MPA 5830 - Solutions to Practice Exercises

Spring 2020

Exercise (1)

Create a map of all the counties in New York. Be sure to title the map and to fill in each county with the total number of COVID19 cases they have seen to date. In addition, draw county borders in white. Use theme_map() and make sure the legend is at the bottom. [Hint: You will need to calculate the total number of cases per county and then join the resulting file with the counties data file to get the latitude/longitudes for the counties.]

library(tidyverse)
library(urbnmapr)
counties %>%
  filter(state_abbv == "NY") -> ny.counties

readr::read_csv("https://raw.githubusercontent.com/nytimes/covid-19-data/master/us-counties.csv") -> covid 

covid %>%
  filter(state == "New York") %>%
  group_by(county, fips) %>%
  summarise(total_cases = sum(cases)) -> cov19

ny.counties %>%
  left_join(cov19, by = c("county_fips" = "fips")) -> ny.df

ggplot() +
  geom_polygon(
    data = ny.df, 
    aes(x = long, y = lat, group = group, fill = total_cases),
    color = "salmon"
  ) +
  coord_fixed(1.3) +
  ggthemes::theme_map() +
  labs(
    title = "Total COVID-19 Cases in New York's Counties",
    fill = "No. of Cases"
       )

Exercise (2)

Run the following code chunk to load data on the murder, assault and rape rates per 100,000 persons. Urbanpop is the percent of the state population that lives in an urban area.

data(USArrests)
names(USArrests)

## [1] "Murder"   "Assault"  "UrbanPop" "Rape"

USArrests$statename <-  rownames(USArrests)

Now create a state-level map of the 50 states making sure to use UrbanPop to fill each state. Title the map and place the legend at the bottom.

states %>%
  left_join(USArrests, by = c("state_name" = "statename")) -> st.df

ggplot() +
  geom_polygon(
    data = st.df, 
    aes(x = long, y = lat, group = group, fill = UrbanPop),
    color = "salmon"
  ) +
  coord_fixed(1.3) +
  ggthemes::theme_map() +
  labs(
    title = "Percent of the Population in Urban Areas",
    subtitle = "(by State)",
    fill = "% Urban"
       )

Exercise (3)

Use the USArrests data to draw scatterplots of (a) Murder versus UrbanPop, (b) Assault versus UrbanPop, and (c) Rape versus UrbanPop. Save each of these scatterplots by name and then use patchwork to create a single canvas that includes all three plots. Make sure you label the x-axis, y-axis, and title each plot.

ggplot() +
  geom_point(
    data = USArrests,
    aes(x = Murder, y = UrbanPop)
  ) +
  labs(title = "Scatterplot of Murder vs. UrbanPop") -> p1

ggplot() +
  geom_point(
    data = USArrests,
    aes(x = Assault, y = UrbanPop)
  ) +
  labs(title = "Scatterplot of Assault vs. UrbanPop") -> p2

ggplot() +
  geom_point(
    data = USArrests,
    aes(x = Rape, y = UrbanPop)
  ) +
  labs(title = "Scatterplot of Rape vs. UrbanPop") -> p3

library(patchwork)
p1 | (p2 / p3)

Exercise (4)

Now create highcharter versions of each of the three scatterplots you created in Exercise (3) above. You should end up with three scatterplots, each on its own canvas.

library(highcharter)

lst <- list()

hchart(USArrests, "scatter", 
       hcaes(x = Murder, y = UrbanPop, group = statename)) -> lst

htmltools::tagList(lst)

hchart(USArrests, "scatter", 
       hcaes(x = Assault, y = UrbanPop, group = statename)) -> lst

htmltools::tagList(lst)

hchart(USArrests, "scatter", 
       hcaes(x = Rape, y = Assault, group = statename)) -> lst

htmltools::tagList(lst)

Exercise (5)

The sftrees dataset includes various pieces of data on trees in San Francisco. In turn, sf.df is a 10% random sample of these trees.

Use leaflet to create a map that shows, as circles, the location of each tree, and make sure when someone clicks on the circle the popup shows the species. Also ensure that the circles are green in color.

readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-01-28/sf_trees.csv') -> sf_trees

sf_trees %>%
  sample_frac(0.10) -> sf.df

library(leaflet)
library(htmltools)
library(widgetframe)

leaflet() %>%
  addTiles() %>%
  addCircles(
    data = sf.df, lng = ~ longitude, lat = ~ latitude,
    color = "green", opacity = 0.02, popup = ~htmlEscape(species)
    ) %>%
  setView(lat = 37.742832, lng = -122.441463, zoom = 12) -> lst 

frameWidget(lst, width = "100%", height = "600")