Loading...
HomeMy WebLinkAboutCC Ex E Sex Offender Walker reportSee discussions, stats, and author profiles for this publication at:https://www.researchgate.net/publication/240822071 The Geographic Link Between Sex Offenders and Potential Victims: A Routine Activities Approach Article in Justice Research and Policy · September 2001 DOI: 10.3818/JRP.3.2.2001.15 CITATIONS 59 READS 90 3 authors, including: Jeffery T Walker University of Alabama at Birmi… 27 PUBLICATIONS 267 CITATIONS SEE PROFILE James W Golden University of Arkansas at Little … 5 PUBLICATIONS 61 CITATIONS SEE PROFILE Available from: Jeffery T Walker Retrieved on: 17 August 2016 Sex offenders and victims • 15 *The Geographic Link Between Sex Offenders and Potential Victims: A Routine Activities Approach Jeffery T. Walker University of Arkansas at Little Rock and Arkansas Statistical Analysis Center James W. Golden Amy C. VanHouten University of Arkansas at Little Rock JUSTICE RESEARCH AND POLICY, Vol. 3, No. 2, Fall 2001 © Justice Research and Statistics Association This research was made possible by a grant from the Bureau of Justice Statistics through the State Justice Statistics Program, grant number 98-MU-MU-K013. Points of view in this document are those of the authors, and do not necessarily represent the official posi- tion or policies of the U.S. Department of Justice, Arkansas Crime Information Center, or the University of Arkansas at Little Rock. *Abstract Despite the publicity surrounding several high-profile sex offender incidents in the 1990s, relatively little research has been done on sex offenders themselves. This ar- ticle reports on a study of sex offenders in one Arkansas county who had children as victims. Specifically, we examined the relationship between where the offenders live and where children congregate to see whether offenders choose to reside in areas with high concentrations of children. We use routine activities theory as a foundation for explaining our results, and conclude with a discussion of the policy implications of the findings. 16 • Justice Research and Policy Although rapists have drawn the attention of the criminal justice system for many years, the activities of other types of sex offenders were brought to the forefront of public attention in the 1990s, spurred by several high-profile kidnappings, molestations, and murders of children. Legislation such as Megan’s Law and the Jacob Wetterling Act propelled sex offenders into the public spotlight by requir- ing that they register with the state and that the public be notified of their exist- ence and often their whereabouts. Relatively little research, however, has focused on sex offenders themselves, and most research that has been conducted has used very small samples or has remained relatively obscured in the tables and graphs of state crime publications. This study attempts to overcome some of those limitations. We examined the population of sex offenders who had children as victims in one metropolitan county in Arkansas to see where offenders live in relation to parks, schools, and day cares. Our argument is that child sex offenders are largely incorrigible and they may attempt to live in areas with a high concentration of children. We use routine activities theory to explain our results, and conclude that this theory is capable of explaining this type of crime and may be beneficial to policymakers in controlling sex offenses against children. *Sex Offenders and Victims In June 2000, Christopher Reardon was arrested for sexually abusing young boys. Interest in the case increased when it was learned that Reardon used his job as a youth minister and YMCA swim coach to befriend the boys before he mo- lested them. Reardon ultimately pled guilty to 120 counts of sexual abuse of a child. This case is only one example of the many sexual abuse cases in which adults use their physical or geographic relationship to gain access to young children. The National Clearinghouse on Child Abuse and Neglect (1996, p. 8) reports that the number of cases of sexually abused children increased from about 119,200 in 1986 to 217,000 in 1993, an 83% increase. Hillman and Solek-Tefft (1988, p.1) estimate that at least 25% of all Americans were sexually abused as children. Even though many children are sexually abused each year, little research has focused on offenders who target children (Crewdson, 1988, p. 115). Sexual abuse has been defined in a variety of ways. McGrath (1991, p. 341), for example, describes it as an act that takes advantage of a “weaker, more vul- nerable person.” He suggests that offenders often choose victims who are unable to fight back, such as children and those who suffer from a mental illness or mental retardation. Colton and Vanstone (1996, p. 2) define child sexual abuse as the “carrying out of sexual acts by adults with children below the age of Sex offenders and victims • 17 consent.” While definitions of child sexual abuse may vary, the actions they de- scribe are ultimately the same: An adult takes advantage of a child for the purpose of sexual gratification. *Targeting Children One of the most important characteristics of child sex offenders is that they generally have a method of gaining access to children. Goldstein (1987, p.91) argues that there are three ways in which sexual offenders gain access to chil- dren: marriage, neighborhood, and occupation. These methods are not mutually exclusive. Offenders may molest their own child, molest the children on the ball team that they coach, and work in a business that targets children. Goldstein (1987, p. 83) proposes that it is not uncommon for an individual to marry for the sole purpose of gaining access to children. Perpetrators who use marriage as an access technique often seek out women who already have children who meet their age and gender preferences. The marriage often lasts only as long as the children are in the victim target range. Offenders also might marry and take in foster children or adopt children. They will rarely produce their own children for this purpose, however, since there is no guarantee that they will be the preferred gender (Goldstein, p. 83). Hillman and Solek-Tefft (1988, p. 27) argue that offenders might also have a relationship with the family of a child victim. A second way sex offenders might gain access to a child is by searching for neighborhoods with children. Goldstein (1987, p. 98) insists that many pedo- philic activities are premeditated. Goldstein conducted part of his research by examining publications used by pedophiles. In these publications, he found sev- eral articles and other material suggesting that offenders target places such as arcades, schools, and playgrounds. Offenders have also been known to use par- ticular activities to entice children. For example, a perpetrator might approach children at bus stops, schools, and playgrounds and perform magic tricks. After gaining their trust, the offender may trick them into performing sex acts (Goldstein, p. 106). When a pedophile chooses to target a victim, age preference is the key factor. Abusers who prefer adolescents are more likely to become scout leaders and coaches, while those attracted to younger children may seek employment in day care centers (Crewdson, 1988, p. 