wisconsin_crashes/R/crash_data_summaries.Rmd

---
title: "Crash Data Summaries"
output:
  html_document:
    toc: true
    toc_depth: 5
    toc_float:
      collapsed: false
      smooth_scroll: true
editor_options:
  chunk_output_type: console
---


# Input Data & Configuration

## Libraries
```{r libs, eval = TRUE, echo = TRUE, results = "show", warning = FALSE, error = TRUE, message = FALSE}
date()
rm(list=ls())
library(tidyverse)
library(RColorBrewer)
library(tidycensus)
library(ggrepel)
county_focus <- c("MILWAUKEE")
municipality_focus <- c("MILWAUKEE")
```

## Load TOPS data
```{r loadTOPS, eval = TRUE, echo = TRUE, results = "show", warning = FALSE, error = TRUE, message = FALSE}
load(file = "data/TOPS/TOPS_data.Rda")
load(file = "data/TOPS/vuln_roles.Rda")
load(file = "data/TOPS/retrieve_date.Rda")
load(file = "data/TOPS/injury_severity.Rda")
injury_severity_pal <- colorFactor(palette = injury_severity$color, levels = injury_severity$InjSevName)

```

## build data summaries for city
```{r citysummaries, eval = TRUE, echo = TRUE, results = "show", warning = FALSE, error = TRUE, message = FALSE}

data_summary <- list(NULL)
# crashes by year that resulted in a pedestrian fatality or severe injury
data_summary[["crash_by_year"]] <- TOPS_data %>%
  filter(MUNINAME %in% municipality_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  group_by(MUNINAME, year, vulnerable_role, ped_inj_name) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by race of pedestrian/bicyclist for focus year
data_summary[["crash_by_race"]] <- TOPS_data %>%
  filter(MUNINAME %in% municipality_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  group_by(MUNINAME, vulnerable_role, ped_inj_name, vulnerable_race) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by race of driver that resulted in a pedestrian fatality or severe injury
data_summary[["crash_by_driver_race"]] <- TOPS_data %>%
  filter(MUNINAME %in% municipality_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(driver_race = ifelse(ROLE1 %in% c("DR"), race_name1, ifelse(ROLE2 %in% c("DR"), race_name2, NA))) %>%
  group_by(MUNINAME, year, vulnerable_role, ped_inj_name, driver_race) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by age of pedestrian/bicyclist
data_summary[["crash_by_age"]] <- TOPS_data %>%
  filter(MUNINAME %in% municipality_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(vulnerable_age = ifelse(ROLE1 %in% vuln_roles, age1, ifelse(ROLE2 %in% vuln_roles, age2, NA))) %>%
  group_by(MUNINAME, year, vulnerable_role, ped_inj_name, vulnerable_age) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by age of driver that resulted in a severe injury or fatality of a pedestrian/bicyclist
data_summary[["crash_by_driver_age"]] <- TOPS_data %>%
  filter(MUNINAME %in% municipality_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(driver_age = ifelse(ROLE1 %in% c("DR"), age1, ifelse(ROLE2 %in% c("DR"), age2, NA))) %>%
  group_by(MUNINAME, year, vulnerable_role, ped_inj_name, driver_age) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by sex of pedestrian/bicyclist
data_summary[["crash_by_sex"]] <- TOPS_data %>%
  filter(MUNINAME %in% municipality_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(vulnerable_sex = ifelse(ROLE1 %in% vuln_roles, SEX1, ifelse(ROLE2 %in% vuln_roles, SEX1, NA))) %>%
  group_by(MUNINAME, year, vulnerable_role, ped_inj_name, vulnerable_sex) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by sex of driver that resulted in a severe injury or fatality of a pedestrian/bicyclist
data_summary[["crash_by_driver_sex"]] <- TOPS_data %>%
  filter(MUNINAME %in% municipality_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(driver_sex = ifelse(ROLE1 %in% c("DR"), SEX1, ifelse(ROLE2 %in% c("DR"), SEX2, NA))) %>%
  group_by(MUNINAME, year, vulnerable_role, ped_inj_name, driver_sex) %>%
  summarise(count = n_distinct(DOCTNMBR))

## export csv files for city ----
for(table_name in as.vector(names(data_summary[-1]))) {
  write_csv(data_summary[[table_name]], file = paste0("data_summaries/city/",table_name, ".csv"))
}
```


## build data summaries for county ----
```{r countysummaries, eval = TRUE, echo = TRUE, results = "show", warning = FALSE, error = TRUE, message = FALSE}

data_summary <- list(NULL)

# crashes by year that resulted in a pedestrian fatality or severe injury
data_summary[["crash_by_year"]] <- TOPS_data %>%
  filter(CNTYNAME %in% county_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  group_by(CNTYNAME, year, vulnerable_role, ped_inj_name) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by race of pedestrian/bicyclist for focus year
data_summary[["crash_by_race"]] <- TOPS_data %>%
  filter(CNTYNAME %in% county_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  group_by(CNTYNAME, vulnerable_role, ped_inj_name, vulnerable_race) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by race of driver that resulted in a pedestrian fatality or severe injury
data_summary[["crash_by_driver_race"]] <- TOPS_data %>%
  filter(CNTYNAME %in% county_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(driver_race = ifelse(ROLE1 %in% c("DR"), race_name1, ifelse(ROLE2 %in% c("DR"), race_name2, NA))) %>%
  group_by(CNTYNAME, year, vulnerable_role, ped_inj_name, driver_race) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by age of pedestrian/bicyclist
data_summary[["crash_by_age"]] <- TOPS_data %>%
  filter(CNTYNAME %in% county_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(vulnerable_age = ifelse(ROLE1 %in% vuln_roles, age1, ifelse(ROLE2 %in% vuln_roles, age2, NA))) %>%
  group_by(CNTYNAME, year, vulnerable_role, ped_inj_name, vulnerable_age) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by age of driver that resulted in a severe injury or fatality of a pedestrian/bicyclist
data_summary[["crash_by_driver_age"]] <- TOPS_data %>%
  filter(CNTYNAME %in% county_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(driver_age = ifelse(ROLE1 %in% c("DR"), age1, ifelse(ROLE2 %in% c("DR"), age2, NA))) %>%
  group_by(CNTYNAME, year, vulnerable_role, ped_inj_name, driver_age) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by sex of pedestrian/bicyclist
data_summary[["crash_by_sex"]] <- TOPS_data %>%
  filter(CNTYNAME %in% county_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(vulnerable_sex = ifelse(ROLE1 %in% vuln_roles, SEX1, ifelse(ROLE2 %in% vuln_roles, SEX1, NA))) %>%
  group_by(CNTYNAME, year, vulnerable_role, ped_inj_name, vulnerable_sex) %>%
  summarise(count = n_distinct(DOCTNMBR))

# crashes by sex of driver that resulted in a severe injury or fatality of a pedestrian/bicyclist
data_summary[["crash_by_driver_sex"]] <- TOPS_data %>%
  filter(CNTYNAME %in% county_focus) %>%
  filter(ped_inj %in% c("A", "K")) %>%
  mutate(driver_sex = ifelse(ROLE1 %in% c("DR"), SEX1, ifelse(ROLE2 %in% c("DR"), SEX2, NA))) %>%
  group_by(CNTYNAME, year, vulnerable_role, ped_inj_name, driver_sex) %>%
  summarise(count = n_distinct(DOCTNMBR))

## export csv files for county ----
for(table_name in as.vector(names(data_summary[-1]))) {
  write_csv(data_summary[[table_name]], file = paste0("data_summaries/county/",table_name, ".csv"))
}
```