library(RCurl)
library(XML)
library(jsonlite)
library(knitr)
library(kableExtra)
library(tidyverse)
library(tidyr)
library(geosphere)
library(leaflet)
library(plyr)
library(tools)
library(prettydoc)
library(ggrepel)Objective
For this project, I will investigate crime, and crime rates in NYC. The main questions I want to answer are:
Which NYC borough has the highest crime rate?
Which category of crimes are most common in New York City as a whole?
Data Sources
The 2 main data sources I will use for this project can be found on NYCOpenData (https://opendata.cityofnewyork.us/data/).
I will explore the “NYPD Arrests Data (Historic)” dataset, and will pull crime statistics from this set in order to explore crime rates in NYC. I will also use the “New York City Population by Borough, 1950 - 2040” dataset in order to calculate crime rates per New York City borough.
The datasets can be found at the following locations:
NYPD Arrests Data (Historic) - https://data.cityofnewyork.us/Public-Safety/NYPD-Arrests-Data-Historic-/8h9b-rp9u
New York City Population by Borough, 1950 - 2040 - https://data.cityofnewyork.us/City-Government/New-York-City-Population-by-Borough-1950-2040/xywu-7bv9
Import the Datasets.
The NYPD Arrests Data (Historic) csv file is 2.05 GB, so I decided to obtain the data via NYCOpenData’s JSON endpoint to save on memory. The New York City Population by Borough, 1950 - 2040 csv file is only 4 KB, but for consistency, I also imported this dataset via the JSON endpoint.
crime_data <- fromJSON("https://data.cityofnewyork.us/resource/qgea-i56i.json")
nyc_population <- fromJSON("https://data.cityofnewyork.us/resource/xywu-7bv9.json")Data Cleansing
NYPD Arrests Dataset
View the number of columns and rows in the crime_data dataset.
dim(crime_data)## [1] 1000 40Now look at the data structure, and column names.
str(crime_data)## 'data.frame': 1000 obs. of 40 variables:
## $ cmplnt_num : chr "522575447" "403507361" "631420068" "995609899" ...
## $ cmplnt_fr_dt : chr "2006-08-29T00:00:00.000" "2006-11-05T00:00:00.000" "2006-09-08T00:00:00.000" "2011-12-13T00:00:00.000" ...
## $ cmplnt_fr_tm : chr "13:00:00" "11:00:00" "23:30:00" "18:40:00" ...
## $ addr_pct_cd : chr "43" "66" "106" "79" ...
## $ rpt_dt : chr "2006-08-30T00:00:00.000" "2006-11-05T00:00:00.000" "2006-09-09T00:00:00.000" "2011-12-13T00:00:00.000" ...
## $ ky_cd : chr "578" "107" "347" "341" ...
## $ ofns_desc : chr "HARRASSMENT 2" "BURGLARY" "INTOXICATED & IMPAIRED DRIVING" "PETIT LARCENY" ...
## $ pd_cd : chr "638" "221" "905" "333" ...
## $ pd_desc : chr "HARASSMENT,SUBD 3,4,5" "BURGLARY,RESIDENCE,DAY" "INTOXICATED DRIVING,ALCOHOL" "LARCENY,PETIT FROM STORE-SHOPL" ...
