In thk686/keitt.ssi.2019: Learning materials for Geospatial Data Analysis in R, part of the UT Summer Statistics Institute

library(tidyverse)
library(tigris)
library(sf)

The tigris package defines functions that will download spatial data. It uses the sp classes, so we convert to sf using st_as_sf and then transform to Texas Centric Albers Equal Area projection. We do not use the full resolution data here, which would be obtained if we did not specify the cb and resolution switches.

texas.counties <- counties(48, cb = TRUE, resolution = '5m') %>%
  st_as_sf() %>%
  st_transform(crs = 3083)

Now we can plot the result.

ggplot(texas.counties) +
  geom_sf(aes(fill = rank(as.integer(AWATER)))) +
  scale_fill_gradient(low = "tan", high = "white") +
  guides(fill = FALSE) +
  theme_bw()

Here's an example from the author of the tigris package. First we retrieve some party affiliation data (certainly dated) from a website. Then we join that data to the spatial data using dplyr's inner_join function.

party <- read_csv("http://personal.tcu.edu/kylewalker/data/txlege.csv", col_types = cols())
districts <- state_legislative_districts("TX", house = "lower", cb = TRUE) %>%
  st_as_sf() %>%
  st_transform(crs = 3083) %>%
  inner_join(y = party, by = c("NAME" = "District")) %>%
  mutate(Party = as.factor(recode(Party, R = "Republican", D = "Democrat")))

Now we have a spatial coverage where each polygon (lower house districts in Texas) has a party affiliation. Let's make a plot.

ggplot(districts) +
  geom_sf(aes(color = Party, fill = Party), alpha = 0.5) +
  scale_fill_manual(values = c("blue", "red")) +
  scale_color_manual(values = c("darkblue", "darkred")) +
  theme_bw()

thk686/keitt.ssi.2019 documentation built on June 28, 2019, 1:28 p.m.