2 part paper

Crime mapping:

fimUraphiral n"rappinU

*lrat*{t}ies u$*d i* visr.raliz* a n i: rn i:r r *f th i rrg*, inr,l*rJing 't*t:ttirs*,tl i gtiln * t, an d paitcrns *f *rim* a*tlrh*ir t;*rr*lat**,

Geographic information

system (GIS):

Th* srfnvar*tts*llhnl lsx.* nadrt *rir** r*xpfiirt# i n * r *,o,*tn *t,l au ail aYsl * t* r***arther* sitrc* ihe i*#**,

The size of the shapes in the graphs (e.g., circles, diamonds) indicate the degree of between- ness centrality. The highest-scoring individual at the beginning of the investigation was Melvyn Foster, who is represented by the square containing the large diamond in the Exhibit 1 1.3A. By 18 months, however, Gary Leon Ridgway becomes more noticeable (depicted by the shaded triangle in the second box). Unfortunately, the task force did not focus on Ridgeway but contin- ued to focus exclusively on Foster. Bichler and her colleagues concluded,

It is hard to know in hindsight if these results would sway the organization momenflrm that led investigators to focus on Foster to the exclusion of Ridgway. . . . It might have prevented the working hypothesis from solidiiring so early on in the investigation by encouraging members of the Green River Killer Task Force to pay greater attention to GaryLeon Ridgway.

While this research cannot predict what would have happened had SNA been available then, it cerainly highlights the utility of SNA for investigative purposes today. We move on next to a research technique that is frequendy being used for predictive purposes in law enforce- ment.

CRIME MAPPING

Many of us have adopted an image of crime mapping that involves a police precinct wall with pushpins stuck all over it identif.ing the location of crime incidents. Shows such as Tlte Wire, Tlte Dimict, and episodes of CSI have also presented the drama behind the use of crime mapping in generating crime counts within various police beats. Crime mapping for intelligenceJed policing has been increasing in the past few decades, but crime mapping for research purposes has a very long history; it is generally used to identifiz the spatial distribu- tion of crime along with the social indicators such as poverty and social fisorganization that are similarly distributed across areas (e.g., neighborhood, census tract). Boba (2009) defines crime mapping as 'othe process of using a geographic information s,'stem (GIS) to conduct special analysis of crime problems and other police-related issues" (7). She also describes the three main functions of crime mapping:

1 . It provides visual and statistical analyses of the spatial nature of crime and other events.

2. It allows the linkage of crime data to other data sources, such as census information on poyerty or school information, which allows relationships among variables to be esablished.

l. It provides maps to visually communicate analysis results.

Although applied crime mapping similar to this has been used for over one hundred years to assist the police in criminal apprehension and crime prevention, the advent of computing technology has enabled crime mapping to become an advanced form of statistical data analysis. The geographic information system (GIS) is the software tool that has made crime mapping increasingly available to researchers since the 1990s.

Today, crime mapping is being used by the majority of urban law enforcement agen- cies to identifiz crimg hot spots (Caplan and Kennedy 2015). Hot spots are geospatial loca- tions within jurisdictions where crimes are more likely to occur compared to other areas. Being able to understand where crime is more likely to occur helps agencies deploy resources more effectively, especially for crime prevention purposes. These hot spots can be specific addresses, blocts, or even clusters of blocks @ck et al. 2005). Of course, crime mapping data with insight from criminological theory is the ideal. fu Eck and his colleagues explain, "Crime

TOPICAL RESEARCH DESIGNS334 SECTION IV

theories are critical for useful crime mapping because they aid interpreation of daa and provide guidance as to vrhat actions are most appropriate" (3). This is important because the ability to undersand why crimes are occurring has a great deal to do with underlying factors related to the environment in which they occur. Kennedy, Caplan andPiza (2012) provide a very illuminating example:

A sole analytical focus on crime hotspots is like observing that children frequently play at the same place every day and then calling that place a hotspot for children playing, but without acknowledging the presence of swings, slides, and open fields-features of the place (i.e., suggestive of a playground) that attract children there instead of other locations absent such entertaining feattres. Q4546)

Through various qrmbols, maps can communicate a great deal of information. Exhibit 11.4, which was published by the National Institute ofJustice, displays some common symbols used by crime analysts (Eck et al. 2005). As the map shows, dots (A) point to specific places where crime is likely to occur, a crime site @ and C) indicates where crime is equally likely to occur within a panicular site, and a crime gradient (D) indicates that the probability of crime is most likely inside the site and decreases as you move toward the edge of the site.

