US Census Bureau API

The United States Census Bureau has come up with a way to provide developers and analysts with census data by means of APIs.

The censusapi R package is a wrapper for those APIs, and allows users to connect to 370+ endpoints, which include for instance the well-known American Community Surveys and Population Estimates.

In this short document, we will quickly explore some statistics about women in business in the state of New York.

Note that you will need to request an API key in order to be able to use the service, after which you will need to perform the following steps:

# Add key to .Renviron

Sys.setenv(CENSUS_KEY=YOURKEYHERE)

# Reload .Renviron
readRenviron("~/.Renviron")

# Check to see that the expected key is output in your R console
Sys.getenv("CENSUS_KEY")

Show available endpoints with the listCensusApis() function

library(censusapi)  # Loading required packages
library(tidyverse)

apis <- listCensusApis()  # Contains list of endpoints

Available Endpoints

In this instance we wish to connect to the "ACS 1-year Subject Tables" endpoint (acs/acs1/subject) for year 2017.

Finding the number of women-owned incorporated businesses in New York State.

Using listCensusMetadata() allows one to retrieve our variable of interest, i.e. S2408_C04_004E (number of women-owned incorporated businesses in NY), from the "ACS 1-year Subject Tables" API endpoint for the year 2017 . The data will then be partitionned into two distinct geographic levels, i.e. county and congressional district.

vars <- listCensusMetadata(name = "acs/acs1/subject",
                           type = 'variables', 
                           vintage = '2017')

# Loading packages for producing tables

library(knitr)
library(kableExtra)
library(DT)

knitr::kable(head(vars), caption = 'Subset of Census Metadata for the 2017 ACS') %>%
  kable_styling(full_width = T) %>% 
  kableExtra::scroll_box(width = "100%", height = "500px")

geo <- listCensusMetadata(name = "acs/acs1/subject",
                           type = 'geography', 
                           vintage = '2017')

DT::datatable(geo, options = list(pageLength = 5)) 

Collecting and Visualizing the data.

Using the get_acs() from the tidycensus package to explore women self-employment data in the State of New-York.

The tidycensus package is a great supplement to censusapi in a sense that it provides spatial information, which can be utilized for visualization purposes.

# Loading required packages

library(tidycensus)
library(tigris)

self_employed_women_county <- get_acs(geography = "county", variables = "S2408_C04_004E", geometry = T, survey = 'acs1', state = "NY")

# Loading packages for plots

library(ggthemes)
library(plotly)
library(sf)
library(viridis)

p_county <- ggplot(self_employed_women_county, aes(x = reorder(NAME, estimate), y = estimate, fill = NAME, text = paste('</br>County: ', NAME, '</br>Estimate: ', estimate))) + 
  geom_bar(stat = 'identity') +
  theme_gdocs() +
  theme(legend.position = 'none', axis.text.x = element_text(angle = 45, hjust = 1)) +
  ylab('Estimate') + xlab("County") +
  ggtitle("Self-Employement estimates for Women-owned incorporated businesses in New York",
          subtitle = 'County-level')

ggplotly(p_county, tooltip = 'text')

county_map <- ggplot(self_employed_women_county, aes(fill = estimate, color = estimate, text = paste('</br>County: ', NAME, '<br>Estimate: ', estimate))) +
  geom_sf() + 
  coord_sf(crs = 26910) +
  theme_void() + 
  theme(aspect.ratio = 1) + 
  scale_fill_viridis() + 
  scale_color_viridis() +
  labs(title = "Number of women-owned incorporated businesses per New York State county.", caption = '*Note that data is missing for some counties.')
  

ggplotly(county_map, tooltip = c('text'))

self_employed_women_congress_district <- get_acs(geography = "congressional district", variables = "S2408_C04_004E", geometry = F, survey = 'acs1', state = "NY")

geom <- congressional_districts(cb = T, year = '2017')

women_congress <- left_join(self_employed_women_congress_district, geom, by = 'GEOID')

plot_cd <- ggplot(women_congress, aes(x = reorder(NAME, estimate), y = estimate, fill = NAME, text = paste('</br>Congressional District: ', NAME, '</br>Estimate: ', estimate))) + 
  geom_bar(stat = 'identity') +
  theme_gdocs() +
  theme(legend.position = 'none', axis.text.x = element_text(angle = 45, hjust = 1)) +
  ylab('Estimate') + xlab("Congressional District") +
  ggtitle(label = "Self-Employement estimates for Women-owned incorporated businesses in New York",
          subtitle = 'Congressional Districts')

ggplotly(plot_cd, tooltip = 'text')

congress_map <- ggplot(women_congress, aes(fill = estimate, color = estimate, text = paste('</br>Congressional District: ', 
                                                                                           NAME, '<br>Estimate: ',estimate))) +
  geom_sf(aes(geometry = geometry)) + 
  coord_sf(crs = 26910) +
  theme_void() + 
  theme(aspect.ratio = 1) + 
  scale_fill_viridis() + 
  scale_color_viridis()
  

ggplotly(congress_map, tooltip = c('text'))

Conclusion

This short document is a non-exhaustive example of the type of information that can be retrieved by means of using the US Census Bureau API. Many different subjects are available, the majority of which possesses data distributed across various geography levels.