• Academics note crime rates in districts of Colombo, Gampaha, Kegalle, Kilinochchi, Anuradhapura, Vavuniya, Hambantota, Kurunegala, Kalutara, Ratnapura, Galle, Kandy, Matara, and Jaffna

A decision-making process that considers where best to target law enforcement-based preventive action with regard to crime by paying more attention to areas that are crime hotspots, which according to the statistics are the districts of Colombo, Gampaha, Kegalle, Kilinochchi, Anuradhapura, Vavuniya, Hambantota, Kurunegala, Kalutara, Ratnapura, Galle, Kandy, Matara, and Jaffna, and which have been deemed unsafe in terms of the prevalence of crime, has the potential to reduce crime.

This recommendation was made in an article on the “Statistical approach of identifying crime hotspots for geographic information system (GIS) mapping in Sri Lanka”, which was authored by M.A.P. Munasingha and N.A.D.N. Napagoda (both attached to the Wayamba University's Applied Sciences Faculty’s Mathematical Sciences Department) and published in the Sri Lankan Journal of Applied Statistics 23(2) in November 2022.

The present wave of crime is a key social problem in Sri Lanka. At present, crimes are well organised and show increasing trends with the rising population and rapid advancement of modern technologies. Crime is a social stimulus and society suffers in many ways as a result. Therefore, taking necessary steps for the reduction of crime in the society is vital to maintain its well-being. Many types of crime are prevalent in society today. Kidnapping, arson, mischief over Rs. 5,000, house-breaking and theft, grievous hurt, being hurt by a knife, homicide and abetting to commit suicide, attempted homicide, rape and incest, riots, robbery, unnatural offences and grave sexual abuse, extortion, and cheating and the misappropriation of trust over Rs. 100,000 are some such crimes. The increase in crime has severely affected the lives of the people. As a result, society is deprived of the opportunity to spend time freely. Therefore, it is timely to formulate programmes to reduce crimes. Crime can be minimised by identifying areas with a high level of crime in order to facilitate the provision of greater security to and in those hotspot areas. 

Munasingha et al.’s research used annual total crime data ranges from 2010 to 2019 in order to generate hotspot maps, and to test their accuracy for predicting where crimes will occur next. The prediction accuracy index (PAI) was used to compare hotspot mapping accuracy based on the types of crime (housebreaking and theft, cheating, robbery, rape, and being hurt by a knife).

Munasingha et al.’s main study focus was to identify the most common types of crime in Sri Lanka and to study their prevalence in districts. 

The increase in crime has severely affected the lives of the people and deprives society of the opportunity to spend its free time. It is timely to formulate programmes to reduce crime. Crime can be minimised by identifying high-crime areas and by providing greater security for those areas. Data can be represented as to whether crime is rising or falling in the targeted areas. This can help criminal justice professionals understand whether their initiatives are successful. Also, map updating is a key resource for crime control.

In this study, the required data was collected from the Department of Police (“Sri Lanka Police, Crime statistics”) and the Department of Census and Statistics. Annual total crime data ranges from 2010 to 2019 were used for further analysis. This dataset contained crime events collected district-wise, with 14 different categories of crime. 

Colombo and Gampaha have the highest crime rates based on both the population and area criteria and have been ranked in the first and second positions, respectively, whereas the Nuwara Eliya District records the lowest based on the population criteria (per 100,000 people). Based on the area criteria, the Mullaitivu District records the lowest.

The safer districts were Polonnaruwa, Ratnapura, Kalutara, Galle, Mannar, Kandy, Matara, Kurunegala, Monaragala, Batticaloa, Matale, Puttalam, Trincomalee, Jaffna, Ampara, Badulla, Mullaitivu, and Nuwara Eliya. The relatively unsafe districts were Colombo, Gampaha, Kegalle, Kilinochchi, Anuradhapura, Vavuniya, and Hambantota.

