Crime_Analysis_Chapel_Hill_script.R

# Set working directory to Source File Location

# Load libraries
library(tidyverse)  # dealing with tabular data
library(lubridate) # handling dates
library(dplyr) # data manipulation
library(tidyr) # tidy data
library(data.table) # table formated result
library(ggplot2) # visualization
library(scales)# graphical scaling

# Importing my .csv data to the Global Environment
# 3 years of data
arrest_data_2018 = read_csv("2018_police_data.csv")
arrest_data_2019 = read_csv("2019_police_data.csv")
arrest_data_2020 = read_csv("2020_police_data.csv")

arrest_data <- rbind(arrest_data_2018,arrest_data_2019,arrest_data_2020)

# **********************
# Analysis of Year 2018 Data

# Run Each Month Analysis at a time to see each plot

# Month Analysis
#extracting month from date and making a new column for it
arrest_data_2018$month_of_arrest <- format(as.POSIXct(arrest_data_2018$Date_of_Arrest, format = "%Y/%m/%d %H:%M:%S"), format = "%m")

# changing months to factors
arrest_data_2018$month_of_arrest <- as.factor(arrest_data_2018$month_of_arrest)

# plotting the arrests in 2018 by months
month_data <- arrest_data_2018 %>%
  group_by(month_of_arrest) %>%
  summarise(Total = n()) # grouping the data into month by count

#see in tabular form
data.table(month_data)

# Visualize the data
ggplot(month_data, aes(month_of_arrest,Total)) +
  geom_bar(stat = "identity", fill = "skyblue", color = "black") +
  ggtitle("Year 2018 Arrests By Month") +
  theme(legend.position = "none") +
  scale_y_continuous(labels = comma)

#Race Analysis
race_data <- arrest_data_2018 %>%
  group_by(Race) %>%
  summarise(Total = n()) # grouping the data into race by count

#filter data to remove NA
race_data = filter(race_data, !is.na(Race))

#see in tabular form
data.table(race_data)

# Visualize the data
ggplot(race_data, aes(Race,Total)) +
  geom_bar(stat = "identity", fill = "aquamarine1", color = "white") +
  ggtitle("Year 2018 Arrests By Race") +
  theme(legend.position = "none") +
  scale_y_continuous(labels = comma)

# **********************
# Analysis of Year 2019 Data
# Month Analysis
#extracting month from date and making a new column for it
arrest_data_2019$month_of_arrest <- format(as.POSIXct(arrest_data_2019$Date_of_Arrest, format = "%Y/%m/%d %H:%M:%S"), format = "%m")

# changing months to factors
arrest_data_2019$month_of_arrest <- as.factor(arrest_data_2019$month_of_arrest)

# plotting the arrests in 2019 by months
month_data <- arrest_data_2019 %>%
  group_by(month_of_arrest) %>%
  summarise(Total = n()) # grouping the data into month by count

#see in tabular form
data.table(month_data)

# Visualize the data
ggplot(month_data, aes(month_of_arrest,Total)) +
  geom_bar(stat = "identity", fill = "pink", color = "black") +
  ggtitle("Year 2019 Arrests By Month") +
  theme(legend.position = "none") +
  scale_y_continuous(labels = comma)

#Race Analysis
race_data <- arrest_data_2019 %>%
  group_by(Race) %>%
  summarise(Total = n()) # grouping the data into race by count

#filter data to remove NA
race_data = filter(race_data, !is.na(Race))

#see in tabular form
data.table(race_data)

# Visualize the data
ggplot(race_data, aes(Race,Total)) +
  geom_bar(stat = "identity", fill = "darkgoldenrod1", color = "white") +
  ggtitle("Year 2019 Arrests By Race") +
  theme(legend.position = "none") +
  scale_y_continuous(labels = comma)


# **********************
# Analysis of Year 2020 Data
# Month Analysis
#extracting month from date and making a new column for it
arrest_data_2020$month_of_arrest <- format(as.POSIXct(arrest_data_2020$Date_of_Arrest, format = "%Y/%m/%d %H:%M:%S"), format = "%m")

# changing months to factors
arrest_data_2020$month_of_arrest <- as.factor(arrest_data_2020$month_of_arrest)

# plotting the arrests in 2020 by months
month_data <- arrest_data_2020 %>%
  group_by(month_of_arrest) %>%
  summarise(Total = n()) # grouping the data into month by count

#see in tabular form
data.table(month_data)

