Index > research
What
is the 3-i intelligence model?
What is intelligence-led policing?
What
is Aoristic analysis?
What is Sleipnir?
What is the Modifiable Areal Unit Problem?
What is the 'Vicinity' approach to
spatial analysis?
What
is Hotspot analysis?
What is the ecological fallacy?
How can I display an aoristic change
in time graphically?
How accurate is geocoding?
How successful are others at
geocoding? A hit rate survey
What is the 3-i intelligence model?
The 3-i model is a model that simply shows the role of the intelligence/crime analyst in an intelligence-led policing environment. The aim of the model is to show that crime reduction will not happen unless analysts interpret the criminal environment and influence the thinking of decision makers who in turn have a positive impact on the criminal environment. Diagram, more details and downloads...
What is intelligence-led policing?
Intelligence-led policing was originally articulated as a law enforcement operational strategy that sought to reduce crime through the combined use of crime analysis and criminal intelligence in order to determine crime reduction tactics that concentrate on the enforcement and prevention of criminal offender activity, with a focus on active and recidivist offenders. Since then, the defintion has evolved. Tell me more...
One of the problems with analysing crime is achieving temporal accuracy when the exact time and date of many crimes is unknown. This can occur if a family goes on holiday for two weeks and return to find their home burgled. Not knowing the exact date can be a problem for analysing crime accurately, but an aoristic search can help, and can also reveal different patterns not seen in other types of analyses. Tell me more...
The Sleipnir technique provides intelligence analysts working on organized crime groups with a comprehensive and transparent method to assist in developing and presenting recommendations and supporting intelligence in a concise manner. Research conducted at the 2011 IALEIA/LEIU conference may help to understand the wider possibilities of the relative importance of the attributes as initially determined through the results of voting by 16 Canadian focus groups. Tell me more...
What is the Modifiable Areal Unit Problem?
When data are aggregated to boundaries such an census tracts and block groups, it is possible that the result of any analysis is determined to some extent by the shape of the boundaries used. Aggregate data to police beats and you will get a different outcome to your analysis than if you had aggregated the data to blocks. Is this a serious problem? Tell me more...
What is the 'Vicinity' approach to spatial analysis?
If crime patterns are to be compared against socio-economic data (such as census data), the accurate geocoding of the crime locations is essential. This is however rarely the case and any errors in the crime point location can end up with the point being located in the wrong polygon and attributed to the wrong socio-economic value. A solution to this is to use a spatial weighting average of the values from the immediate vicinity of the crime location, reducing the potential for errors. Tell me more...
Hotspot analysis is an important part of crime analysis. Hotspots can be used to aggregate high volumes of crime and simplify the display of large amounts of crime, and can also be used to share sensitive information with other agencies without compromising the privacy of the data. The accurate identification of a crime hotspot is an essential part of the work, and this is one area where local statistical analysis can be useful. Tell me more...
What is the ecological fallacy?
The ecological fallacy is a situation that can occur when a researcher or analyst makes an inference about an individual based on aggregate data for a group. It can have many variations. For example, if a researcher identified that an area had a high level of burglary then might be tempted to assume that many burglars live in the area. The main problem however is when researchers make assumptions about an individual who lives in the area based on aggregate data about the region. Tell me more...
How can I display an aoristic change in time graphically?
Mapping changes in time are difficult enough as it is, without also considering the aoristic complications. Mapping aoristic changes is similar to adding a temporal component to a map. There are two main solutions. To map the change in crime distribution with a bar graph attached to the crime hotspot, or to use animation software to generate a movie with time progressing as the movie progresses. Tell me more...
Crime mapping starts usually with geocoding - the process of turning an address or crime location in a police recorded crime database into an electronic point on a map. Special software 'geocoding engines' within GIS perform this task, but sometimes the accuracy is variable. The accuracy then goes on to impact on the accuracy of hotspot algorithms and other forms of spatial analysis that are conducted afterwards. So how accurate is geocoding? What a GIs places a dot on a map, how close it is to the actual address? Well, the mean of a 20,000 address test set was 20-30 metres (67 to 100 feet) depending on geocoding setting. Tell me more...
How successful are others at geocoding? A hit rate survey
On 20th May 2002 I posted a short message on the NIJ Crime Mapping Research Center (now the Mapping and Analysis for Public Safety [MAPS] program) list server, asking list members for their average success rates with geocoding. E-mails received ranged from a single word answer, to short essays (one person wrote 1,300 words on the issue). By 28th May 2002, I had received replies from 39 individuals, describing their geocoding experiences with numerous different agencies. In the end I had usable responses with regard to 43 different agencies. The mean average geocoding hit rate was 87.5%, with a standard deviation of 14.1%. The lowest was 41%, while the highest was 99.7%. Slightly more than two thirds of the responses were 90% or greater. Tell me more...
Frank Morgan coined the term 'near repeat' in an article in 2000. He noted, in a study of burglaries in the Australian city of Perth, that several one-time burglaries occured near to, and shortly after, a burglary at a repeat victimization location. Further studies in Australia by Mike Townsley and colleagues, and Shane Johnson and Kate Bowers in the UK, cemented the realization that the risk
of repeat burglary victimization is transmitted from a burgled location to nearby locations for a
limited amount of time. The near repeat phenomenon now appears to be relevant to a wide variety of crimes, including burglary, shootings, vehicle theft, and even the distribution of IED attacks in Baghdad. There is an advanced simulation modelling approach to determining the spatial and temporal characteristics of the near repeat phenomenon, and helpfully there is now a free software product to assist with the analysis. Tell me more...
