Maps

Author

Affiliation

Prof Amanda Luby

Carleton College
Stat 220 - Fall 2026

library(tidyverse)
library(maps)
library(mapproj)
library(ggthemes)

Create Original Map

geom_map() matches each row’s map_id to a region in a separate map data set. expand_limits() is required, or ggplot won’t know the x/y range to show. coord_map() gives a nicer projection than the default, and theme_map() removes the axes.

states <- map_data("state")
ggplot(states) +
  geom_map(aes(map_id = region), map = states, color="gold2", fill="navyblue") +
  expand_limits(x = states$long, y = states$lat) +
  coord_map() +
  theme_map()

Your turn

Edit the code so that each state is a different color

Tip

states already has a region column — you can map it to fill the same way it’s mapped to map_id

states <- map_data("state")
ggplot(states) +
  geom_map(aes(map_id = region), map = states, color="gold2", fill="navyblue") +
  expand_limits(x = states$long, y = states$lat) +
  coord_map() +
  theme_map()

Your turn: ACS data

Load Data

acs_state_data = read_csv("https://raw.githubusercontent.com/stat220-f26/stat220-f26.github.io/refs/heads/main/data/acs_state_data_2022_5y.csv") 
acs_state_data$state = tolower(acs_state_data$NAME)

Variable information:

med_age: median age
med_income: median age
race_X: estimated population of (self-reported) race
born_in_state: estimated population who were born in state
X_age_married: average age at first marriage for self-reported male and female respondents
hs_diploma, associate_degree, prof_degree, etc.: estimated number with a high school diploma, associate’s degree, professional degree, etc.
internet_X: estimated number of people who have internet services

ACS Data Starter Map

Important note: this is the same geom_map() you just used, now with real data. acs_state_data only needs a column with the state name (state). geom_map() looks up the matching shape in map = states for you.

ggplot(acs_state_data) + 
  geom_map(
    aes(map_id = state, fill = NAME),
    map = states
  ) +
  expand_limits(x = states$long, y = states$lat) +
  coord_map() +
  theme_map()

Your turn 1

Comment out the expand_limits line using #. What happened?

Your turn 2

Edit your chloropleth map by:

Choosing a different variable in acs_state_data to map to fill
Choosing a different color scale
Updating the title, axis labels, and legend title

Tip

Try it: Proportional symbol maps

So far we’ve used geom_map() to fill in regions. But sometimes you want to show where specific things are located, with size showing how big they are. To do this, we add a second layer with geom_point(), using latitude/longitude coordinates.

Load Data

The maps package (already loaded) includes us.cities, a data set of US cities and their populations (as of 2006). We’ll filter to cities with a population over 100,000, and exclude Alaska and Hawaii so the map isn’t stretched out.

data(us.cities)

big_cities <- us.cities |>
  filter(!country.etc %in% c("AK", "HI"), pop > 100000)

name: city and state
pop: population
lat, long: the city’s location

Starter map: base + points

Tip

Notice that geom_map() and geom_point() each have their own data. The base map uses states; the points use big_cities. This is the same pattern you’ll need on your homework.

ggplot(states) +
  geom_map(aes(map_id = region), 
               map = states, 
               color = "white", 
               fill = "gray90") +
  geom_point(data = big_cities, 
              aes(x = ___, y = ___, size = ___), 
              color = "navyblue", 
              alpha = 0.6) +
  scale_size(range = c(1, 10), labels = scales::comma) +
  expand_limits(x = states$long, y = states$lat) +
  coord_map() +
  theme_map()

Error in parse(text = input): <text>:7:24: unexpected input
6:   geom_point(data = big_cities, 
7:               aes(x = __
                          ^

Your turn

Fill in the ___ to overlay the points on the map
Adjust the point size range, alpha, and color until you’re happy with how the map looks

# your turn

--- title: Maps --- ```{r} library(tidyverse) library(maps) library(mapproj) library(ggthemes) ``` # Create Original Map `geom_map()` matches each row's `map_id` to a region in a separate `map` data set. `expand_limits()` is required, or ggplot won't know the x/y range to show. `coord_map()` gives a nicer projection than the default, and `theme_map()` removes the axes. ```{r} states <- map_data("state") ggplot(states) + geom_map(aes(map_id = region), map = states, color="gold2", fill="navyblue") + expand_limits(x = states$long, y = states$lat) + coord_map() + theme_map() ``` ## Your turn Edit the code so that each state is a different color ::: callout-tip `states` already has a `region` column — you can map it to `fill` the same way it's mapped to `map_id` ::: ```{r} states <- map_data("state") ggplot(states) + geom_map(aes(map_id = region), map = states, color="gold2", fill="navyblue") + expand_limits(x = states$long, y = states$lat) + coord_map() + theme_map() ``` # Your turn: ACS data ## Load Data ```{r} acs_state_data = read_csv("https://raw.githubusercontent.com/stat220-f26/stat220-f26.github.io/refs/heads/main/data/acs_state_data_2022_5y.csv") acs_state_data$state = tolower(acs_state_data$NAME) ``` Variable information: - `med_age`: median age - `med_income`: median age - `race_X`: estimated population of (self-reported) race - `born_in_state`: estimated population who were born in state - `X_age_married`: average age at first marriage for self-reported male and female respondents - `hs_diploma`, `associate_degree`, `prof_degree`, etc.: estimated number with a high school diploma, associate's degree, professional degree, etc. - `internet_X`: estimated number of people who have internet services ## ACS Data Starter Map **Important note:** this is the same `geom_map()` you just used, now with real data. `acs_state_data` only needs a column with the state name (`state`). `geom_map()` looks up the matching shape in `map = states` for you. ```{r} ggplot(acs_state_data) + geom_map( aes(map_id = state, fill = NAME), map = states ) + expand_limits(x = states$long, y = states$lat) + coord_map() + theme_map() ``` ### Your turn 1 Comment out the `expand_limits` line using `#`. What happened? ### Your turn 2 Edit your chloropleth map by: - Choosing a different variable in `acs_state_data` to map to `fill` - Choosing a different color scale - Updating the title, axis labels, and legend title ::: callout-tip ::: # Try it: Proportional symbol maps So far we've used `geom_map()` to fill in regions. But sometimes you want to show *where specific things are located*, with size showing how big they are. To do this, we add a second layer with `geom_point()`, using latitude/longitude coordinates. ## Load Data The `{maps}` package (already loaded) includes `us.cities`, a data set of US cities and their populations (as of 2006). We'll filter to cities with a population over 100,000, and exclude Alaska and Hawaii so the map isn't stretched out. ```{r} data(us.cities) big_cities <- us.cities |> filter(!country.etc %in% c("AK", "HI"), pop > 100000) ``` - `name`: city and state - `pop`: population - `lat`, `long`: the city's location ## Starter map: base + points ::: callout-tip Notice that `geom_map()` and `geom_point()` each have their *own* `data`. The base map uses `states`; the points use `big_cities`. This is the same pattern you'll need on your homework. ::: ```{r} #| error: true ggplot(states) + geom_map(aes(map_id = region), map = states, color = "white", fill = "gray90") + geom_point(data = big_cities, aes(x = ___, y = ___, size = ___), color = "navyblue", alpha = 0.6) + scale_size(range = c(1, 10), labels = scales::comma) + expand_limits(x = states$long, y = states$lat) + coord_map() + theme_map() ``` ### Your turn - Fill in the `___` to overlay the points on the map - Adjust the point `size` range, `alpha`, and color until you're happy with how the map looks ```{r} # your turn ```