SafeStats: our data

Transport for London provide us with a monthly dataset covering all crime and disorder-related incidents that have occurred on, or are related to, London's buses.

• The data is non-aggregated and therefore contains the data for each of the incidents that make up the overall count
• The data goes back to April 2001 and is coded to the Easting/Northing point level (and therefore available for all geographic levels)
• Other useful information available includes the bus and its route

Categories include:

Anti-social behaviour: incidents of anti-social behaviour both on and off the bus; youth-related; alcohol-related

Assaults: assaults on staff and passengers, on and off the bus

Criminal damage: incidents of graffiti to buses and other buildings; objects being thrown at buses; other damage to vehicles or buildings (such as bus stops and shelters)

Fraud: broken down into incidents discovered by members of TfL staff and those by Revenue Protection Inspectors

Theft and handling: incidents of pickpocketing, thefts from staff and thefts from passengers

Robbery: incidents of personal robbery, business robbery and robbery of staff and passengers

The London Fire Brigade provide us with a monthly dataset covering the details of each of the incidents that they’re called out to.

• The data is non-aggregated and so contains the data for each of the incidents that make up the overall count. A datasets aggregated to Borough level is available on the London Datastore.
• The data is held for as far back as April 2001 and is coded to the Easting/Northing point level of the location attended (and therefore available for all geographic levels)

Categories include (but not limited to):

Hoax Calls: all call-outs determined as hoax or malicious, representing a form of anti-social behaviour

Arson: all call-outs relating to all deliberate fires, representing a form of anti-social behaviour

Deliberate house fires: all call-outs to the London Fire Brigade relating to house fires started deliberately, representing a form of anti-social behaviour

The British Transport Police provide us with a monthly dataset covering crime and disorder offences and incidents that take place in London’s National Rail and Underground network London (including DLR).

• The data is non-aggregated (so contains data for each of the incidents that make up the overall count)
• The data goes back to June 2001 and is coded to the nearest station (so is also aggregated up to borough level)

The categories are:

Anti-social behaviour: begging, drunken behaviour, stone throwing and trespassing

Assaults: offences such as racially-aggravated assault

Burglary: offences of burglary from transport buildings such as warehouses, booking offices, stores and workshops

Criminal damage: offences such as graffiti

Drugs: offences such as cannabis possession and class A offences

Theft: areas such as theft of business property, pedal cycles, personal property and vehicles

Robbery: offences such as personal robberies

Sex offences: offences such as exposure, sexual assault, rape and other public indecencies

Weapon possession: offences such as possession of an offensive weapon

As part of the Information Sharing to Tackle Violence (ISTV) project funded by the Mayors Office for Policing and Crime (MOPAC), hospitals are encouraged to record information about the location of violent assaults (where these take place in public). Previous research has shown that sharing this information along with anonymised data about the time/date and type of injury and method of attack can be used by the police and public safety bodies to enrich ongoing preventative work and identify new priorities.

SafeStats was chosen by the Home Office to source, store, process and visualise data from over 25 hospitals in London to enable analysis by colleagues in fields from preventative health, emergency medicine and health analytics. By hosting this data within SafeStats, it can be viewed and analysed by users across London alongside over 15 million SafeStats records from the providers listed above.

A unique geocoding process has been developed that takes the freetext incident location information and evaluates its potential for address-matching quality, subsequently placing it on a map at point, street centrepoint, electoral ward, or borough level depending on its quality. This information, as well as other fields relating to the incident are available via the ISTV interface or in raw data form on the SafeStats Query Forum.