## $ crm_atpt_cptd_cd : chr "COMPLETED" "COMPLETED" "COMPLETED" "COMPLETED" ...
## $ law_cat_cd : chr "VIOLATION" "FELONY" "MISDEMEANOR" "MISDEMEANOR" ...
## $ boro_nm : chr "BRONX" "BROOKLYN" "QUEENS" "BROOKLYN" ...
## $ loc_of_occur_desc : chr "INSIDE" "INSIDE" "FRONT OF" "INSIDE" ...
## $ prem_typ_desc : chr "RESIDENCE - APT. HOUSE" "RESIDENCE - APT. HOUSE" "STREET" "CHAIN STORE" ...
## $ juris_desc : chr "N.Y. POLICE DEPT" "N.Y. POLICE DEPT" "N.Y. POLICE DEPT" "N.Y. POLICE DEPT" ...
## $ jurisdiction_code : chr "0" "0" "0" "0" ...
## $ parks_nm : logi NA NA NA NA NA NA ...
## $ housing_psa : chr "NA" "NA" "NA" "NA" ...
## $ x_coord_cd : chr "1018029" "982556" "1028213" "1000788" ...
## $ y_coord_cd : chr "240747" "171385" "186786" "189718" ...
## $ susp_race : chr "UNKNOWN" NA NA NA ...
## $ susp_sex : chr "M" NA NA NA ...
## $ latitude : chr "40.827414051" "40.637096864" "40.679260229" "40.687401619" ...
## $ longitude : chr "-73.877945775" "-74.006105014" "-73.841500185" "-73.940369194" ...
## $ lat_lon :'data.frame': 1000 obs. of 2 variables:
## ..$ latitude : chr "40.827414051" "40.637096864" "40.679260229" "40.687401619" ...
## ..$ longitude: chr "-73.877945775" "-74.006105014" "-73.841500185" "-73.940369194" ...
## $ patrol_boro : chr "PATROL BORO BRONX" "PATROL BORO BKLYN SOUTH" "PATROL BORO QUEENS SOUTH" "PATROL BORO BKLYN NORTH" ...
## $ vic_age_group : chr "25-44" "45-64" NA NA ...
## $ vic_race : chr "BLACK HISPANIC" "ASIAN / PACIFIC ISLANDER" "UNKNOWN" "UNKNOWN" ...
## $ vic_sex : chr "F" "F" "E" "D" ...
## $ :@computed_region_efsh_h5xi: chr "11610" "18180" "24013" "18181" ...
## $ :@computed_region_f5dn_yrer: chr "58" "2" "62" "69" ...
## $ :@computed_region_yeji_bk3q: chr "5" "2" "3" "2" ...
## $ :@computed_region_92fq_4b7q: chr "31" "7" "41" "49" ...
## $ :@computed_region_sbqj_enih: chr "26" "39" "64" "51" ...
## $ cmplnt_to_dt : chr NA "2006-11-05T00:00:00.000" "2006-09-09T00:00:00.000" "2011-12-13T00:00:00.000" ...
## $ cmplnt_to_tm : chr NA "17:40:00" "00:01:00" "18:49:00" ...
## $ susp_age_group : chr NA NA NA NA ...
## $ hadevelopt : chr NA NA NA NA ...
## $ transit_district : chr NA NA NA NA ...
## $ station_name : chr NA NA NA NA ...List the dataset column names so we can see which columns are useful for analysis purposes.
colnames(crime_data)## [1] "cmplnt_num" "cmplnt_fr_dt"
## [3] "cmplnt_fr_tm" "addr_pct_cd"
## [5] "rpt_dt" "ky_cd"
## [7] "ofns_desc" "pd_cd"
## [9] "pd_desc" "crm_atpt_cptd_cd"
## [11] "law_cat_cd" "boro_nm"
## [13] "loc_of_occur_desc" "prem_typ_desc"
## [15] "juris_desc" "jurisdiction_code"
## [17] "parks_nm" "housing_psa"
## [19] "x_coord_cd" "y_coord_cd"
## [21] "susp_race" "susp_sex"
## [23] "latitude" "longitude"
## [25] "lat_lon" "patrol_boro"
## [27] "vic_age_group" "vic_race"
## [29] "vic_sex" ":@computed_region_efsh_h5xi"
## [31] ":@computed_region_f5dn_yrer" ":@computed_region_yeji_bk3q"
## [33] ":@computed_region_92fq_4b7q" ":@computed_region_sbqj_enih"
## [35] "cmplnt_to_dt" "cmplnt_to_tm"
## [37] "susp_age_group" "hadevelopt"
## [39] "transit_district" "station_name"We don’t need all of the information in the crime_data dataset for our analysis, so we take what we need, and leave what we don’t.
Clean the crime_data dataset so that we can use it for our analysis.
