Data Loading and Cleaning

suppressPackageStartupMessages(library(tidyverse))

## Warning: package 'readr' was built under R version 4.1.2

#Load Data
shootings <- read.csv("../../../dataset/fatal-police-shootings-data.csv")
clean <- filter(shootings, age != "", armed != "", gender != "", race != "", city != "", flee != "")
clean <- na.omit(clean)
#Remove a spot not in the US
clean <- clean[-which(clean$id == 5618),]

#subset the variables about location
location <- clean[c("date","city","state","longitude","latitude")]

#omit NA and blank(missing) values
location <- na.omit(location)

3 observations are deleted because of NA.

Exploratory Data Analysis

#Summary Statistics
table(clean$manner_of_death)

## 
##             shot shot and Tasered 
##             4675              270

table(clean$race) 

## 
##    A    B    H    N    O    W 
##   88 1322  915   73   42 2505

table(clean$state)

## 
##  AK  AL  AR  AZ  CA  CO  CT  DC  DE  FL  GA  HI  IA  ID  IL  IN  KS  KY  LA  MA 
##  31  91  65 214 733 175  17  17  11 343 179  23  31  41 107  99  52  84  94  32 
##  MD  ME  MI  MN  MO  MS  MT  NC  ND  NE  NH  NJ  NM  NV  NY  OH  OK  OR  PA  RI 
##  81  17  81  62 117  61  25 145   9  26  14  54  93  87  93 145 148  79 101   2 
##  SC  SD  TN  TX  UT  VA  VT  WA  WI  WV  WY 
##  73  12 132 427  60  93   7 127  86  36  13

table(clean$flee)

## 
##         Car        Foot Not fleeing       Other 
##         728         734        3288         195

summary(clean$age)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    6.00   27.00   34.00   36.67   45.00   91.00

#plot shooting points on US map
suppressPackageStartupMessages(library(rgdal))
library(usmap) #import the package
library(ggplot2) #use ggplot2 to add layer for visualization

coord <- location[c("longitude","latitude")]
coord <- usmap_transform(coord)
plot_usmap() +
  geom_point(data = coord, aes(x = longitude.1, y = latitude.1),
             color = "red", alpha = 0.25) + theme(legend.position = "right")

This US map shows that shootings happen intensively along west coast and 1/3 east part of the country.

clean %>%
  group_by(gender) %>%
  count() %>%
  ggplot(aes(x = "", y = n, fill = gender)) +
  geom_bar(width = 1, stat = "identity") +
  coord_polar("y", start=0) +
  scale_fill_brewer(palette="Blues") +
  theme_minimal() +
  geom_text(aes(y = n/2 + c(0, cumsum(n)[-length(n)]), 
            label = scales::percent(n/nrow(clean))), size=5)

This pie chart shows the percentage of male and female who involved in cases. We can see that men take 95% of all the cases.

ggplot(clean, aes(x = fct_reorder(state, age), y = age)) + 
  geom_point() +
  geom_boxplot() +
  xlab("State")+
  theme(axis.text.x = element_text(angle = 90))

This boxplot of states and age shows that, the average ages of crimes from the dataset do NOT variate a lot between different states, except for RI. We do not know the reason why the averge age is especially higher than other states, maybe we can figure it out after combing with more dataset such as income level.

clean %>%
  group_by(state, race) %>%
  count() %>%
  ggplot(aes(x = fct_reorder(state, n, .fun = sum), y = n, fill = race)) +
  xlab("state") +
  ylab("race number") +
  geom_bar(position = "dodge", stat = "identity", width = 0.75) +
  theme_bw() +
  scale_fill_brewer(palette = "Spectral") +
  theme(axis.text.x = element_text(angle = 90))

This bar chart shows the number of cases caused by different races in each state, arranging by the total number of cases, from low to high. We can see that CA, TX and Fl are the top 3 cities that shootings happen. To our surprise, although we observe that the average age in RI is the highest, it actually has the fewest cases. Among different races, it seems that white people caused most of the shootings from our dataset.

ggplot(clean, aes(x = signs_of_mental_illness, y = age)) +
  geom_boxplot() +
  xlab("Sign of Mental Illness")

ggplot(clean) + geom_bar(aes(x = flee, fill = flee))

The boxplot of sign of mental illness and age shows that whether the crime has mental illness or not does NOT reflect their age information. We may use other plot or model to determine if these two factors have correlation or not. The last bar chart is a counting for flee variable. We observe that most crimes do not flee; however, we can not decide the reason for this from the dataset we have for now. For those crimes who choose to flee, they seems prefer to drive cars or just run away.