What Is Geographic Profiling?

Crime is not randomly distributed and offenses are more likely to occur in certain places and at certain times. The apprehension process can be significantly accelerated if geographic profiling is used to organize an abundance of information via geographical links (Ramsland n.d.). This technique is an effective tool in ascertaining a suspect’s residence and/or place of employment because it is conducted in a scientific manner on the basis of well-established psychological principles (Sammons n.d.). The geographical patterns in data can be analyzed using the following principle elements:

“mobility, mental maps, locality demographics and distance” (nij.gov 2006). “A geographical analysis highlights the crime location, any physical boundaries that were present (that might not otherwise be noticed), and the types of roads and highways that come into both the abduction and body dump sites. Hypotheses are developed which can be tested against evidence and modified or rejected as the evidence dictates.” (Sammons n.d.) Objective measurements are frequently used as a component of geographic profiling in order to pinpoint precisely the locus of criminal activity. (Sammons n.d.) One of the methods used to obtained predictions is a geographic profiling system. The area most likely to contain the offender’s home is computed by the system using algorithms and indicated in the probability surfaces. The probability of an offender residing at a particular location generally decreases with increasing distance from an offense, so these algorithms are referred to as distance decay functions (Bennell, Corey & Keyton 2007). According to the available research the accuracy of geographic profiling systems reduces the overall area that police have to search by up to 90% (Bennell, Corey & Keyton 2007). This method can assist police through the use of the most effective tools to target crime prevention resources in the appropriate areas. When we understand the link between offenses and location it is beneficial in our understanding of why certain locations attract more crime than others. Investigative efforts can be targeted when we are able to predict characteristics of the offender(s) responsible for a series of crimes. The use of geographic profiling also facilitates in law enforcement’s ability to prioritize suspects by “an individual’s most likely place of residence, their knowledge of a particular area and to determine which offenses are linked and which are not, so that they can determine whether or not a certain crime may have been carried out by the same offender(s)” (Sammons n.d.). GIS capability, database management, powerful visualization tools and analytic engine are incorporated together and compose what is known as computerized geographic profiling (Harris 1999). The use of geographic profiling can be deployed through a number of investigative strategies. Some examples may include: “department of motor vehicle searches, mass DNA screening prioritization, canvasses and searches, zip code prioritization, information request mailouts, patrol saturation and surveillance, address-based searches of police record systems and suspect and tip prioritization” (Harris 1999). In order for computerized geographic profiling to be most effective crime locations should first be broken down by type such as: body dump sites for a murder, victim encounter or murder. Each of these crime locations should be entered into the system by latitude/longitude, digitization or by entering the address (Harris 1999). “Scenarios are created and weighted based on crime locations and the use of theoretical and methodological principles (Harris 1999). A probability chart known as a z-score histogram is created from the suspect addressed and is evaluated according to their “hit” percentage on the chart

Mental Maps

A mental map, is a cognitive image of one’s surroundings that is developed through experiences, travel routes, reference points and centers of activity. This is another significant factor in geographical profiling. (nij.gov 2006) Each person has their own mental map that contains the places he or she feels safe and takes for granted regardless of offender status. These mental maps do not accurately represent reality, but rather the perspective and experience of the individual. (Sammons n.d.) An offender’s mental map can be reconstructed and interpreted with the assistance of the familiarity of the location in relation to his or her awareness of space (Allaire, Beauregard, Leclerc et al 2007). Inferences about a criminal can be made about them by reviewing the distribution of their offenses since they were likely influenced by that individual’s mental map of an area (Sammons n.d.) Crime site selection is influenced by these mental maps as the offender must first be aware of the site before a target can be victimized. (Allaire, Beauregard, Leclerc et al 2007). Many times an offender’s mental map will change as he becomes more confident which in turn will increase his range of criminal activity. A criminal may be either geographically stable or transient or may start as geographically stable and become transient with time and number of criminal acts. (nij.gov 2006) An offender is influenced by several factors which determine his tendency toward stability or mobility such as: “their experience with travel, means for getting places, sense of personal security, and predatory motivations.” (nij.gov 2006)

Routine Activity Theory and Circle Theory

Routine Activity Theory otherwise known as RAT is another approach that is taken into consideration by geographical profiling. Three circumstances must coincide according to the RAT principle: “a motivated offender, a suitable victim and the absence of a capable guardian” (Sammons n.d.). RAT assumes “that in any area there are a certain number of people motivated to commit crime” (Sammons n.d.) This theory sees offenses as just another activity that a person might do on a regular basis without any attempt to explain the reason why. Another belief of this theory is that the routine activity of victims is important because people tend to stick with familiar territory. Clues about where an offender lives can be provided by an analysis of all the crime scenes (Ramsland n.d.). “There is a difference between perceived distance and actual distance and certain components influence how this disparity can affect the commission of a crime. The perception of distance varies from one person to the next and can be influenced by any of the following: familiarity with a specific region, types of roads, availability of transportation and number of barriers such as bridges or state boundaries” (nij.gov 2006) Since most people’s activities are confined to a few fairly limited area “where they work, where they live, where they socialize” a person’s offenses will be limited in that same geographical area according to RAT (Sammons n.d.). A successful example of geographical profiling using RAT involves a computerized system called Dragnet. The location of the offenses allows Dragnet to predict where an offender is likely to lived based on that information.