# Extract the columns containing the data that we need.
crime_data_refined <- select(crime_data, rpt_dt, ofns_desc, pd_desc, ofns_desc, law_cat_cd, boro_nm, addr_pct_cd, latitude, longitude, lat_lon)
# Remove NAs from the data.
sapply(crime_data_refined, function(x) sum(is.na(x)))## rpt_dt ofns_desc pd_desc law_cat_cd boro_nm addr_pct_cd
## 0 9 0 0 2 0
## latitude longitude lat_lon
## 9 9 18clean_crime_data <- na.omit(crime_data_refined)
sapply(clean_crime_data, function(x) sum(is.na(x)))## rpt_dt ofns_desc pd_desc law_cat_cd boro_nm addr_pct_cd
## 0 0 0 0 0 0
## latitude longitude lat_lon
## 0 0 0summary(clean_crime_data)## rpt_dt ofns_desc pd_desc
## Length:980 Length:980 Length:980
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :character
## law_cat_cd boro_nm addr_pct_cd
## Length:980 Length:980 Length:980
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :character
## latitude longitude
## Length:980 Length:980
## Class :character Class :character
## Mode :character Mode :character
## lat_lon.latitude lat_lon.longitude
## Length:980 Length:980
## Class :character Class :character
## Mode :character Mode :characterLook at a sample of the clean data.
sample_n(clean_crime_data, 5)## rpt_dt ofns_desc
## 1 2010-06-01T00:00:00.000 GRAND LARCENY
## 2 2009-07-27T00:00:00.000 OFF. AGNST PUB ORD SENSBLTY &
## 3 2006-08-18T00:00:00.000 MISCELLANEOUS PENAL LAW
## 4 2007-09-12T00:00:00.000 HARRASSMENT 2
## 5 2006-11-25T00:00:00.000 CRIMINAL MISCHIEF & RELATED OF
## pd_desc law_cat_cd
## 1 LARCENY,GRAND FROM BUILDING (NON-RESIDENCE) UNATTENDED FELONY
## 2 AGGRAVATED HARASSMENT 2 MISDEMEANOR
## 3 CRIMINAL CONTEMPT 1 FELONY
## 4 HARASSMENT,SUBD 3,4,5 VIOLATION
## 5 MISCHIEF, CRIMINAL 4, OF MOTOR MISDEMEANOR
## boro_nm addr_pct_cd latitude longitude lat_lon.latitude
## 1 MANHATTAN 1 40.703684905 -74.009667273 40.703684905
## 2 MANHATTAN 17 40.755516375 -73.970988622 40.755516375
## 3 BROOKLYN 84 40.691094994 -73.996074542 40.691094994
## 4 BRONX 42 40.832036889 -73.886736442 40.832036889
## 5 BROOKLYN 83 40.70685174 -73.919028249 40.70685174
## lat_lon.longitude
## 1 -74.009667273
## 2 -73.970988622
## 3 -73.996074542
## 4 -73.886736442
## 5 -73.919028249NYC Population Dataset
View the number of columns and rows in the nyc_population dataset.
dim(nyc_population)## [1] 6 22Look at the data structure, and column names.
str(nyc_population)## 'data.frame': 6 obs. of 22 variables:
## $ age_group : chr "Total Population" "Total Population" "Total Population" "Total Population" ...
## $ borough : chr "NYC Total" " Bronx" " Brooklyn" " Manhattan" ...
## $ _1950 : chr "7891957" "1451277" "2738175" "1960101" ...
## $ _1950_boro_share_of_nyc_total: chr "100.00" "18.39" "34.70" "24.84" ...
## $ _1960 : chr "7781984" "1424815" "2627319" "1698281" ...
## $ _1960_boro_share_of_nyc_total: chr "100.00" "18.31" "33.76" "21.82" ...
## $ _1970 : chr "7894862" "1471701" "2602012" "1539233" ...
## $ _1970_boro_share_of_nyc_total: chr "100.00" "18.64" "32.96" "19.50" ...
## $ _1980 : chr "7071639" "1168972" "2230936" "1428285" ...
## $ _1980_boro_share_of_nyc_total: chr "100.00" "16.53" "31.55" "20.20" ...
## $ _1990 : chr "7322564" "1203789" "2300664" "1487536" ...
## $ _1990_boro_share_of_nyc_total: chr "100.00" "16.44" "31.42" "20.31" ...
## $ _2000 : chr "8008278" "1332650" "2465326" "1537195" ...
## $ _2000_boro_share_of_nyc_total: chr "100.00" "16.64" "30.78" "19.20" ...
## $ _2010 : chr "8242624" "1385108" "2552911" "1585873" ...
## $ _2010_boro_share_of_nyc_total: chr "100.00" "16.80" "30.97" "19.24" ...
## $ _2020 : chr "8550971" "1446788" "2648452" "1638281" ...
## $ _2020_boro_share_of_nyc_total: chr "100.00" "16.92" "30.97" "19.16" ...
## $ _2030 : chr "8821027" "1518998" "2754009" "1676720" ...
## $ _2030_boro_share_of_nyc_total: chr "100.00" "17.22" "31.22" "19.01" ...
## $ _2040 : chr "9025145" "1579245" "2840525" "1691617" ...
## $ _2040_boro_share_of_nyc_total: chr "100.00" "17.50" "31.47" "18.74" ...List the dataset column names.
colnames(nyc_population)## [1] "age_group" "borough"
## [3] "_1950" "_1950_boro_share_of_nyc_total"
## [5] "_1960" "_1960_boro_share_of_nyc_total"
## [7] "_1970" "_1970_boro_share_of_nyc_total"
## [9] "_1980" "_1980_boro_share_of_nyc_total"
## [11] "_1990" "_1990_boro_share_of_nyc_total"
## [13] "_2000" "_2000_boro_share_of_nyc_total"
## [15] "_2010" "_2010_boro_share_of_nyc_total"
## [17] "_2020" "_2020_boro_share_of_nyc_total"
## [19] "_2030" "_2030_boro_share_of_nyc_total"
## [21] "_2040" "_2040_boro_share_of_nyc_total"Clean the nyc_population dataset.
# We only need 2 columns from the table (Borough and Population).
nyc_population_refined <- select(nyc_population, borough, '_2020')
# Remove NAs from the data.
sapply(nyc_population_refined, function(x) sum(is.na(x)))## borough _2020
## 0 0clean_pop_data <- na.omit(nyc_population_refined)
sapply(clean_pop_data, function(x) sum(is.na(x)))## borough _2020
## 0 0# Get rid of the NYC Total population row as we don't need it.
clean_pop_data = clean_pop_data[-1,]
# Rename the columns.
colnames(clean_pop_data) <- c('Borough', 'Population')
sample_n(clean_pop_data, 5)## Borough Population
## 1 Queens 2330295
## 2 Brooklyn 2648452
## 3 Staten Island 487155
## 4 Bronx 1446788
## 5 Manhattan 1638281Data Analysis
Crime Frequency by Borough
# Count the occurences of a borough name to give us an idea of the crime rate in the borough.
crimes_per_borough <- table(clean_crime_data$boro_nm)
head(crimes_per_borough)##
## BRONX BROOKLYN MANHATTAN QUEENS STATEN ISLAND
## 211 308 233 176 52Create a new dataframe that contains a population column for each borough.
crimes_dataframe <- as.data.frame(crimes_per_borough, stringsAsFactors=FALSE)
population_dataframe <- as.data.frame(clean_pop_data, stringsAsFactors=FALSE)
# Add column names to the crimes_dataframe.
colnames(crimes_dataframe) <- c('Borough', 'Crimes')
# Convert borough names to title case form so that we can join on 'Borough'
# with the population_dataframe.
crimes_dataframe$Borough <- str_to_title(crimes_dataframe$Borough)
# Join the 2 tables together.
crimes_population <- join(crimes_dataframe, population_dataframe, by = 'Borough', type = 'right')
crimes_population$Crimes <- crimes_dataframe$Crimes
kable(crimes_population, "html", escape = F) %>%
kable_styling("striped", full_width = T) %>%
column_spec(1, bold = T)| Borough | Crimes | Population |
|---|---|---|
| Bronx | 211 | 1446788 |
| Brooklyn | 308 | 2648452 |
| Manhattan | 233 | 1638281 |
| Queens | 176 | 2330295 |
| Staten Island | 52 | 487155 |
Calculate the crime rate for each Borough, and add a CrimeRate column to the crimes_population dataframe.
