Data Points

A number of studies have shown that our perception of crime is very often much different to reality. In a 2016 US survey⁽¹⁾, 57 percent of respondents believed that the level of national crime had become worse since 2008, when in fact violent and property crime declined by double-digit percentages according to the FBI Uniform Crime Reports (UCR).

The same holds true across The Pond. Data released by the Office for National Statistics⁽²⁾ shows that many more people in England and Wales worry about being a victim of crime than will actually experience it. In 2016, 0.3 percent of the population were victims of robbery, but 9 percent of survey respondents (i.e. 30 times higher) were worried that they would be a victim in the coming year.

The crime data landscape is fragmented, which does not help our understanding of crime rates and spatial incidence. Although the FBI UCR provide standardized data at an aggregated geographical scale, more granular incident-level data are made available via various police agencies e.g. Sheriff's Offices. However, each reports crimes in a different manner, which makes comparison difficult - until now.

We collected these incident level records from disparate agencies across the US, standardized the data and combined them with UCR data using a sustainable and repeatable automated process to build our new and improved version of CrimeIndex.

CrimeIndex allows users to compare local-level crime (block-group) against both state and national rates, providing a more accurate and contextual view of prevalence, not only for all crime in total, but across the following crime categories:

 

• Violent Crime
• Criminal Homicide
• Rape
• Robbery
• Aggravated Assault
• Property Crime
• Burglary
• Larceny/Theft
• Motor Vehicle Theft
• Arson

 

Users can now compare their local crime rate with state or national levels.

There are clear use cases for this product – insurance, real estate and so on– but when combined with additional demographic data sets, they provide an additional dimension to population insights.

What use cases can you see CrimeIndex helping with?

 

(1) Pew Research Center Survey Conducted Oct 25 – Nov 8, 2016
(2) Office for National Statistics "Trends in prevalence and worry about robbery, Crime Survey for England and Wales, years ending March 2009 to March 2016"

------------------------------
Andy Peloe
Senior Product Manager - World Demographics
Pitney Bowes
------------------------------

Hi Andy - I could see CrimeIndex helping in 2 use cases: 

1. Routing engines for driving or walking routes could use it to determine which blockgroups to avoid routing through (the routing data provider could downgrade roads in those blockgroups for their through-routing) – I don't think the routing data providers today make use of CrimeIndex as an input.
2. Delivery companies could use this as an indicator of an address where they need to use 2 delivery people in the truck (perhaps to have 1 driver and 1 assistant in the truck so packages are not stolen), or that they cannot leave packages unsecured at a home.

Both are unfortunate to have to think about but perhaps prudent.

Thanks

Tom

------------------------------
Tom Gilligan
Pitney Bowes Software, Inc.
White River Junction, VT, USA
------------------------------

Original Message:
Sent: 07-18-2019 06:36
From: Andrew Peloe
Subject: It's Criminal

A number of studies have shown that our perception of crime is very often much different to reality. In a 2016 US survey⁽¹⁾, 57 percent of respondents believed that the level of national crime had become worse since 2008, when in fact violent and property crime declined by double-digit percentages according to the FBI Uniform Crime Reports (UCR).

The same holds true across The Pond. Data released by the Office for National Statistics⁽²⁾ shows that many more people in England and Wales worry about being a victim of crime than will actually experience it. In 2016, 0.3 percent of the population were victims of robbery, but 9 percent of survey respondents (i.e. 30 times higher) were worried that they would be a victim in the coming year.

The crime data landscape is fragmented, which does not help our understanding of crime rates and spatial incidence. Although the FBI UCR provide standardized data at an aggregated geographical scale, more granular incident-level data are made available via various police agencies e.g. Sheriff's Offices. However, each reports crimes in a different manner, which makes comparison difficult - until now.

We collected these incident level records from disparate agencies across the US, standardized the data and combined them with UCR data using a sustainable and repeatable automated process to build our new and improved version of CrimeIndex.

CrimeIndex allows users to compare local-level crime (block-group) against both state and national rates, providing a more accurate and contextual view of prevalence, not only for all crime in total, but across the following crime categories:

 

• Violent Crime
• Criminal Homicide
• Rape
• Robbery
• Aggravated Assault
• Property Crime
• Burglary
• Larceny/Theft
• Motor Vehicle Theft
• Arson

 

Users can now compare their local crime rate with state or national levels.

There are clear use cases for this product – insurance, real estate and so on– but when combined with additional demographic data sets, they provide an additional dimension to population insights.

What use cases can you see CrimeIndex helping with?

 

(1) Pew Research Center Survey Conducted Oct 25 – Nov 8, 2016
(2) Office for National Statistics "Trends in prevalence and worry about robbery, Crime Survey for England and Wales, years ending March 2009 to March 2016"

------------------------------
Andy Peloe
Senior Product Manager - World Demographics
Pitney Bowes
------------------------------