dhomer: DHOME Data

## code to prepare `public_schools` dataset goes here

library(sf)
library(tigris)
library(here)
library(readr)
library(janitor)
library(data.table)

# https://nces.ed.gov/ccd/CCDLocaleCode.asp
# South Carolina Public Schools
# Downloaded 2021-07-13

explorer_fname <- here(path('data-raw'), "raw_sc_public_schools.csv")
raw_schools <- readr::read_csv(explorer_fname) %>%
  clean_names()

# Correcting latitude/longitude error revealed during check
raw_schools$latcod[964] = 33.98846900935035
raw_schools$loncod[964] = -80.98127978465922

sc_tracts <- tracts(state = 45)

coords <- raw_schools %>%
  filter(is.na(loncod) == F & is.na(latcod) == F) %>%
  st_as_sf(coords = c('loncod', 'latcod'), crs = st_crs(sc_tracts))

system.time({
  intersected <- st_within(coords, sc_tracts)
})

schools <- coords %>%
  mutate(intersection = as.integer(intersected),
         geoid = if_else(is.na(intersection), "",
                         sc_tracts$GEOID[intersection]))

public_schools_unchecked <- st_join(schools, sc_tracts, by = c('geoid','GEOID')) %>%
  dplyr::mutate(lat = sf::st_coordinates(.)[,2],
                lon = sf::st_coordinates(.)[,1]) %>%
  select(-c('intersection','STATEFP','COUNTYFP','TRACTCE','GEOID','NAME','NAMELSAD','MTFCC','FUNCSTAT','ALAND','AWATER','lcity05','lstate05','mlocale','ulocale','status05')) %>%
  rename(tract_latitude = INTPTLAT, tract_longitude = INTPTLON) %>%
  st_drop_geometry()

public_schools = data.frame()

for (i in 1:nrow(public_schools_unchecked) ) {
  if(32.0346 <= public_schools_unchecked$lat[i] && public_schools_unchecked$lat[i] <= 35.215402 && -83.35391 <= public_schools_unchecked$lon[i] && public_schools_unchecked$lon[i] <= -78.54203) {
    public_schools <- rbind(public_schools, public_schools_unchecked[i,])
  }
  else {
    glimpse(public_schools_unchecked[i,])
  }
}

fwrite(public_schools, file = here(path('data-raw'), 'sc_public_schools.csv'))

usethis::use_data(public_schools, overwrite = TRUE)