add a simple legend on ggplot with linetype

Question

How do I add a legend to this ggplot? I've searched everywhere but couldn't find a simple way that matches my current code. I've got three variable, they all have the same colour with different linetypes.

ggplot()+geom_line(data=datapop, aes(Year, OECD),size = 0.7, color="#69b3a2") +
    geom_line(data=datapop, aes(Year, World),size = 0.7, color="#69b3a2", linetype="dashed")+
    geom_line(data=datapop, aes(Year, Switzerland),size = 0.7, color="#69b3a2", linetype="twodash")+
    xlab("Years")+ ylab("Aging") +theme_minimal()+ labs(color="Legend text")+
    scale_x_date(date_breaks = "2 years",date_labels = "%Y")+
    theme(axis.text.x=element_text(angle=60, hjust=1))

Answer 1

Another approach would be the data.table solution. Naturally, you first have to install and load the data.table package.

install.packages("data.table")
library(data.table)

I have created this dummy data for your situation, and converted the character type of 'Year' to the Date format, and converted whole data to a data.table :

datapop <- data.frame(Year = c("1980", "1982", "1984"), OECD = c(2,3,4), World = c(3,5,8), Switzerland = c(2,2.5,3)) %>% mutate(Year = as.Date(Year, format = "%Y")) %>% as.data.table()

I melted the these columns of 'OECD', 'World' and 'Switzerland' to a single column of 'variable' data.table 's melt() function, using 'Year' column as the ID column:

datapop_melted <- melt(datapop, id.vars = "Year")

Then, I simply plotted this data using ggplot2 package. In the aes() section I have given linetype = variable argument, so that it creates the legend according to the line types by itself:

ggplot(data=datapop_melted) +
      geom_line(aes(x = Year, y = value, linetype = variable)) + 
      labs(title="Years vs Aging", x ="Years", y = "Aging", linetype = "Location") +
      scale_x_date(date_breaks = "2 years", date_labels = "%Y") +
      theme(axis.text.x=element_text(angle=60, hjust=1)) +
      theme_minimal()

You can see the resulting plot in the link .

Answer 2

UPDATE

Here is a solution with mtcars .

data("mtcars")

ggplot( data = mtcars) +
  geom_line(aes(gear, mpg, linetype = "mpg"), color = "royalblue", size = 0.7 ) + 
  geom_line(aes(gear, drat, linetype = "drat" ), color = "royalblue", size = 0.7 ) + 
  geom_line(aes(gear, qsec, linetype = "qsec" ), color = "royalblue",  size = 0.7 )+ 
  xlab("gear") + ylab("Outcomes") + 
  scale_linetype_manual( name = "Legend text", 
                         values = c( "mpg" = "solid",
                                     "drat" = "dashed",
                                     "qsec" = "twodash" ) ) +
  theme_minimal() + 
  theme(axis.text.x=element_text(angle=60, hjust=1))

For you example, I would do the following. To adjust your linetype you should use scale_linetype_manual() . The argument name is for the legend's title, whereas values controls the linetype by calling the names you used in the original aes() within geom_line() .

ggplot( data = datapop)+
  geom_line(aes(Year, OECD, linetype = "OECD" ), color = "royalblue", size = 0.7 ) + 
  geom_line(aes(Year, World, linetype = "World" ), color = "royalblue", size = 0.7 ) + 
  geom_line(aes(Year, Switzerland, linetype = "Switzerland" ), color = "royalblue", size = 0.7 ) + 
  xlab("Year") + ylab("Aging") + 
  scale_color_manual( name = "Legend text",
                      values = c( "OECD" = "solid",
                                  "World" = "dashed",
                                  "Switzerland" = "twodash" ) ) +
  theme_minimal() + 
  theme(axis.text.x=element_text(angle=60, hjust=1))

This is the way you should do it without using any other package rather than ggplot2 .

Does it look better now?