ggplot2 add colors to legend

Question

I am creating a multi-line chart with legend, but I'm not sure how to force the squares in the legend to be the same color as my lines.

ggplot(df, aes(x=Month)) +
  geom_line(aes(y=`Less than a high school diploma`/100, color="Less Than HS Diploma"), size=2, alpha=0.5, linetype=1) +
  geom_line(aes(y=`High school graduates, no college`/100, color="HS Diploma"), size=2, alpha=0.5, linetype=1) + 
  geom_line(aes(y=`Some college or associate degree`/100, color="Some College / Associate's Degree"), size=2, alpha=0.5, linetype=1) +
  geom_line(aes(y=`Bachelor's degree and higher`/100, color="Bachelor's Degree and Higher"), size=2, alpha=0.5, linetype=1) +
  scale_color_manual(name="Educational Attainment", values = c("Less Than HS Diploma"="deepskyblue", "HS Diploma" = "firebrick1", "Some College / Associate's Degree"="mediumpurple", "Bachelor's Degree and Higher"="springgreen4")) +
  ggtitle("Unemployment Rate by Educational Attainment") +
  xlab("Time") +
  ylab("Unemployment Rate") +
  scale_y_continuous(labels = scales::percent) +
  theme(plot.title = element_text(hjust = 0.5), legend.position="bottom")

Answer 1

The main issue is the alpha argument together with geom_line. If you want the keys to show up as lines you set alpha to 1 in the legend via guides(color = guide_legend(override.aes = list(alpha = c(1, 1, 1, 1)))). If you want colored rectangles for the keys this could be achieved by adding key_glyph = "rect" to your geom_line layers

Using the economics dataset as example data:

library(ggplot2)

ggplot(economics, aes(x=date)) +
  geom_line(aes(y=`psavert`/100, color="Less Than HS Diploma"), size=2, alpha=0.5, linetype=1) +
  geom_line(aes(y=`uempmed`/100, color="HS Diploma"), size=2, alpha=0.5, linetype=1) + 
  geom_line(aes(y=`psavert`/10, color="Some College / Associate's Degree"), size=2, alpha=0.5, linetype=1) +
  geom_line(aes(y=`uempmed`/10, color="Bachelor's Degree and Higher"), size=2, alpha=0.5, linetype=1) +
  scale_color_manual(name="Educational Attainment", values = c("Less Than HS Diploma"="deepskyblue", "HS Diploma" = "firebrick1", "Some College / Associate's Degree"="mediumpurple", "Bachelor's Degree and Higher"="springgreen4")) +
  guides(color = guide_legend(override.aes = list(alpha = c(1, 1, 1, 1)))) +
  ggtitle("Unemployment Rate by Educational Attainment") +
  xlab("Time") +
  ylab("Unemployment Rate") +
  scale_y_continuous(labels = scales::percent) +
  theme(plot.title = element_text(hjust = 0.5), legend.position="bottom")

And with key_glyph="rect":

library(ggplot2)

ggplot(economics, aes(x=date)) +
  geom_line(aes(y=`psavert`/100, color="Less Than HS Diploma"), size=2,alpha=0.5,  linetype=1, key_glyph = "rect") +
  geom_line(aes(y=`uempmed`/100, color="HS Diploma"), size=2, alpha=0.5, linetype=1, key_glyph = "rect") + 
  geom_line(aes(y=`psavert`/10, color="Some College / Associate's Degree"), size=2, alpha=0.5, linetype=1, key_glyph = "rect") +
  geom_line(aes(y=`uempmed`/10, color="Bachelor's Degree and Higher"), size=2, alpha=0.5, linetype=1, key_glyph = "rect") +
  scale_color_manual(name="Educational Attainment", values = c("Less Than HS Diploma"="deepskyblue", "HS Diploma" = "firebrick1", "Some College / Associate's Degree"="mediumpurple", "Bachelor's Degree and Higher"="springgreen4")) +
  ggtitle("Unemployment Rate by Educational Attainment") +
  xlab("Time") +
  ylab("Unemployment Rate") +
  scale_y_continuous(labels = scales::percent) +
  theme(plot.title = element_text(hjust = 0.5), legend.position="bottom")

^{Created on 2020-10-22 by the reprex package (v0.3.0)}

Answer 2

The values argument from scale_color_manual should have color names instead of the line names, which you don't need to pass. Example:

scale_color_manual(name="Educational Attainment", values = c("red","yellow","white",...))

Answer 3

This should work (No output included as no data was shared). If you want the legend filled, you must also enable inside aes() the option fill. After that you can scale the colors for filling with scale_fill_manual() and use labs() to give them a common name. Here the code:

library(ggplot2)
#Code
ggplot(df, aes(x=Month)) +
  geom_line(aes(y=`Less than a high school diploma`/100,
                color="Less Than HS Diploma",
                fill="Less Than HS Diploma"), size=2, alpha=0.5, linetype=1) +
  geom_line(aes(y=`High school graduates, no college`/100,
                color="HS Diploma",
                fill="HS Diploma"), size=2, alpha=0.5, linetype=1) + 
  geom_line(aes(y=`Some college or associate degree`/100,
                color="Some College / Associate's Degree",
                fill="Some College / Associate's Degree"), size=2, alpha=0.5, linetype=1) +
  geom_line(aes(y=`Bachelor's degree and higher`/100,
                color="Bachelor's Degree and Higher",
                fill="Bachelor's Degree and Higher"), size=2, alpha=0.5, linetype=1) +
  scale_color_manual(name="Educational Attainment",
                     values = c("Less Than HS Diploma"="deepskyblue",
                                "HS Diploma" = "firebrick1",
                                "Some College / Associate's Degree"="mediumpurple",
                                "Bachelor's Degree and Higher"="springgreen4")) +
  scale_fill_manual(name="Educational Attainment",
                     values = c("Less Than HS Diploma"="deepskyblue",
                                "HS Diploma" = "firebrick1",
                                "Some College / Associate's Degree"="mediumpurple",
                                "Bachelor's Degree and Higher"="springgreen4"))+
  ggtitle("Unemployment Rate by Educational Attainment") +
  xlab("Time") +
  ylab("Unemployment Rate") +
  scale_y_continuous(labels = scales::percent) +
  theme(plot.title = element_text(hjust = 0.5), legend.position="bottom")+
  labs(color='Class',fill='Class')