = c(17,21,18,20,19,21,22,21,27,23,24),
  Iceland = c(7,7.5,7,8,8,7.5,8,8.5,9,9.5,9),
  Austria = c(8,8.5,8,8,8,8.5,9,9.5,10,10.5,11),
  check.names = FALSE
)
dput(temperature)

My goal is to create a line graph

I used that code

df <- pivot_longer(temperature, 
                        cols = c(Spain, Cyprus, Iceland, Austria), 
                        values_to = "Temperature",
                        names_to = "Countries")
ggplot(df,
       aes(
         x = Year,
         y = Temperature,
         color = Countries
       ))  
  geom_line()   
  ggtitle("Trend of Temperature in 4 European countries")  
  xlab("Year")  
  ylab("Temperature[C]")

Again the colors are really ugly. I would like to change it. I used that code

df <- pivot_longer(temperature, 
                        cols = c(Spain, Cyprus, Iceland, Austria), 
                        values_to = "Temperature",
                        names_to = "Countries")
ggplot(df,
       aes(
         x = Year,
         y = Temperature,
         color = Countries
       ))  
  geom_line()   
  ggtitle("Trend of Temperature in 4 European countries")  
  xlab("Year")  
  ylab("Temperature[C]")    scale_fill_manual(values=c("aqua marine", "teal","lime","yellow"))

But the colors didn't change???

Because I thought that this is the way how to change colors

CodePudding user response：

You should use scale_color_manual in this case, instead of scale_fill_manual. Be explicit in your values param, passing a named vector:

  scale_color_manual(values=c("Austria" = "red", "Cyprus" = "blue",
                             "Iceland" = "green","Spain" = "orange"))

Check colors() for named colors available. You could use "#008080" in place of "teal" and "limegreen" in place of "lime", if you like.

CodePudding user response：

The reason why the color doesn't change is because you cannot use scale_fill_manual. Change it to scale_color_manual and you will see the change.

ggplot(df,
       aes(
         x = Year,
         y = Temperature,
         color = Countries
       ))  
  geom_line()   
  ggtitle("Trend of Temperature in 4 European countries")  
  xlab("Year")  
  ylab("Temperature[C]")   scale_color_manual(values=c("aqua marine", "red","blue","yellow"))

Note: that I had to change some colors because it couldn't find "lime" and "teal".

PS. The difference between them is that:

scale_fill_manual() is used for plots where you have things to fill, e.g. box plots, bar plots, violin plots, dot plots, etc.
scale_color_manual() is used for lines and points.

Here more info.

CodePudding user response：

You're well on your way! You are making three mistakes:

You added twice in a row in your last line of code
You are using scale_fill_manual(...), but the aesthetic that is being used in the plot is color. Thus, the correct function would be scale_colour_manual(...).
The colours that you specified are not R colours. You can get an overview of all named colours in R using colors(). Check an individual colour using e.g. "aquamarine" %in% colors()

Thus, the correct code would be:

ggplot(df,
       aes(
         x = Year,
         y = Temperature,
         color = Countries
       ))  
  geom_line()   
  ggtitle("Trend of Temperature in 4 European countries")  
  xlab("Year")  
  ylab("Temperature[C]")   scale_colour_manual(values=c("aquamarine", "cyan4","limegreen","yellow"))