crime_rate <- as.numeric(crimes_population$Crimes) / as.numeric(crimes_population$Population) * 100000
crimes_population$CrimeRate <- round(as.numeric(format(crime_rate, scientific = FALSE)), 2)
kable(crimes_population, "html", escape = F) %>%
kable_styling("striped", full_width = T) %>%
column_spec(1, bold = T)| Borough | Crimes | Population | CrimeRate |
|---|---|---|---|
| Bronx | 211 | 1446788 | 14.58 |
| Brooklyn | 308 | 2648452 | 11.63 |
| Manhattan | 233 | 1638281 | 14.22 |
| Queens | 176 | 2330295 | 7.55 |
| Staten Island | 52 | 487155 | 10.67 |
Plot a bar graph displaying the crime rates for each borough.
crime_by_borough_plot <-ggplot(crimes_population, aes(x = Borough, y = CrimeRate, fill = Borough)) +
geom_bar(stat = 'identity') +
labs(title = 'Crime Rates by Borough',
x = 'Borough',
y = 'Crime Rate')
crime_by_borough_plot
New York City crime by category
cat_nyc <- select(clean_crime_data, law_cat_cd)
crime_category_table <- table(cat_nyc$law_cat_cd)
crime_category_table <- sort(crime_category_table, decreasing = TRUE)
crime_by_category <- as.data.frame(crime_category_table, stringsAsFactors = FALSE)
colnames(crime_by_category) <- c("Category", "Frequency")
crime_by_category$Percentage <- crime_by_category$Frequency / sum(crime_by_category$Frequency) * 100
crime_by_category## Category Frequency Percentage
## 1 MISDEMEANOR 572 58.36735
## 2 FELONY 291 29.69388
## 3 VIOLATION 117 11.93878Bar graph displaying arrest frequency by crime category in NYC.
crime_by_category_plot <-ggplot(crime_by_category, aes(x = Category, y = Frequency, fill = Category)) +
geom_bar(stat="identity") +
labs(title = 'Crime by Category',
x = 'Category',
y = 'Frequency')
crime_by_category_plot
Pie chart displaying arrest frequency by crime category in NYC.
crime_category_pie_chart <-ggplot(crime_by_category, aes(x = '', y = Frequency, fill = Category)) +
geom_bar(stat ='identity') +
coord_polar(theta = 'y') +
scale_x_discrete('') +
labs(title = 'Crime by Category',
x = 'Category',
y = 'Frequency')
crime_category_pie_chart
Over half of all arrests in NYC are for misdemeanor offences (crimes such as petty theft, disorderly conduct, public intoxication, assault, etc.). Just under half as many felony arrests were made (crimes such as murder, burglary, arson, etc.), and the remainder of arrests were for violations (disorderly conduct, loitering, etc.). The low level of violation arrests is not surprising as these are low level crimes which rarely result in arrest.
These statistics make sense as felonies are less likely crimes, misdemeanors are less serious crimes, but often result in arrest, and violations rarely lead to arrest.
