Follow me on twitter @bradleyboehmke
Follow me on twitter @bradleyboehmke
A hybrid between a bar chart and a Cleveland dot plot is the lollipop chart. A lollipop chart typically contains categorical variables on the y-axis measured against a second (continuous) variable on the x-axis. Similar to the Cleveland dot plot, the emphasis is on the dot to draw the readers attention to the specific x-axis value achieved by each category. The line is meant to be a minimalistic approach to easily tie each category to its relative point without drawing too much attention to the line itself. A lollipop chart is great for comparing multiple categories as it aids the reader in aligning categories to points but minimizes the amount of ink on the graphic.
This tutorial introduces the basics of the lollipop chart and compares them to bar charts and dot plots. I also show how to go from a basic lollipop chart to a more refined, publication worthy graphic. If you’re short on time jump to the sections of interest:
To reproduce the code throughout this tutorial you will need to load the following packages. The primary package of interest is ggplot2, which is a plotting system for R. You can build lollipop charts with base R graphics, but when I’m building more refined graphics I lean towards ggplot2. Note that I use the development version of ggplot2 which offers some nice title, subtitle, and caption options which I cover in the last section. You can download the development version with this line of code: devtools::install_github("hadley/ggplot2")
library(dplyr) # for data manipulation
library(tidyr) # for data tidying
library(ggplot2) # for generating the visualizations
In addition, throughout the tutorial I illustrate the graphics with the midwest data set provided in the ggplot2 package.
head(midwest)
## # A tibble: 6 x 28
## PID county state area poptotal popdensity popwhite popblack
## <int> <chr> <chr> <dbl> <int> <dbl> <int> <int>
## 1 561 ADAMS IL 0.052 66090 1270.9615 63917 1702
## 2 562 ALEXANDER IL 0.014 10626 759.0000 7054 3496
## 3 563 BOND IL 0.022 14991 681.4091 14477 429
## 4 564 BOONE IL 0.017 30806 1812.1176 29344 127
## 5 565 BROWN IL 0.018 5836 324.2222 5264 547
## 6 566 BUREAU IL 0.050 35688 713.7600 35157 50
## # ... with 20 more variables: popamerindian <int>, popasian <int>,
## # popother <int>, percwhite <dbl>, percblack <dbl>, percamerindan <dbl>,
## # percasian <dbl>, percother <dbl>, popadults <int>, perchsd <dbl>,
## # percollege <dbl>, percprof <dbl>, poppovertyknown <int>,
## # percpovertyknown <dbl>, percbelowpoverty <dbl>,
## # percchildbelowpovert <dbl>, percadultpoverty <dbl>,
## # percelderlypoverty <dbl>, inmetro <int>, category <chr>
Most readers would have little problem understanding either of the basic versions of the lollipop chart, dot plot or the bar chart. Consider if we want to view the top 25 counties in Ohio for percentage of college educated folks. After a little data manipulation (note that I order the counties by percent college educated (percollege) and then make the county variable a factor with the levels ordered accordingly; this will allow us to order the bars and dots in the following charts appropriately)…
ohio_top25 <- midwest %>%
filter(state == "OH") %>%
select(county, percollege) %>%
arrange(desc(percollege)) %>%
top_n(25) %>%
arrange(percollege) %>%
mutate(county = factor(county, levels = .$county))
We could view the data as a horizontal bar chart…
# bar chart
ggplot(ohio_top25, aes(county, percollege)) +
geom_bar(stat = "identity") +
coord_flip()
as a dot plot…
# dot plot
ggplot(ohio_top25, aes(percollege, county)) +
geom_point()
or as a lollipop chart. In the lollipop chart we use geom_segment to plot the lines and we explicitly state that we want the lines to start at x = 0 and extend to the percollege value with xend = percollege. We simply need to include y = county and yend = county to tell R the lines are horizontally attached to each county.
# lollipop chart
ggplot(ohio_top25, aes(percollege, county)) +
geom_segment(aes(x = 0, y = county, xend = percollege, yend = county), color = "grey50") +
geom_point()
Consider the case where we want to compare counties in Ohio to see how they differ from the state average. For this the emphasis becomes the state average so we can do a data manipulation to generate the state average value and test if each county above or below that value.
ohio <- midwest %>%
filter(state == "OH") %>%
select(county, percollege) %>%
arrange(percollege) %>%
mutate(Avg = mean(percollege, na.rm = TRUE),
Above = ifelse(percollege - Avg > 0, TRUE, FALSE),
county = factor(county, levels = .$county))
head(ohio)
## # A tibble: 6 x 4
## county percollege Avg Above
## <fctr> <dbl> <dbl> <lgl>
## 1 VINTON 7.913256 16.89045 FALSE
## 2 ADAMS 8.741730 16.89045 FALSE
## 3 NOBLE 8.845888 16.89045 FALSE
## 4 HOLMES 9.330709 16.89045 FALSE
## 5 PERRY 10.107671 16.89045 FALSE
## 6 MONROE 10.513927 16.89045 FALSE
We can now incorporate this data into graphic by mapping the x = argument within geom_segment to the state average and then color the counties based on if they are above or below average.