Dragnet created a map that suggested probabilities that the offender responsible was based in different regions according to the information it received on a number of linked rapes in Las Vegas.

The focus of the investigation was able to be narrowed to a single apartment block as a result of the investigating officer’s knowledge of the local area and the offender in turn was subsequently arrested

(Sammons n.d.). Canter and Larkin (1193) came up with the circle theory of environmental range which proposed that the majority of the time, if a circle is drawn that encompasses all of a series of linked crimes, the offender will be based somewhere within the circle. (Sammons n.d.) “There is a fair amount of support for this view. Godwin and Canter (1997) found that 85 per cent of the offenders they studied lived inside the circle encompassing their offenses” (Sammons n.d.). Serial rapes and arson attacks in Australia were reviewed by Koscis and Irwin (1997) which confirmed the circle theory. The only exception appeared to be burglary since burglars only lived in the circle defined by the offenses about half the time. A review the offense locations of 53 serial murderers in Germany was conducted by Snook et al (2005).The results of his study discovered that the killer lived within six miles of where the bodies were found in 63% of the cases. “Younger offenders travel shorter distances and killers with higher IQ travel further.” Information about the dispersal of offenses may indicate some general characteristics of the offender responsible because it has been found that experience and intelligence influence killers’ attempts to disguise their crimes (Sammons n.d.)

Importance of Geographic Profiling and how it can be most effective

Geographical profiling has a particular importance in the United States because there are many different law enforcement agencies which share little data between them, so this method enables a connection between crimes to be established that might not have otherwise been linked (Sammons n.d.). In order to construct an accurate geographical profile all of the following elements need to be included in the assessment: “computerized analysis, study of area maps, analysis of neighborhood demographics for both the abduction site and body dump site, examination of the crime scenes, complete familiarity with the case file and interviews with investigators and witnesses.” (Ramsland n.d.) In order for geographical profiling to be effective the following should be taken into consideration: it requires accurate data on the offenses that have been committed in an area, police data on crime is likely underreported, so the data used to generate the crime maps is likely to be incomplete as a result, since police will have a vast amount of data available it may be difficult for them to determine what should be left out when they attempt to construct a crime map and additional problems can arise from inconsistencies in how the locations of the crimes may be recorded by the police (Sammonds n.d.). Information about five or more crime locations needs to be available in order for geographic profilers to be most effective according to a 2005 study conducted by Rossmo (Bennell, Corey & Keyton, et al. 2007). He also claimed that as additional crime locations were incorporated into the prediction that there would be an even further increase in accuracy. (Bennell, Corey & Keyton, et al. 2007). This assumption is not supported by the analysis of CrimeStat’s performance across maps with varying numbers of crimes. Participant performance is shown to increase from three to five crimes, but the increase vanishes when making predictions from seven crimes according to the findings in this study (Bennell, Corey & Keyton, et al. 2007). .

A localized improvement in performance was found with five crime locations, but it is not evidence of a large positive correlation between the number of crime locations and predictive accuracy. Accuracy does not increase as more crime locations are added to the information used to make a prediction (Bennell, Corey & Keyton, et al. 2007).