Crimes Leading to The Least Arrests in NYC
least_crimes <- sort(table(clean_crime_data$pd_desc), decreasing = TRUE)
least_crimes <- data.frame(least_crimes[least_crimes < 10])
colnames(least_crimes) <- c('Crime', 'Frequency')
least_crimes$Percentage <- least_crimes$Frequency / sum(least_crimes$Frequency) * 100
kable(least_crimes, "html", escape = F) %>%
kable_styling("striped", full_width = T) %>%
column_spec(1, bold = T)| Crime | Frequency | Percentage |
|---|---|---|
| FORGERY,ETC.-MISD. | 9 | 4.109589 |
| FRAUD,UNCLASSIFIED-FELONY | 8 | 3.652968 |
| LARCENY,GRAND BY THEFT OF CREDIT CARD | 8 | 3.652968 |
| LEAVING SCENE-ACCIDENT-PERSONA | 8 | 3.652968 |
| MISCHIEF, CRIMINAL 3 & 2, OF M | 8 | 3.652968 |
| FORGERY,ETC.,UNCLASSIFIED-FELO | 7 | 3.196347 |
| FRAUD,UNCLASSIFIED-MISDEMEANOR | 6 | 2.739726 |
| VIOLATION OF ORDER OF PROTECTI | 6 | 2.739726 |
| BURGLARY,COMMERCIAL,NIGHT | 5 | 2.283105 |
| CONTROLLED SUBSTANCE,POSSESS. | 5 | 2.283105 |
| LARCENY,GRAND BY BANK ACCT COMPROMISE-REPRODUCED CHECK | 5 | 2.283105 |
| MISCHIEF,CRIMINAL, UNCL 2ND | 5 | 2.283105 |
| RESISTING ARREST | 5 | 2.283105 |
| BRIBERY,PUBLIC ADMINISTRATION | 4 | 1.826484 |
| BURGLARY,RESIDENCE,NIGHT | 4 | 1.826484 |
| LARCENY,PETIT OF VEHICLE ACCES | 4 | 1.826484 |
| MARIJUANA, POSSESSION | 4 | 1.826484 |
| ROBBERY,PUBLIC PLACE INSIDE | 4 | 1.826484 |
| SEXUAL ABUSE 3,2 | 4 | 1.826484 |
| THEFT OF SERVICES, UNCLASSIFIE | 4 | 1.826484 |
| AGGRAVATED HARASSMENT 1 | 3 | 1.369863 |
| ARSON, MOTOR VEHICLE 1 2 3 & 4 | 3 | 1.369863 |
| BURGLARY,RESIDENCE,UNKNOWN TIM | 3 | 1.369863 |
| CHILD, ENDANGERING WELFARE | 3 | 1.369863 |
| LARCENY,GRAND BY ACQUIRING LOS | 3 | 1.369863 |
| LARCENY,GRAND OF MOTORCYCLE | 3 | 1.369863 |
| MARIJUANA, SALE 4 & 5 | 3 | 1.369863 |
| RAPE 1 | 3 | 1.369863 |
| STOLEN PROPERTY 3,POSSESSION | 3 | 1.369863 |
| TRAFFIC,UNCLASSIFIED MISDEMEAN | 3 | 1.369863 |
| WEAPONS POSSESSION 3 | 3 | 1.369863 |
| ADM.CODE,UNCLASSIFIED MISDEMEA | 2 | 0.913242 |
| ARSON 2,3,4 | 2 | 0.913242 |
| ASSAULT POLICE/PEACE OFFICER | 2 | 0.913242 |
| CONTEMPT,CRIMINAL | 2 | 0.913242 |
| FORGERY,M.V. REGISTRATION | 2 | 0.913242 |
| LARCENY,GRAND PERSON,NECK CHAI | 2 | 0.913242 |
| LARCENY,PETIT BY ACQUIRING LOS | 2 | 0.913242 |
| LARCENY,PETIT BY DISHONEST EMP | 2 | 0.913242 |
| LARCENY,PETIT OF BICYCLE | 2 | 0.913242 |
| NY STATE LAWS,UNCLASSIFIED FEL | 2 | 0.913242 |
| PUBLIC ADMINISTATION,UNCLASS M | 2 | 0.913242 |
| PUBLIC ADMINISTRATION,UNCLASSI | 2 | 0.913242 |
| RECKLESS ENDANGERMENT 1 | 2 | 0.913242 |
| ROBBERY,COMMERCIAL UNCLASSIFIED | 2 | 0.913242 |
| ROBBERY,DWELLING | 2 | 0.913242 |
| STRANGULATION 1ST | 2 | 0.913242 |
| UNAUTHORIZED USE VEHICLE 2 | 2 | 0.913242 |
| WEAPONS POSSESSION 1 & 2 | 2 | 0.913242 |
| AGGRAVATED SEXUAL ASBUSE | 1 | 0.456621 |
| BAIL JUMPING 1 & 2 | 1 | 0.456621 |
