Complementing crime reduction plans

By NORMAN KAMUYA

Over the last couple of years we have witnessed increasing numbers and sophistication on crime. Scholars have made great progress in explaining why some individuals are more likely than others to commit crimes, but not much in explaining why the crime rate in a jurisdiction rises or falls.

A quick look at any police unit website for statistics on crime figures will leave you formulating a myriad of hypothesis.

The law enforcement community is increasingly inundated with information while simultaneously facing reduced budgets. Waging war on terrorism and new forms of crime has created the additional burdens of requiring compilation, analysis and integration of even more information. No analyst or team of analysts could swiftly and accurately sift through all the data to uncover patterns that might indicate, for example, how to best deploy the police force to prevent crime or determine whether or not a threat is real.

Any crime prevention department or unit inevitably needs a solution that could identify these trends and patterns both expediently and inexpensively. For Example every time someone is arrested or convicted of a crime, they are entered into the Criminal History Records Repository. This "rap sheet" – commonly known as OB, contains all the necessary information, including identification data (date of birth, name, home district among others), charges, and the court dispositions associated with every charge. Sooner than later this information builds up at exponential rates.

Paper systems

More often than not, a law enforcement unit running legacy or paper systems will encounter problems trying to mine any useful data. There could be thousands of individuals included in the repository associated with information on thousands of criminal charges and a similar number of arrest episodes.

Using Data analysis to generate any insight requires finding the ‘right tool for the job’. Off-course crime recording systems currently exist, but more often that not they are not specifically designed to analyse data. More comprehensive data analysis can be performed using tools such as PASW (Formerly SPSS - Statistical Package for the Social Sciences).

It may be worth investing in it or an equivalent application, although statistical packages more easily allow some forms of data interrogation and visual display. PASW predictive analytics software represents a revolution in the ability to access previously unobtainable data and pull meaning and value from it.

Several methods for measuring crime exist, including household surveys, hospital or insurance records and compilations by law enforcement agencies. Public surveys are sometimes conducted to estimate the amount of crime not reported to police. Such surveys are usually more reliable for assessing trends and rarely encompass all crime and necessary to procure statistics useful for local crime prevention.

The writer is a predictive analytics expert. Email: Nkamuya@spss-eastafrica.com