CHAPTER 11 . SOCIAL NETWORK ANALYSIS, CRIME MAPPING, AND BIG DATA 335

fennifer A. Herbert, MA, Grime Intelligence Analyst, Grime Analysis and Strategic Evaluation Unit

Source: Courtesy of Jennifer A. Herbert

Jennifer Herbert gradu- ated with a double maj or in political sci- ence and justice stud- ies from James Madison University in 2007 . She had aspirations of becoming a police offi- cer and eventually a detective. She was hired as a police officer after graduation, but she real-

ized while at the police academy that she wanted to pursue the crime analysis career path in law enforce- ment. She became a crime analyst at Chesterfield County Police Department in Virginia. While work- ing full time as an analyst, Jennifer pursued a mas- ter's degree in intelligence at the American Military University. She then accepted a promotion to crime intelligence analyst at Henrico County Police Division. After working as a crime analyst for six years, Jennifer cannot imagine doing anything else.

Every day is different when working as a crime intelligence analyst. Some days, Herbert analyzes phone records and maps a suspect's whereabouts dur- ing the time of a crime. Other days, she maps the lat- est residential burglary trend and predicts where the next burglary will occur. She also completes research

projects that examine quality-of-life issues for the community, including estimating crimes per 1,000 residents by neighborhood. Herbert's role as a crime analyst is equally important in preventing crime and in helping patrol officers to apprehend offenders. She thinks the most rewarding part of her job is helping people who have been victimi zed by ,pprehending offenders and improving the quality of life for county residents. Jennifer has some good advice for students interested in careers involving analysis:

If crime analysis interests you, ask your local police department if you can do an intern- ship (paid or unpaid) to gain experience. Be sure to network with other crime analysts and let them know you are interested in pursuing a career in crime analysis. Courses in all forms of data analysis and GIS (geo- graphic information systems) are almost essential to a career in crime analysis. Even if you did not take GIS classes during your undergraduate studies, many community colleges offer introductory and advanced classes in GIS. Other qualifications that wilt help you stand out as an applicant include competency in basic statistics and profi- ciency in data analysis programs, including Microsoft Excel, Access, and SPSS.

Chance of crime being part of pattern is high at site, but zero at other places.

Everywhere outside the chance of being a crime site that is part of the pattern is zero.

Everywhere inside is equally Iikely to be a crime site that is part of the pattern.

Chance of a site being Part of the pattern changes radically at the edge.

Source: Eck et al. 2005, Exhibit 3.

CASE STUDY

Gradient indicates that the probability of being a crime site that is part of the Pattern varies from low near the edge to high near the center.

Chance of crime site being part of pattern is equal anywhere along street segments, and zero elsewhere.

Social Disorganization and the Chicago School

fu we noted earlier, crime mapping for general research purposes has a long history in crimi- nological research. Although they were not the first to use crime mapping, Shaw and McKay

1tl+i; conducted a landmark analysis in criminology on juvenile delinquency in Chicago neighborhoods back in the 1 93 0s. These researchers mapped thousands of incidents of iuve- nll. d"li.,qo"ncy and analyzed relationships between delinquenry and various social condi- tions such as social disorganization. After analyzing rates of police arrests for delinquenry, Shaw and McKayused police iecords to determine the names and addresses of those arrested

in Chicago between L927 and 1935. They observed a striking pattern that persisted over the

years, as shown in Exhibit 11.5. Exhibit 11.5 displays one of the maps Shaw and McKay (1942) created that illumi-

nates the spatial disuibution of delinquenry within concentric circles of Chicago that spread out io the suburbs from the city's center. As noted in Exhibit 11.5, there is a linear dicrease in rates of delinquenry as the distance from the Loop (city center) increases. When rares of other community characteristics were similarly mapped (e.g', infant mor-

taliry tuberculosis cases, percentage of families who own their own homes, percentage of

TOPICAL RESEARCH DESIGNS336 SECTION IV

ExhibitTL.4 Symbols Used in Grime Maps

foreign-born residents, percentage of families on relief), the conclusions rilr'ere obvious. Shaw and McKay concluded,

It may be observed, in the first instance, that the variations in rates of officially recorded delinquents in communities of the city correspond very closely with varia- tions in economic status. The communities with the highest rates of delinquents are occupied by these segments of the population whose position is most disadvantageous in relation to the distribution of economic, social, and cultural values. Of all the com- munities in the city, these have the fewest facilities for acquiring the economic goods indicative of status and success in our conventional culture. . . . In the low-income areas, where there is the greatest deprivation and frustration, where, in the history of the city, immigrant and migrant groups have brought together the widest variety ofdivergent cultural traditions and institutions, and where there exists the greatest disparity between the social values to which the people aspire and the availability of facilities for acquiring these values in conventional ways, the development of crime as an organized way of life is most marked. (3 18-19)

6.6

4.3

3.9

6.1

LL J I T F :f o U) tr o LL

U) LU F E,

B. Zone rates of police arrests, 1927

Source: Shaw and McKay 1942.