Generalizability

When a crime occurs in a specific location, the area surrounding that location may experience an increased risk of a similar crime occurring for a distinct period of time which is known as a “near repeat” phenomenon. Trends in spatiotemporal proximity which involve both space and time are referred to as near repeat crimes (Cook, Nobles & Ward et al. 2011). A rival explanation for the observation of near repeat crime is spree crime which is a pattern characterized by high frequency of criminal activity involving the same offender across a short time span such as hours or days. This phenomenon can be generalized across multiple crime types and effect various temporal bands on near repeat pattern. “The temporal dimension of near repeat crime may vary across different types of offenses” (Cook, Nobles & Ward et al. 2011). The majority of crime types have a component of repeat victimization except for manslaughter or murder. A disproportionate amount of crime has been shown by research to occur in specific geographic areas known as “hot spots.” These areas have a higher than average risk of victimization and vary in size, but are typically blocks or street segments. The identification of hot spots allows law enforcement to understand where crimes are most likely to take place, so they can more effectively target its resources. (Cook, Nobles & Ward et al. 2011). Repeat victimization is different than hot spots which include multiple targets and crime types and they are not concentrated in a specific amount of time even if they are concentrated in space. An overall consensus was found that homes that are burglarized have a higher likelihood of being burglarized again in the future. In the month following an original burglary Johnson et al. (1997) found the risk of re-victimization to be elevated. Another study conducted by Morgan (2001) also found that re-victimization was most likely to occur in the month following a burglary, however the data reflected that areas of higher overall burglary rates had more stability in their likelihood of repeat victimization (Cook, Nobles & Ward et al. 2011). A near repeat pattern for burglaries that extended at least 200m for two weeks existed in all of the study locations examined by Johnson et al (2007). This was the result in all ten areas within the five counties despite the fact that the patterns differed in the geographic areas. A pattern of increased risk of victimization in the surrounding area of a residence after a burglary was discovered by Morgan. Morgan referred to these incidents as “near repeats.” In the week following a burglary the homes in the surrounding area were particularly at an increased risk of burglary (Cook, Nobles & Ward et al. 2011). They also determined that affluent areas as opposed to deprived areas appeared to reflect more evident space-time clustering. “Townsley et al. (2003) found that there was an increase in burglary incidents within 200 m (approximately 650 feet) and 2 months of an original burglary. Similarly, Johnson and Bowers (2004) found increased risk of burglaries for dwellings within 400 feet of a previously burgled home for 1 to 2 months following the incident, especially on the same side of the road” (Cook, Nobles & Ward et al. 2011). The overall burglary rate can be reduced by 25% if repeat victimization can be prevented and is also a highly important crime reduction strategy. A similar result to burglary was found when Johnson et al. (2009) looked at the spatiotemporal relationship of theft from motor vehicles. Over a two week period evidence of near repeat crimes was found occurring within 800m of the original incident. Ratcliffe and Rengert (2008) analyzed the spatial and temporal distributions of shooting incidents in Philadelphia to explore the near repeat phenomenon for traditional violent crime (Cook, Nobles & Ward et al. 2011). The analysis for this crime type was partially guided by the theoretical information provided by Ratcliffe and Rengert. This study discovered there is a significantly increased likelihood of another shooting within one block of the initial shooting for two weeks after the incident. It cannot be determined whether a near repeat pattern for shooting is common to all geographic areas or whether characteristics specific to the study location influenced the pattern since this was the first study to examine shootings (Cook, Nobles & Ward et al. 2011). A global near repeat phenomenon for shootings may exist if the Philadelphia pattern is exhibited in different geographical areas, but if they did not find similar results then individual locations would need to identify their own unique near repeat patterns if they existed at all. The near repeat pattern for auto theft spans a greater spatial distance than the other crimes which is likely attributed to the goal of stealing specific vehicles, but patterns of offending will vary according to motivation. Any near repeat pattern for robbery occurs with small spatial and temporal bands and exhibits a small, doughnut like spatial pattern as a result of the spontaneity of robbery (Cook, Nobles & Ward et al. 2011). “Prior research has shown robbery to cluster temporally within 1 to 2 days, and spatially, close to the original incident, but not within the immediate 500m” (Cook, Nobles & Ward et al. 2011). The spatial similarity of near repeat patterns for robbery and auto theft regardless of crime type illustrates that offenders committing different crimes may still share a comparable decision making process. Offenders may still work within distinct reference areas regardless of crime type which is the basis for geographic profiling and offender identification through circle theory. Repeat offenders select their targets in familiar areas that tend to be closer to the offender’s residence (Cook, Nobles & Ward et al. 2011). The result is that geographic patterns can be found in crimes that are linked to the same perpetrator, which can then be used to identify the probable location of the offender’s residence. Patterns in both near repeat crimes and spree offending can likely be attributed to the same offender. “Bowers and Johnson (2004) found that near repeat burglaries exhibited the same modus operandi as the original crimes, indicating that near repeat burglaries were likely being committed by the same offenders or groups of offenders. Bernasco (2008) found that same-offender involvement is directly tied to spatial and temporal distances between burglaries. Additionally, Johnson et al. (2009) found that crimes occurring closest to one another in space and time were most likely to be attributed to the same offender” (Cook, Nobles & Ward et al. 2011). A successful first offense increases the likelihood of repeat offending in the general area although the same exact location may not be re-victimized.

