In almartin82/njschooldata: a simple interface for accessing NJ DOE school data in R

Update, March 10 2018

Pre-pulled data has been removed from /data to make njschooldata snappier - the size of the data files were creating a major performance for package development because R reloads them every time devtools::load_all() is called.

Archive

just some documentation of what I've saved to ~~sysdata.Rda~~ various .Rda files

#libraries and functions etc
require(knitr)
#switch this to run the vignette
knitr::opts_knit$set(eval = FALSE)

ASSESSMENT

(run everything in new_all_assess.Rmd)

ENROLLMENT

enr_files <- list()

for (i in c(2017:1999)) {
  enr <- fetch_enr(i) 

  enr_files[[paste0('enr', i)]] <- enr
}

rbind enr together

enr_all <- enr_files[["enr2015"]]

for (i in names(enr_files)[2:length(enr_files)]) {
  print(i)
  enr_all <- dplyr::rbind_all(list(enr_all, enr_files[[i]]))

  print(nrow(enr_all))
} 

HS GRAD

hs_list <- list()
for (i in c(1998:2014)) {
  print(i)
  hs_list[[i]] <- fetch_grate(i)
}

hs_grad <- dplyr::rbind_all(hs_list)

save(enr_files, file = 'data\\enr_files.Rda')
save(enr_all, file = 'data\\enr_all.Rda')
save(hs_grad, file = 'data\\hs_grad.Rda')
save(all_assess_raw, file = 'data\\all_assess_raw.Rda')
save(all_assess_tidy, file = 'data\\all_assess_tidy.Rda')

prog codes need padded zeros

library(stringr)
library(rprojroot)

root <- rprojroot::find_rstudio_root_file()

prog_codes <- readr::read_csv(file.path(root, 'data-raw', 'program_codes_clean.csv'))

prog1 <- prog_codes[prog_codes$end_year >= 2010, ]
prog2 <- prog_codes[prog_codes$end_year < 2010, ]
prog2$program_code <- as.numeric(prog2$program_code) * 1

prog1$program_code <- stringr::str_pad(
  prog1$program_code, width = 2, side = 'left', pad = '0'
)

prog_codes <- rbind(prog1, prog2)

save(prog_codes, file = file.path(root, 'data', 'program_codes.Rda'))

for debugging; step through fetch_enr steps

foo1 <-  get_raw_enr(2009) 
foo2 <- foo1 %>% clean_enr_names()
foo3 <- foo2 %>% split_enr_cols()
foo4 <- foo3 %>% clean_enr_data()
foo5 <- foo4 %>% process_enr_program()
foo6 <- foo5 %>% arrange_enr()

almartin82/njschooldata documentation built on Nov. 23, 2023, 1:33 a.m.