CHAPTER 11 . SOCIAL NETWORK ANALYSIS, CRIME MAPPING, AND BIG DATA 337

Exhibit 11.5 Zone Rates of Police Arrests in Ghicag or 1937

W

ENTIRE | 7.8 I rr

FOR 4.3

ilt

I ntelligence-led policing:

iJsing data, anal'ysis, anrl

r:rininal tfuer"rry t* uuirJ* # *ii r: * ali * r:'ati * n anr| decisi*n making,

CASE STUDY

Predicting Break and Entries (BNEs) Contemporary researchers and law enforcement ofEcials interested in issues related to crime and criminology have access to more sophisticated computer technology that allows for the creation of more enhanced crime maps. In fact, the easy availability of mapping tools, includ- ing mobile GIS technology, is providing many more opportunities for intelligence-led policing, which includes using "data, analysis, and criminal theory . . . to guide police alloca- tion and decision making" @itterer, Nelson, and Nathoo 2015, l2l). The purpose of crime maps, however, remains the same: to illuminate the relationship between categories of crime and corresponding characteristics such as poverty and disorganization across given locations.

Break and entries (BNE ) are one type of crime that are patterned; in facg one of the things we know about them is that the probability of a repeat BNE increases for either the original BNE site andlor for homes near it for several weels after the original BNE. To deter- mine the effectiveness of mapping in predicting residential and commercial BNEs, Fitterer and her colleagues (2015) used daa from the Vancouver Police Districts (VPD) alongside data on other characteristics ofthe districts, including variables such as population density, property values, dominant housing types, and street Iight density.

UsingBNEdatabylocation,hour,day,month,andyearfrom200l to2}l2,rheresearch- ers then created a map of the Vancouver area composed of 200m by 200m grids, placing the data within each grid. For example, Exhibit 11.6 displays the map depicting residential BNE hot spots for the time period. The goal of their research, however, was not merely to describe the BNE data but to use data from the early time period to predict later occurrences of BNEs. For example, Fitterer and her colleagues found that the probability of a repeat BNE occur- ring up to 850m from the originating BNE increased 53"/" for the next 24 hours after the first event. While there was still an increase in the likelihood of near BNEs over time, this increased risk decreased to 24% after a week of the original BNE. Imporandy, the propor- tion of historical crime did significandy predict future crime. The authors concluded,

We found [that] both residential and commercial crimes had a strong spatial cluster- ing over short time periods, suggesting a near-repeat offense dynamic . . . [indicating] that perpetrators prefer to reoffend where they have local knowledge about residents' routine activities, possessions, and can confirm successful property entry. (130)

Residential BNE Hot Spots Quadratic Density * 100,000

ffio-21 w22- 46 Effi47 - 82 w83 - 138 E 139 - 247

Source: Adapted from Fitterer et aI. (2015), Figure 9, page 129.

338 SECT!ON !V . TOPICAL RESEARCH DESIGNS

Exhibit 11.6 Vancouver's 2OOL-2O11, Break and Enter Hot Spot Map for Residential Break-Ins

CASE STUDY

Using Google Earth to Track Sexual Offending Recidivism While the GIS software that was utilized by Fitterer et al. (2015) has many research advan- tages for displaying the spatial distributions of crime, other researchers have begun to take advantage of other mapping tools, including Google Earth. One such endeavor was con- ducted by Duwe, Donnay, and Tewksbury (2008), who sought to determine the effects of Minnesota's residency restriction statute on the recidivism behavior of registered sex offend- ers. Many states have passed legislation that restricts where sex offenders are allowed to live. These policies are primarily intended to protect children from child molesters by deterring direct contact with schools, day care centers, parls, and so on. Most of these statutes are applied to all sex offenders, regardless oftheir'offendinghistory or perceived risk ofreoffense.

The impact of such laws on sexual recidivism, however, remains unclear. Duwe et al. (2008) attempted to fill this gap in our ftnowledge. They examinedZ}4 sexaffenders who had been reincarcerated for a new sex offense between 1990 and 2005 and asked several research questions, including "Where did offenders initially establish contact with their victims, and where did they commit the offense?" and "What were the physical distances between an offender's residence and both the offense and first contact locations?" (488). The research- ers used Google Earth to calculate the distance between an offender's place of residence, the place where first contact with the victim occurred, and the location of offense.

Duwe and his colleagues (2008) investigated four criteria to classifir a reoffense as pre- ventable: (1) the means by which offenders established contact with their victims, (2) the

CHAPTER 11 . SOCIAL NETWORK ANALYSIS, CRIME MAPPING, AND BIG DATA 339