Residential Proximity

Research has consistently shown that the number of criminal offenses that an offender commits decreases as distance from an offender’s residence increases with the exception of violent and/or sex offenders (Donnay, Duwe & Tewksbury 2008). The determination of offense location was influenced more by characteristics of events and relationships rather than characteristics of victims and opportunities (Donnay, Duwe & Tewksbury 2008).Offenses occured an average of more than three miles from offender’s homes according to data from five hundred sixty five rapes committed by serial rapists. Of this same population 86% marauded outward into an area of an average of 180 square miles rather than offending against victims that lived nearby (Donnay, Duwe & Tewksbury 2008). “In New Zealand, serial sexual assault offenders committed their offenses an average of 3 kilometers (1.86 miles) away from their residences. For other types of violent crime, Groff and McEwen (2006) reported that homicide offenders committed the offense, on average, 0.69 miles from their homes. In addition, Tita and Griffiths (2005) showed that across 9 years of homicides in Pittsburgh, homicide offenders rarely killed in their own neighborhoods” (Donnay, Duwe & Tewksbury 2008).When offenders commit their crimes it is likely to result in different spatial patterns because factors related to opportunity and risk will vary even if motivational level across offender types is constant. Since the targets of serial burglars are immobile they will exhibit more commuting behavior than serial murderers (Bennell, Corey & Keyton, et al. 2007). A burglar can travel into the same area to commit more crimes in the future because while the crime is being committed he or she can locate potential targets through the observation of other residences in the area. While serial murderers may observe other targets while offending it is not very likely that these potential victims will remain in this same place for long (Bennell, Corey & Keyton, et al. 2007). Another factor that will determine whether an individual will exhibit commuting behavior is the level of perceived risk associated with committing crimes in the same geographic area (Bennell, Corey & Keyton, et al. 2007). Serial burglars would not be dissuaded from repeatedly commuting into the same area to commit their crimes because they are largely unaffected by the risks associated with committing multiple offenses in close proximity to one another (Bennell, Corey & Keyton, et al. 2007). Serial murderers would typically have to exhibit marauding behavior to avoid detection. As a result of the attention their crimes receive they would not be able to display the same behavior as the serial burglars (Bennell, Corey & Keyton, et al. 2007).

Rational Choice Approach

There is a relationship between the behavioral and geographic aspect of criminal behavior referred to as the rational choice approach. This approach recognizes that the offender’s behaviors are dependent on environmental cues associated with the criminal event like: nature (indoor vs. outdoor locations) and familiarity with the offense location (Allaire, Beauregard, Leclerc et al 2007).Target selection is highly dependent on the physical environment and there is a pattern in both spatial and temporal distribution of offenders and victims. As a result of the connection between the types of location and the types of strategy exhibited by an offender means that offender strategies might be triggered by the types of location at which the offender and victim met (Allaire, Beauregard, Leclerc et al 2007).One study conducted by Ouimet and Proulx (1994) reflects the correlates of spatial behavior in violent crimes by showing that a majority of child molesters offended in or near their residences. There are several advantages to an offender’s home over competing locations which make it the best possible location to commit an offense (Allaire, Beauregard, Leclerc et al 2007).One of these advantages may be facilitating the security children might feel which might make them more willing to participate in sexual contact. “There is an association between interactional, transactional and adaptive nature of human behavior. The level of violence of the crime is positively associated with the distance traveled by the offender from his home to the target.” (Allaire, Beauregard, Leclerc et al 2007).The offender has to adapt his crime strategies and use the appropriate approach method for the situation. Information processing and decision making occurs through experience whether or not a person is a criminal. For example, child molesters may have to travel farther if they are not able to find a suitable victim near their homes. It will become harder for the offender to convince a child take a car trip with a stranger and get him or her to return to the offender’s home the farther he has to travel (Allaire, Beauregard, Leclerc et al 2007). Geographic profiling must take into account the linkage between location types and both offender victim search methods and attack methods. Hunting patterns may be helpful to determine which crime locations are the best predictors of an offender’s anchor point (Allaire, Beauregard, Leclerc et al 2007). “The relationship between offending and geographic behavior may serve as the basis for integrated criminal-geographic profiling as unique investigative strategy” (Allaire, Beauregard, Leclerc et al 2007).

CONCLUSION

Geographic profiling determines the most probable area that an offender lives through the use of an investigative methodology that reviews the locations of a connected series of crimes. This methodology is based on a model that describes the hunting behavior of the offender. It is generally applied in cases of serial murder, rape, arson, robbery and/or bombing cases, but may also be implemented in single crimes that involve multiple scenes or other significant geographic characteristics (Harris 1999). Geographical profiling is an attempt to make predictions about an offender based on information obtained from the crime scene such as the location and timing of the offense (Sammons n.d.). People in general take more short trips than long trips in their daily lives according to the distance decay concept, so offenders are more likely to live close to the sites of their crimes than far away (Harris 1999). Geographic profiling is essential in that it refocuses the scope of the case from the whole metropolitan area to a small area of the community which in turn substantially reduces the amount of time and resources required to conduct the investigation. (Ramsland n.d.) Geographic profiling provides a means for managing the large volume of information generated in major crime investigations and should be regarded as one of several powerful decision support tools available to the detective. It is best employed in conjunction with other police methods and does not in itself solve cases. When properly decoded geographic crime patterns are clues that can be used to point the detective in the direction of the offender (Harris 1999).