# Visualize the data
ggplot(month_data, aes(month_of_arrest,Total)) +
  geom_bar(stat = "identity", fill = "wheat2", color = "black") +
  ggtitle("Year 2020 Arrests By Month") +
  theme(legend.position = "none") +
  scale_y_continuous(labels = comma)

#Race Analysis
race_data <- arrest_data_2020 %>%
  group_by(Race) %>%
  summarise(Total = n()) # grouping the data into race by count

#filter data to remove NA
race_data = filter(race_data, !is.na(Race))

#see in tabular form
data.table(race_data)

# Visualize the data
ggplot(race_data, aes(Race,Total)) +
  geom_bar(stat = "identity", fill = "coral1", color = "white") +
  ggtitle("Year 2020 Arrests By Race") +
  theme(legend.position = "none") +
  scale_y_continuous(labels = comma)

# **********************
# Analysis of 2018 -2020 Data
# Month Analysis
#extracting month from date and making a new column for it
arrest_data$month_of_arrest <- format(as.POSIXct(arrest_data$Date_of_Arrest, format = "%Y/%m/%d %H:%M:%S"), format = "%m")

# changing months to factors
arrest_data$month_of_arrest <- as.factor(arrest_data$month_of_arrest)

# plotting the arrests in 2018 - 2020 by months
month_data <- arrest_data %>%
  group_by(month_of_arrest) %>%
  summarise(Total = n()) # grouping the data into month by count

#see in tabular form
data.table(month_data)

# Visualize the data
ggplot(month_data, aes(month_of_arrest,Total)) +
  geom_bar(stat = "identity", fill = "thistle1", color = "black") +
  ggtitle("Year 2018-2020 Arrests By Month") +
  theme(legend.position = "none") +
  scale_y_continuous(labels = comma)

#Race Analysis
colors = c("darkolivegreen1", "lavender", "pink", "white", "orange", "purple", "lemonchiffon1") 
table(arrest_data$Race)
pie(table(arrest_data$Race), main = "Year 2018-2020 Arrests By Race", col=colors, labels="")
legend(1, 1, c("Asians","Blacks","Hispanics","Islanders","Others","Unknown"), cex = 0.4, fill = colors)
#dev.off() #Switch on or off when necessary
#adjust margins if needed to prevent error message

# Age Analysis
age_summary_by_race <- arrest_data%>%
  group_by(Race)%>%
  summarise(Average=mean(Age, na.rm=TRUE),
            Maximum=max(Age, na.rm = TRUE),
            Minimum=min(Age, na.rm = TRUE),
            Median=median(Age, na.rm = TRUE))
age_summary_by_race

write_csv(age_summary_by_race, "age_summary_by_race.csv")

age_summary_by_gender <- arrest_data%>%
  group_by(Gender)%>%
  summarise(Average=mean(Age, na.rm=TRUE),
            Maximum=max(Age, na.rm = TRUE),
            Minimum=min(Age, na.rm = TRUE),
            Median=median(Age, na.rm = TRUE))
age_summary_by_gender

write_csv(age_summary_by_gender, "age_summary_by_gender.csv")

# Weapon Present Analysis
weapon_data <- arrest_data %>%
  group_by(Weapon_Present) %>%
  summarise(Total = n()) # grouping the data into Weapon Type by count

#filter data to remove NA
weapon_data = filter(weapon_data, !is.na(Weapon_Present))
weapon_data

write_csv(weapon_data, "weapon_data.csv")

# Under the Influence (UI) Analysis
UI_data <- arrest_data %>%
  group_by(Drugs_or_Alcohol_Present) %>%
  summarise(Total = n()) # grouping the data into Drugs or Alcohol Present data by count

#filter data to remove NA
UI_data = filter(UI_data, !is.na(Drugs_or_Alcohol_Present))
UI_data

colors1 = c("aquamarine2", "white", "lightgoldenrod1")
pie(table(arrest_data$Drugs_or_Alcohol_Present), main = "Year 2018-2020 Arrests By UI", col=colors1, labels="")
legend(1, 1, c("No","Unknown","Yes"), cex = 0.4, fill = colors1)
# adjust margins if needed to prevent error message

# Crime Type Analysis
crime_type_data <- arrest_data %>%
  group_by(Primary_Charge) %>%
  summarise(Total = n()) # grouping the data into Primary Charge by count

#filter data to remove NA
crime_type_data = filter(crime_type_data, !is.na(Primary_Charge))
crime_type_data
view(crime_type_data)