Based on the country’s total crime rate based on the population criteria, safe districts have their crime rate below the overall total crime rate and unsafe Districts have their crime rate higher than the overall total crime rate. The overall crime rate per 100,000 population in 2012 is 205.8. According to the classification, there are 18 safe and seven unsafe districts in Sri Lanka.

Colombo recorded the highest number of crimes, which is 447.2071, while the count of Gampaha accounts for 359.9929, and the Mannar and Mullaitivu Districts have the lowest number of crimes.

The majority of the crimes committed in Sri Lanka from 2010 to 2019 are related to crimes in which there is housebreaking and theft, and cheating.

According to that result, five types of crime that have a higher impact on society were identified. They are housebreaking and theft, cheating, robbery, being hurt by a knife, and rape. It was concluded that there is very strong evidence to suggest that there is a difference between at least one of the pair of groups.

According to the results, only one pair does not have a significant difference, and that is being hurt by a knife and rape. All other pairs have significant differences.

The results show that there is a very strong positive correlation between robbery and housebreaking and theft, and being hurt by a knife and rape. All types of crime have shown a positive and high correlation for the occurrence of crimes in the consecutive years from 2010 to 2019.

The criminal databases of house breaking, being hurt by a knife and robbery are showing a declining trend and it can be concluded that there is a tendency for those crimes to decrease in the future. However, the crime data point of rape and cheating fluctuates around a straight line, and since they do not represent a clear trend, it is not possible to determine whether such crime will be low or high in the future.

There was however a decrease in the number of crimes in all types of crime over the next few years. A heat map of total crime based on the area criteria shows that crimes are more prevalent in the Western Province. Furthermore, the Kegalle, Galle, Kandy, Matara, and Jaffna districts have a significant number of total crimes per area. Moreover, most of the crimes can be observed in the Colombo and Gampaha Districts and which then spread over to the down South area in decreasing magnitude.

According to the results, the risk of all five types of crime appears to be high in the Kurunegala and Gampaha Districts. The Colombo District appears to be at risk for four types of crime, namely cheating, house breaking, being hurt by a knife and robbery. The Kilinochchi, Mullaitivu, Mannar, Vavuniya, and Trincomalee Districts have the lowest incidence of rape, robbery, and being hurt by a knife. The Kilinochchi, Mullaitivu, Mannar, Matale, and Monaragala districts have the lowest incidence of cheating. The Mullaitivu, Mannar, Vavuniya, and Batticaloa Districts have the lowest incidence of housebreaking.

Robbery has the highest PAI value. There is a 63% probability of robbery occurring in the hotspot areas. The robbery-related crime hotspot map is consistently better in its ability to predict spatial patterns of robbery-related crime than any of the other hotspot maps for types of crime in their ability to predict crimes of their respective type. Rape has the lowest PAI value. There is a 37% probability of rape occurring in the hotspot areas. The hit rates can be used to measure the probability of crime using the resources in the hotspot areas.

Kurunegala, Colombo, Kalutara, Gampaha, and Ratnapura are the hotspot areas of total crimes from 2010 to 2019. Kilinochchi, Mullaitivu, Mannar, Vavuniya, and Trincomalee are the districts with the lowest incidence of total crimes.

Crimes have been disturbances and threats to all Sri Lankans all over the country. All five types of crime have shown a positive and high correlation with each other for the occurrence of crimes and there was a significant difference between the mean of all pairs with the exception of rape and being hurt by a knife. Data on crime involving housebreaking, being hurt by a knife and robbery had a downward linear trend pattern. The other two types of crime trends had fluctuated. The risk of all five types of crime appeared to be high in the Kurunegala and Gampaha districts. The Colombo District appeared to be at risk for four types of crime, namely, cheating, housebreaking, being hurt by a knife, and robbery. Considering the total crimes in Sri Lanka, the area defined as “hot” covered 18.3% of the total area and there was a 52.7% probability of the said crimes being committed in the hotspot areas. The robbery hotspot map was consistently better at predicting where robbery related crime will occur in the future, than the hotspot maps for the other types of crime.