| BURGLARY,COMMERCIAL,DAY | 1 | 0.456621 |
| BURGLARY,COMMERCIAL,UNKNOWN TI | 1 | 0.456621 |
| CONTROLLED SUBSTANCE, INTENT T | 1 | 0.456621 |
| CONTROLLED SUBSTANCE,INTENT TO | 1 | 0.456621 |
| CONTROLLED SUBSTANCE,SALE 3 | 1 | 0.456621 |
| FACILITATION 4, CRIMINAL | 1 | 0.456621 |
| FORGERY,DRIVERS LICENSE | 1 | 0.456621 |
| IMPRISONMENT 2,UNLAWFUL | 1 | 0.456621 |
| LARCENY, GRAND OF AUTO - ATTEM | 1 | 0.456621 |
| LARCENY,GRAND BY FALSE PROMISE | 1 | 0.456621 |
| LARCENY,GRAND FROM PERSON,PURS | 1 | 0.456621 |
| LARCENY,GRAND FROM STORE-SHOPL | 1 | 0.456621 |
| LARCENY,GRAND OF TRUCK | 1 | 0.456621 |
| LARCENY,GRAND OF VEHICULAR/MOTORCYCLE ACCESSORIES | 1 | 0.456621 |
| LARCENY,PETIT BY CHECK USE | 1 | 0.456621 |
| LARCENY,PETIT BY CREDIT CARD U | 1 | 0.456621 |
| LARCENY,PETIT BY FALSE PROMISE | 1 | 0.456621 |
| LARCENY,PETIT FROM TRUCK | 1 | 0.456621 |
| MARIJUANA, POSSESSION 1, 2 & 3 | 1 | 0.456621 |
| MENACING 1ST DEGREE (VICT PEAC | 1 | 0.456621 |
| MISCHIEF, CRIMINAL 3&2, BY FIR | 1 | 0.456621 |
| MISCHIEF, CRIMINAL 4, BY FIRE | 1 | 0.456621 |
| OBSTR BREATH/CIRCUL | 1 | 0.456621 |
| PETIT LARCENY-CHECK FROM MAILB | 1 | 0.456621 |
| RAPE 2 | 1 | 0.456621 |
| RECKLESS DRIVING | 1 | 0.456621 |
| RECKLESS ENDANGERMENT 2 | 1 | 0.456621 |
| ROBBERY,NECKCHAIN/JEWELRY | 1 | 0.456621 |
| ROBBERY,ON BUS/ OR BUS DRIVER | 1 | 0.456621 |
| ROBBERY,POCKETBOOK/CARRIED BAG | 1 | 0.456621 |
| ROBBERY,RESIDENTIAL COMMON AREA | 1 | 0.456621 |
| SODOMY 1 | 1 | 0.456621 |
| STOLEN PROPERTY 2,1,POSSESSION | 1 | 0.456621 |
| TRESPASS 1,CRIMINAL | 1 | 0.456621 |
| UNAUTHORIZED USE VEHICLE 3 | 1 | 0.456621 |
Finally, I wanted to look at the crimes that lead to the least amount of arrests in NYC. It is interesting that these are all low level crimes. This would suggest that they are crimes that people rarely get arrested for, or that they are so low level that people rarely report these crimes.
Conclusion
The number of arrests for low level crimes are consistently low throughout this report. This would suggest that such crimes often go unreported, or they are so low level that they do not warrent arrest. The fact that misdemeanors lead to the most arrests is of no surprise. Such crimes are low level, but they lead to arrest. The fact that they are low level probably means people are more likely to commit these crimes with little effort to hide their actions.
When it comes to safe boroughs in New York, it appears that Queens is the safest borough, followed by Staten Island. The most dangerous borough is the Bronx, followed by Manhattan.
I would not have expected Manhattan to have a high crime rate, so given more time, this would have been an interesting point to explore further. Either there are interesting reason for this lurking below the surface, or the data is somehow skewed.