1 Introduction

This report completes the practice exercises inspired by a three-day workshop taught by Dr. Kyle Walker on working with the 2021 American Community Survey (ACS) in R. Part A looks at 5-year ACS data for the percent of the population with a graduate degree by county in California. Part B uses 5-year ACS data of median rent by census tract in San Francisco County to create an interactive choropleth map.

2 Part A

2.1 Data Preparation

2.1.1 1. Use get_acs() to fetch data on the percentage of the population that have a graduate degree (“DP02_0066P”) at the county level.

# query ACS data on percent graduate degrees by county in California 

CA_Grad <- get_acs(
    geography = "county",
    state = "CA",
    variables = c(percent_graduate = "DP02_0066P"),
    year = 2021)

2.2 Data Analysis

2.2.1 Which counties have the largest percentages of graduate degree holders?

# use tidyverse to arrange data from highest to lowest

arrange(CA_Grad, desc(estimate))
## # A tibble: 58 × 5
##    GEOID NAME                             variable         estimate   moe
##    <chr> <chr>                            <chr>               <dbl> <dbl>
##  1 06085 Santa Clara County, California   percent_graduate     26.6   0.3
##  2 06041 Marin County, California         percent_graduate     26.3   0.9
##  3 06075 San Francisco County, California percent_graduate     24.1   0.5
##  4 06081 San Mateo County, California     percent_graduate     23.1   0.5
##  5 06001 Alameda County, California       percent_graduate     21.3   0.3
##  6 06113 Yolo County, California          percent_graduate     21     0.9
##  7 06003 Alpine County, California        percent_graduate     18.7   8.5
##  8 06087 Santa Cruz County, California    percent_graduate     17.8   0.8
##  9 06013 Contra Costa County, California  percent_graduate     16.9   0.4
## 10 06073 San Diego County, California     percent_graduate     15.7   0.3
## # ℹ 48 more rows

Results: Santa Clara, Marin, San Francisco, San Mateo, and Alameda Counties have the highest percentage of graduate degree holders in California.

2.2.2 Which counties have the smallest percentages?

# use tidyverse to to arrange data form lowest to highest

arrange(CA_Grad, estimate)
## # A tibble: 58 × 5
##    GEOID NAME                          variable         estimate   moe
##    <chr> <chr>                         <chr>               <dbl> <dbl>
##  1 06021 Glenn County, California      percent_graduate      2.7   0.8
##  2 06035 Lassen County, California     percent_graduate      3.1   0.6
##  3 06011 Colusa County, California     percent_graduate      3.5   1  
##  4 06025 Imperial County, California   percent_graduate      4.1   0.5
##  5 06047 Merced County, California     percent_graduate      4.4   0.4
##  6 06103 Tehama County, California     percent_graduate      4.5   0.9
##  7 06031 Kings County, California      percent_graduate      4.6   0.6
##  8 06039 Madera County, California     percent_graduate      4.8   0.6
##  9 06107 Tulare County, California     percent_graduate      5     0.4
## 10 06069 San Benito County, California percent_graduate      5.2   0.8
## # ℹ 48 more rows

Results: Glenn, Lassen, Colusa, Imperial, and Merced Counties have the lowest percentage of graduate degree holders in California.

2.2.3 Make a margin of error plot.

# use ggplot to make a plot with error bars using ACS graduate degree data for counties in California

CA_plot_errorbar <- ggplot(CA_Grad, aes(x = estimate, 
                                        y = reorder(NAME, estimate))) + 
  geom_errorbar(aes(xmin = estimate - moe, xmax = estimate + moe),
                width = 0.5, linewidth = 0.5) +
  geom_point(color = "lightblue", size = 2) + 
  scale_x_continuous(labels = label_percent(scale = 1)) + 
  scale_y_discrete(labels = function(x) str_remove(x, " County, California|, California")) + 
  labs(title = "Percent of Population with a Graduate Degree, 2017-2021 ACS",
       subtitle = "Counties in California",
       caption = "Data acquired with R and tidycensus. Error bars represent margin of error around estimates.",
       x = "ACS estimate",
       y = "") + 
  theme_classic(base_size = 12)

# view plot
CA_plot_errorbar
**Figure 2.** County-level estimates of percentage of population with a graduate degree, California.

Figure 2. County-level estimates of percentage of population with a graduate degree, California.

2.2.3.1 Does this method work well for your state?

This method works okay for my state, but there are a lot of counties, so the plot needs to be large and some counties have a wide margin of error.