ggplot(ohio, aes(percollege, county, color = Above)) +
geom_segment(aes(x = Avg, y = county, xend = percollege, yend = county), color = "grey50") +
geom_point()
Another comparison approach is if we wanted to compare the top 10 counties for each of the midwest states in our data set. In this case we have to do additional manipulation as each state has a couple county names that are common.
top10 <- midwest %>%
select(state, county, percollege) %>%
group_by(state) %>%
arrange(desc(percollege)) %>%
top_n(10) %>%
arrange(percollege) %>%
unite(county_st, county, state, remove = FALSE) %>%
mutate(county_st = factor(county_st, levels = .$county_st))
head(top10)
## Source: local data frame [6 x 4]
## Groups: state [2]
##
## county_st state county percollege
## <fctr> <chr> <chr> <dbl>
## 1 WARRICK_IN IN WARRICK 23.81557
## 2 HENDRICKS_IN IN HENDRICKS 24.22003
## 3 PORTER_IN IN PORTER 24.45589
## 4 ST JOSEPH_IN IN ST JOSEPH 24.61167
## 5 SUMMIT_OH OH SUMMIT 24.73314
## 6 CUYAHOGA_OH OH CUYAHOGA 25.08539
We can now plot our data and facet by state to get small multiples representing the top 10 counties for each state. Here I abbreviate the names for brevity.
ggplot(top10, aes(percollege, county_st)) +
geom_segment(aes(x = 0, y = county_st, xend = percollege, yend = county_st), color = "grey50") +
geom_point() +
scale_y_discrete(labels = abbreviate) +
facet_wrap(~ state, scales = "free_y")
Depending on the number of categories (i.e. counties) you are trying to graphically display, and the range of the x-axis, it can be helpful to add value markers to the points to clarify the difference between the points.
OH_top10 <- midwest %>%
select(state, county, percollege) %>%
filter(state == "OH") %>%
arrange(desc(percollege)) %>%
top_n(10) %>%
arrange(percollege) %>%
mutate(county = factor(county, levels = .$county))
ggplot(OH_top10, aes(percollege, county, label = round(percollege, 1))) +
geom_segment(aes(x = 0, y = county, xend = percollege, yend = county), color = "grey50") +
geom_point() +
geom_text(nudge_x = 1.5)
Alternatively you can enlarge the dots to include the labelling inside of them.
ggplot(OH_top10, aes(percollege, county, label = paste0(round(percollege, 0), "%"))) +
geom_segment(aes(x = 0, y = county, xend = percollege, yend = county), color = "grey50") +
geom_point(size = 7) +
geom_text(color = "white", size = 2)
Now let’s take one of these outputs and create a nice, publication worthy graphic. Here I take the graphic above that compares the percent of college educated adults in Ohio counties to the state average and do a little refinement. I use geom_segment and annotate to create a more simplified and appealing legend, add a title, subtitle and caption followed by some basic theme adjustments.
ggplot(ohio, aes(percollege/100, county, color = Above)) +
geom_segment(aes(x = Avg/100, y = county, xend = percollege/100, yend = county), color = "grey50") +
geom_point() +
annotate("text", x = .25, y = "ALLEN", label = "Above Average", color = "#00BFC4", size = 3, hjust = -0.1, vjust = .75) +
annotate("text", x = .25, y = "FULTON", label = "Below Average", color = "#F8766D", size = 3, hjust = -0.1, vjust = -.1) +
geom_segment(aes(x = .25, xend = .25 , y = "ASHLAND", yend = "DEFIANCE"),
arrow = arrow(length = unit(0.2,"cm")), color = "#00BFC4") +
geom_segment(aes(x = .25, xend = .25 , y = "KNOX", yend = "PUTNAM"),
arrow = arrow(length = unit(0.2,"cm")), color = "#F8766D") +
scale_x_continuous(labels = scales::percent, expand = c(0, 0), limits = c(.07, .33)) +
labs(title = "College Educated Adults in Ohio Counties",
subtitle = "The average percent of college educated adults in Ohio is 16.89%. Franklin, Greene, Geauga, and \nDelaware counties lead Ohio with over 30% of their adults being college educated while Vinton, \nAdams, Holmes, and Perry trailing with less than 10% of their adults being college educated.",
caption = "U.S. Census Bureau: 2000 Census") +
theme_minimal() +
theme(axis.title = element_blank(),
panel.grid.minor = element_blank(),
legend.position = "none",
text = element_text(family = "Georgia"),
axis.text.y = element_text(size = 8),
plot.title = element_text(size = 20, margin = margin(b = 10), hjust = 0),
plot.subtitle = element_text(size = 12, color = "darkslategrey", margin = margin(b = 25, l = -25)),
plot.caption = element_text(size = 8, margin = margin(t = 10), color = "grey70", hjust = 0))
Lollipop charts are a nice alternative to bar charts and dot plots. As previously mentioned, when trying to visualize data across many categories they, much like dot plots, provide a nice minimalistic visualization of the data.