Estimating Chinese speaking population in NYC using R/tidycensus

NYC_ChinseseSpeakers.R

library(tidyverse)
library(tidycensus)

nyc_counties <- c("005", "047", "061", "081", "085")

acsvars <- load_variables(2015, "acs5", cache=TRUE)

options(tigris_use_cache = TRUE)
nyc_tract <- get_acs(geography="tract",
                     variables=c("B16001_001", "B16001_066", "B16001_067", "B16001_068"),
                     year=2015,
                     survey="acs5",
                     state="36",
                     county = nyc_counties,
                     output="wide", geometry = TRUE)

nyc_tract_chinese <- nyc_tract %>%
  # filter(B16001_001E>0) %>%
  select(GEOID, lang_total=B16001_001E, lang_chinese=B16001_066E, geometry) %>%
  mutate(PctChinseSpeaking=lang_chinese/lang_total*100) %>%
  arrange(GEOID)

summary(nyc_tract_chinese$PctChinseSpeaking)
nyc_tract_chinese

library(sf)

nyc_tract_chinese %>%
  ggplot(aes(fill=PctChinseSpeaking)) +
  geom_sf()