weekly screen in rate report

weekly_data_report.R

@@ -0,0 +1,128 @@
+#!/usr/local/bin/Rscript
+
+setwd('/tmp')
+
+suppressPackageStartupMessages(library(argparser))
+p <- arg_parser('create JFS Neighborhood Report')
+p <- add_argument(p,'file_name',help='name of geocoded csv file')
+args <- parse_args(p)
+
+# args <- list()
+# args$file_name <- "simulated_jfs_data_geocoded.csv"
+
+# INTAKE_ID
+# SCREENING_DECISION = SCREENED OUT or SCREENED IN
+# ALLEGATION_ADDRESS, formatted in a string without punctuation, it includes city, state, and zip code
+
+message("\nNeighborhood results with < 5 instances have been censored for privacy purposes\n")
+
+suppressPackageStartupMessages(library(readr))
+suppressPackageStartupMessages(library(tidyverse))
+suppressPackageStartupMessages(library(sf))
+suppressPackageStartupMessages(library(knitr))
+
+d <- read_csv(args$file_name,
+              col_types = cols(INTAKE_ID = col_character(),
+                               SCREENING_DECISION = col_character(),
+                               DECISION_DATE = col_character(),
+                               PERSON_ID = col_character(),
+                               RACE = col_character(),
+                               ADDRESS_START = col_date(),
+                               MANDATED_REPORTER = col_character(),
+                               #  REPORTER_PERSON_ID = col_character(),
+                               address_type = col_character(),
+                               address = col_character(),
+                               bad_address = col_logical(),
+                               PO = col_logical(),
+                               lat = col_double(),
+                               lon = col_double(),
+                               score = col_double(),
+                               precision = col_character(),
+                               precise_geocode = col_logical(),
+                               fips_tract_id = col_character(),
+                               fraction_assisted_income = col_double(),
+                               fraction_high_school_edu = col_double(),
+                               median_income = col_double(),
+                               fraction_no_health_ins = col_double(),
+                               fraction_poverty = col_double(),
+                               fraction_vacant_housing = col_double(),
+                               dep_index = col_double()
+              ))
+
+###
+d_test <- read.csv('test/simulated_jfs_data_geocoded.csv')
+d <- d_test
+###
+d <- dplyr::mutate(d, DECISION_DATE = dht::check_dates(DECISION_DATE))
+
+# rmarkdown::render(input = '/app/aggregate_data_report.rmd',
+#                   params = list(d = d),
+#                   envir = new.env())
+
+
+tract_to_neighborhood <- readRDS('/app/tract_to_neighborhood.rds')
+neighborhood_shp <- readRDS('/app/ham_neighborhoods_dep_index_shp.rds')
+
+# Overall Summary
+date_min <- min(d$DECISION_DATE)
+date_max <- max(d$DECISION_DATE)
+
+# consider 'SCREENED IN AR' same as 'SCREENED IN'
+for (i in 1:nrow(d)) {
+  if (d$SCREENING_DECISION[i] == 'SCREENED IN AR') {
+    d$SCREENING_DECISION[i] <- 'SCREENED IN'
+  }
+}
+
+d_missing_alleg_add <- d %>%
+  filter(address_type == 'ALLEGATION_ADDRESS',
+         is.na(address)) %>%
+  select(PERSON_ID:address_type)
+
+d_fill_in_address <- d %>%
+  filter(PERSON_ID %in% d_missing_alleg_add$PERSON_ID,
+         address_type == 'CHILD_ADDRESS') %>%
+  group_by(PERSON_ID) %>%
+  arrange(desc(ADDRESS_START)) %>%
+  slice(1) %>%
+  select(INTAKE_ID, address:dep_index)
+
+d_missing_alleg_add <- left_join(d_missing_alleg_add, d_fill_in_address, by = 'PERSON_ID')
+
+d <- d %>%
+  filter(address_type == 'ALLEGATION_ADDRESS' & !is.na(address) |
+           address_type == 'CHILD_ADDRESS') %>%
+  bind_rows(d_missing_alleg_add) %>%
+  filter(address_type == 'ALLEGATION_ADDRESS')
+
+### Geocoding Summary
+d <- filter(d, !duplicated(PERSON_ID)) %>%
+  select( -ADDRESS_START, -address_type) %>%
+  mutate(week = lubridate::week(DECISION_DATE))
+
+### Weekly Counts
+d_neigh <- d  %>%
+  mutate(fips_tract_id = as.character(fips_tract_id)) %>%
+  filter(!is.na(lat)) %>%
+  left_join(tract_to_neighborhood, by='fips_tract_id') %>%
+  filter(!is.na(DECISION_DATE))
+
+screen_neighborhood <- d_neigh %>%
+  group_by(neighborhood, week) %>%
+  summarize(n_screened_in = sum(SCREENING_DECISION == 'SCREENED IN',na.rm = TRUE),
+            n_calls = n(), .groups = "drop")  %>%
+  mutate(screen_in_rate = round(n_screened_in/n_calls,2)) %>%
+  mutate_at(vars(n_screened_in, n_calls),
+            ~ifelse(.x < 5, NA, .x))
+
+##
+
+d_csv <- screen_neighborhood %>%
+  select(Neighborhood = neighborhood,
+         `Number of Calls` = n_calls,
+         `Number of Calls Screened In` = n_screened_in,
+         `Screen-In Rate` = screen_in_rate)
+
+path <- "/tmp/"
+write.csv(d_csv, paste(path,"weekly_report.csv", sep = ''))
+