People first policing
Constabulary cleans up its data and gains new intelligence
Gloucestershire Constabulary uses SAS to ensure data quality and achieve new levels of business intelligence and reporting, helping improve accuracy and operational efficiency to enhance intelligence-led policing
The UK’s 43 police forces are responsible for managing their own data resources – no easy task in a mobile digital world, where data volumes proliferate and, when it comes to information, criminals can be economical with the truth. Gloucestershire Constabulary, which employs 1,300 officers, 800 civilian staff and 170 Special Constables, serves a population of 560,000. Effective community policing is at the heart of operations, and that depends on having high quality data and up-to-date intelligence. Indeed, reliable systems to provide such insights enable resources to be used as effectively as possible, to improve performance, to report accurately to the Home Office, and ultimately share information with other forces.
“Gloucestershire was one of the first forces to recognise the data quality issue,” says Reg Barnard, IS Development Manager. “This has since become a far bigger national issue, particularly with the Police National Database (PND) from 2011. I think we’ve stayed at the forefront of dealing with data quality in support of national objectives.” These include the IMPACT programme to improve the police’s ability to manage and share information and the Management of Police Information (MoPI) national standard; both driving even greater demands for higher quality data, reliable intelligence and information sharing with other forces and agencies following the Bichard Inquiry. SAS is helping Gloucestershire to ensure the quality and accuracy of data, with the insights gained informing management, governance and performance.
We wanted a solution that would not only enable us to address data quality but also business intelligence. Good policing is about quality intelligence: making a link between crime and offender.
Reg Barnard
IS Development Manager
Data quality in the dock
In 2002, Gloucestershire Constabulary began its Vision 5 programme to “promote effective and efficient services.” Along with Freedom of Information requirements, this was the key catalyst to invest in improved data quality. “Most organisations face data quality issues but in the police the usual problems like bad spelling and mis-keying can be magnified because many of the people we deal with are less than honest,” says Barnard. Criminals give all kinds of false information that can't be easily verified, with incompatible legacy systems for different purposes complicating the situation. SAS recommended the force deploy SAS® Enterprise Data Integration Server. “We wanted a solution that would not only enable us to address data quality but also lay the foundations for business intelligence," Barnard says. “Good policing is about quality intelligence: making a link between crime and offender.”
Nick Churchill, a former detective and now the force’s Database Administrator, used SAS to integrate three legacy systems: Unity, the core system covering crime and custody; VPFPO for vehicle parking and fixed penalty offences; and the domestic violence database with details of victims and offenders. “We didn’t make it easy for SAS,” Churchill says. “All three systems used incompatible software: Oracle, Ingres and SQL.” Data from source systems is subjected to SAS data quality procedures to profile, cleanse and standardise. Users can adjust sensitivity on matching fields, like name or address, to identify the biggest problems first then work downwards. This means you can quickly pick up obvious alternative spellings but also less obvious matches. “You’d have a tough time finding such mistakes with any other software, let alone manually,” Churchill says. Questionable records are routed to the owner for audit and correction. “SAS is very powerful for clustering near-matches to help us correct and update data – so it saves a huge amount of time and resources,” Churchill adds. “And it’s really easy to use, considering how powerful it is. You don't need specialist skills. In no time you can produce reports that both impress and are valuable to senior managers.”
“With poor quality data we wasted time trying to find out things we already knew,” Barnard says. “SAS put us straight on track to achieve the operational efficiency we wanted and, in practical terms, this can mean ensuring a police officer arrives at the right address when trying to apprehend a criminal.”
Putting people first
Barnard continues, “Key challenges we face now include the requirements of our People First initiative, which is part of the Citizen Focus approach.” High quality data and improved intelligence through analytics mean the force can operate more effectively and deliver against such requirements. At the same time, the force is undergoing big changes, moving from three divisions to six policing areas. “Performance, in terms of the detention statistics, are no longer the ‘be all and end all’ in terms of measuring effectiveness,” Barnard says. “We are working to focus even more on community policing and engagement.” Ongoing policy changes need to be reflected in the information systems required.
Improving detection rates
“In terms of business intelligence, we’ve moved to the latest version of SAS, and are moving SAS towards being an enterprise decision-support system for the organisation,” Barnard says. “We want to provide valuable insights so people can make good decisions, as well as know how well we’re performing, including an understanding of how the community views us and how confident people feel.” The SAS® Professional Services team worked with the Constabulary on the latest SAS upgrade. Barnard continues, “SAS is used to extract data in support of the PND (Police National Database), it helps with MoPI by improving data quality, and supports our RRD (Review, Retain and Delete) policy. On the intelligence side, SAS is supporting the force as it goes through the reorganisation, measuring the metrics we’re being judged by, and creating that enterprise-wide decision support system to support not only senior offices at the top but people throughout the force.” He adds, “We’re also now providing statistical returns to the Home Office directly thorough SAS, providing further efficiency gains by having just the one system and having a single version of the truth.”
Recent results have borne out Gloucestershire’s approach. When Her Majesty’s Inspectorate of Constabulary assessed all 43 police forces in 2010, it highlighted how Gloucestershire has reduced overall crime in the last two years, with notable success in violence and robbery, and achieved excellent detection rates. It also indicated that public confidence is improving, a finding backed by the British Crime Survey that showed an increase from 45.8 percent to 51.5 percent in the number of Gloucestershire people confident their police and council were dealing with anti-social behaviour and crime.
Barnard concludes, “The value of data is now becoming recognised to a much greater extent, particularly in light of the national initiatives like IMPACT. And all public organisations are under financial pressure to save money. We can either provide less, or the same and hopefully more with the limited resources we have. The only way we can do that is by improving efficiency. A key way of achieving that is by making better use of our information, to measure and improve our performance – and SAS plays a key part in enabling that.”
Challenge
- Improve data quality, remove inaccuracies and provide new levels of insight to support improved law enforcement.
- Continue providing performance information and intelligence while adapting to new requirements and priorities.
Solution
Benefits
More consistent and accurate data to drive more intelligence-led and targeted policing, supporting overall management and performance improvements as well as providing new insights and supporting national initiatives including the IMPACT programme, Management of Police Information and the Police National Database.