Crime in the United States

by Volume and Rate per 100,000 Inhabitants, 1997–2016

In order to gain a better picture of how crime is commited in the united states and understand which types of crimes are most common we can use this FBI data set which includes crime data from 1997-2016 and examine trends in crime during these years.

# Import libraries to be used for the project import matplotlib.pyplot as plt import seaborn as sns import pandas as pd

#Import the crime data from FBI website and create a new data frame and rename the columns crime_df= pd.read_csv('crime.csv', header=None) crime_df.columns = ["Year","Population","Violent Crime","Violent Crime Rate","Murder and nonnegligent manslaughter","Murder and nonnegligent manslaughter Rate","Rape Revised def","Rape Revised def rate","Rape Legacy def","Rape Legacy def rate", "Robbery","Robbery rate","Aggrevated Assult","Aggrevated Assult rate","Property Crime","Property crime rate","Burglary", "Burglary rate", "Larceny-theft","Larceny-theft rate", "Motor vehicle theft", "Motor vehicle theft rate"]

# drop row one and rename data frame to df1 for ease in future manipulation df1= crime_df[1:] df1

# replace all commas df1 = df1.replace(',','', regex=True)

# convert each columnn to an integer column df1["Population"]= df1["Population"].astype (int)

df1["Violent Crime"]= df1["Violent Crime"].astype (int)

df1["Murder and nonnegligent manslaughter"]= df1["Murder and nonnegligent manslaughter"].astype (int)

df1["Rape Legacy def"]= df1["Rape Legacy def"].astype (int)

df1["Robbery"]= df1["Robbery"].astype (int)

df1["Aggrevated Assult"]= df1["Aggrevated Assult"].astype (int)

df1["Property Crime"]= df1["Property Crime"].astype (int)

df1["Burglary"]= df1["Burglary"].astype (int)

df1["Larceny-theft"]= df1["Larceny-theft"].astype (int)

df1["Motor vehicle theft"]= df1["Motor vehicle theft"].astype (int)

# preview to ensure everything looks good df1

# create a plot of crime over time to see which types of crime are commited most often df1.plot(kind='line',x="Year",y=["Violent Crime","Murder and nonnegligent manslaughter","Rape Legacy def", "Robbery","Aggrevated Assult","Property Crime","Burglary", "Larceny-theft", "Motor vehicle theft"]) plt.xlabel("Year") plt.ylabel("Crime") plt.suptitle('Crime rate from 1997 to 2016') plt.legend( ncol=1, loc='best');

This chart shows that the most common crimes commited from 1997-2016 were property crimes and Larceny-Theft

# create a chart of summary statistics for each column in the data frame df1.describe()

Using Seaborn to Visualize a pandas data frame

To futher look into the data set as well as learn how to plot with seaborn we will take a closer look at the top two types of crime commited from 1997 to 2016

# set the x value to be years and the y value to be property crime df1['x'] = df1["Year"] df1['y'] = df1["Property Crime"]

# using seaborn scatter plot X vs. Y. sns.lmplot('x', 'y', data=df1, fit_reg=False)

According to this chart it appears that property crime decreased from 1997 to 2016

# create a density plot using seaborn sns.kdeplot(df1.y)

A Density Plot visualises the distribution of data over a continuous interval or time period. The peaks of a Density Plot help display where values are concentrated over the interval. This plot shows that overtime the peak number of property crime was between 1.0 and 1.1 le7

# this shows another version of the density plot with the histogram in the background sns.distplot(df1.y)

Comparing the top 2 types of crime from 1997 to 2016

# rename the variables x being larceny theft and y being property crime df1['x'] = df1["Larceny-theft"] df1['y'] = df1["Property Crime"]

# using seaborn scatter plot X vs. Y. sns.lmplot('x', 'y', data=df1, fit_reg=False)

This graph shows a strong positive corelation between Larceny theft and property crimes. This makes sense because larceny theft is included in the the defination of property crimes and is the largest property crime commitited according to this FBI data set

Footnotes:

Populations are U.S. Census Bureau provisional estimates as of July 1 for each year except 2000 and 2010, which are decennial census counts.
The violent crime figures include the offenses of murder, rape (legacy definition), robbery, and aggravated assault.
The figures for the offense of rape were estimated using the revised Uniform Crime Reporting Program's (UCR) definition of rape.
The murder and nonnegligent homicides that occurred as a result of the events of September 11, 2001, are not included in this table.
Although arson data are included in the trend and clearance tables, sufficient data are not available to estimate totals for this offense. Therefore, no arson data are published in this table.

Crime in the United States

by Volume and Rate per 100,000 Inhabitants, 1997–2016

This chart shows that the most common crimes commited from 1997-2016 were property crimes and Larceny-Theft

Using Seaborn to Visualize a pandas data frame

To futher look into the data set as well as learn how to plot with seaborn we will take a closer look at the top two types of crime commited from 1997 to 2016

According to this chart it appears that property crime decreased from 1997 to 2016

Comparing the top 2 types of crime from 1997 to 2016

This graph shows a strong positive corelation between Larceny theft and property crimes. This makes sense because larceny theft is included in the the defination of property crimes and is the largest property crime commitited according to this FBI data set

Footnotes:

Data Retreived from: https://ucr.fbi.gov/crime-in-the-u.s/2016/crime-in-the-u.s.-2016/topic-pages/tables/table-1

Recommended on Deepnote

Stock market analysis

10 ways to create NumPy arrays

Wide Residual Networks