2.2.4 2. Use the plotly package to convert the plot created in Step 1 to an interactive chart.

ggplotly(CA_plot_errorbar, tooltip = "x")

Figure 2. County-level estimates of percentage of population with a graduate degree, California. Hover over each point to see estimate value.

3 Part B - Spatial Census Data

3.1 Data Preparation

3.1.1 1. Use the load_variables() function to find a variable that interests me.

# load and view Data Profiles to select a state and demographic of interest
pers <- load_variables(2021, "acs5/profile")

head(pers)
## # A tibble: 6 × 3
##   name         label                                                     concept
##   <chr>        <chr>                                                     <chr>  
## 1 DP02PR_0001  Estimate!!HOUSEHOLDS BY TYPE!!Total households            SELECT…
## 2 DP02PR_0001P Percent!!HOUSEHOLDS BY TYPE!!Total households             SELECT…
## 3 DP02PR_0002  Estimate!!HOUSEHOLDS BY TYPE!!Total households!!Married-… SELECT…
## 4 DP02PR_0002P Percent!!HOUSEHOLDS BY TYPE!!Total households!!Married-c… SELECT…
## 5 DP02PR_0003  Estimate!!HOUSEHOLDS BY TYPE!!Total households!!Married-… SELECT…
## 6 DP02PR_0003P Percent!!HOUSEHOLDS BY TYPE!!Total households!!Married-c… SELECT…

I’ve decided to use the estimate for median rent (US Dollars) in San Francisco County, California

3.1.2 2. Use get_acs() to fetch spatial ACS data on my chosen variable.

# query 5-year ACS Census data for median rent by census tract in San Francisco County, California, include county Shapefiles using "geometry = TRUE"

SF_Rent <- get_acs(
  geography = "tract",
  variables = c(median_rent = "DP04_0134"),
  state= "CA",
  county = "San Francisco",
  geometry = TRUE)

3.1.3 3. Use mapview() to display your data interactively.

# explore data with the mapview package

mapview(SF_Rent, zcol = "estimate")

Results: It looks like estimates range from 336 to 3,500, but the lowest values are in a tract that is a large outdoor recreation area that doesn’t contain houses, so I will remove that GEOID. There is also an island far off the coast that I want to remove

# filter data with tidyr
SF_Rent_Clean <- SF_Rent %>% filter(GEOID != "06075980501") %>% 
  filter(GEOID != "06075980401")

# check with mapview again
mapview(SF_Rent_Clean, zcol = "estimate")

Results: Now, values range from ~530-3,500.

3.2 Data Analysis

3.2.1 4. Use ggplot to create a choropleth map.

# First, I want to show my estimate column in US dollars

SF_Rent_Clean$Median_Rent <- dollar(SF_Rent_Clean$estimate)

# Next, I want to make a custom tooltip 

SF_Rent_Clean <- SF_Rent_Clean %>% mutate(Median_Rent = dollar(estimate),
                                          Rent_MOE = dollar(moe),
                                          tooltip_text = paste0(Median_Rent, " +/-", Rent_MOE))

# I will use ggplot to create a choropleth map since my estimates are derived values.

ggsf <- ggplot(SF_Rent_Clean, aes(fill = estimate, data_id = GEOID,
                              tooltip = tooltip_text)) + 
  geom_sf_interactive() + #<<
  scale_fill_viridis_c(option = "mako", direction = -1) + 
  theme_classic() + 
  labs(fill = "Median rent\n2021 ACS",
       title = "San Francisco County Census Tracts")

# I will use giraph to show interactive popups.

girafe(ggobj = ggsf) %>% #<<
  girafe_options(opts_hover(css = "fill:cyan;"))

Figure 2. San Francisco County Census data from the 5-year ACS showing median rent per tract. Hover over each county to view dollar estimate and margin of error

Results: Rent in San Francisco County is generally very high(above $3,000 in most tracts), but some areas, especially downtown, appear to have lower median rent prices.

4 Citations

SSDAΝ. (2023, March 1). 2023 webinar 1 – Working with the 2021 American Community Survey with R and Tidycensus. [Video]. YouTube. https://www.youtube.com/watch?v=1WKAeM8yR-0

SSDAΝ. (2023, March 2). 2023 webinar 2 – Mapping and spatial analysis with ACS data in R [Video]. YouTube. https://www.youtube.com/watch?v=ahPquLT6gxA