117). Lanning (1992, p. 18) describes sex offenders as frequently associating with young people. The sex offender may hang around schoolyards, arcades, shop- ping centers, or anywhere else that children are present. He suggests that child sex offenders have well-developed techniques for obtaining victims (Lanning, 18 • Justice Research and Policy p.19). He or she may also have hobbies and interests that specifically appeal to children, such as collecting specific toys or dolls. A third way in which sex offenders might target victims is by choosing occu- pations allowing access to children. Goldstein (1987, p. 108) argues that offend- ers often become involved with programs catering to the children in their target group, or they may become employed at a business that caters to children. Goldstein further notes that if offenders are unable to get into a business that caters to children, they might start one of their own. Colton and Vanstone (1996) profiled seven men who were convicted of sexually abusing children. In this study, all of the subjects’ occupations gave them access to children. These occupations included working as a schoolteacher, youth worker, church employee, and as a volunteer for various activities with youth organizations. Kelley, Brant, and Waterman (1993) argue that sexual abuse can occur in any setting where children can be found and that sexual abuse perpetrators pur- posely work to know the locations of concentrations of children. One place that caters to children is day care centers. Several studies have been conducted to determine the incidence of sexual abuse in the day care setting Approximately 1 child in 10 is cared for outside the home (Crewdson, 1988, p.117). There is a consensus that children in day care are at a higher risk for sexual abuse (Kelley et al., 1993, p.71). Sexual abuse data indicate that an aver- age of 5.5 children per 10,000 enrolled in day care are sexually abused (Murray, 2000, p. 213). Kelley et al. (p.71) report that many of the cases involving mul- tiple or serial sexual abuse involve day care centers. In one of the most notori- ous cases, Virginia McMartin, her daughter, her grandson, and many of the preschool’s employees were charged with sexually abusing the children in their care (Crewdson, p.134). This case was even more tragic given the evidence that the daycare employees and other people in the community were involved in an elaborate child sexual abuse ring, and that the day care had been set up for the sole purpose of using children for sex (Crewdson, p.135). Finkelhor, Williams, and Burns (1988) examined 500 day care facilities in an attempt to determine if day cares were high-risk environments for sexual abuse. They found 2,500 cases of child sexual abuse in the 500 facilities. Of the perpetrators, 16% were the day care’s owner or director. In one case a man with a previous conviction for child molestation convinced his wife to start a baby- sitting service, illustrating the importance of the relationship between convicted child sex offenders and day care centers (Finkelhor et al.). Investigators in the case concluded that since all children at the day care were abused, the perpetrator’s sole purpose in establishing the day care was to obtain access to children to sexually abuse. Sex offenders and victims • 19 *Routine Activities and Crime In 1979, Cohen and Felson presented a theory that took a different view of criminal behavior. Instead of only examining the activities of offenders or victims, Cohen and Felson proposed that criminal behavior is a union of criminal, victim, time, and place. This became known as routine activities theory. The focus of routine activities theory is the natural activity patterns of people. Cohen and Felson (1979, p.593) define routine activities as “any recurrent and prevalent activities which provide for basic population and individual needs.” They argue that criminal activities follow patterns similar to those of “everyday suste- nance activities,” meaning that criminals go about their lives in the same kinds of movement and behavior patterns as those who go to work, school, shop, and undertake hobbies. It is when the “routine activities” of a criminal overlap the “routine activities” of a victim that the potential for a criminal violation occurs. According to Cohen and Felson, routine activities bring people together in both criminal and non-criminal behaviors. “Hence the timing of work, schooling, and leisure may be of central importance for explaining crime rates” (Cohen & Felson, p.591). It is this convergence of victims and offenders in time and place that is central to routine activities theory, and to the research at hand. Cohen and Felson (1979) propose that their theory is most applicable to “direct-contact predatory violations,” that is, when an offender has direct physi- cal contact with a victim. They believe that routine activity patterns are affected by three elements: motivated offenders, suitable targets, and the absence of ca- pable guardians against a violation. A change in any one of these could increase or decrease the crime rate. While it has been popularly held that an increase in the number of offenders could increase the crime rate regardless of the number of vic- tims, it may also be true that an increase in the number of suitable targets in an area could increase the crime rate, even without an increase in the number of offenders. In a follow-up study published in 1987, Felson extended the original theory by focusing on urban characteristics, including locations that attract both poten- tial offenders and victims. In this research, Felson proposes that criminals often position themselves to naturally come into contact with potential victims: “Just as lions look for deer near their watering hole, criminal offenders dis- proportionably find victims in certain settings . . ..” (1987, p. 914). He goes on to suggest that certain areas of the city bring offenders and targets together so effectively that they almost draw the people to the area, they make it easy for offenders to find suitable victims, and they create areas of the city that are very high crime risk areas. Public facilities such as parks and schools were among the areas mentioned. As such, routine activities theory was expanded to show that not only do offenders and 20 • Justice Research and Policy victims often come together as a result of their everyday activities (and by extension, criminals could increase their odds of coming into such contact by manipulating their activities), but places in the city could also promote such interaction. Following the work by Felson, much of the research related to routine activi- ties theory began to examine more specific locations and/or offender types (see, for example, Sherman, Gartin, and Buerger, 1989, and Smith, Frazee, and Davison, 2000). While this line of research is related to the research at hand (for example, Roncek and Maier [1991] found that increases in potential targets [in their case bars and taverns] in residential areas can significantly increase the amount of crime), no literature currently applies routine activities theory to child sex of- fenders. This study attempts to begin to fill that void by arguing that child sex offenders may put themselves in a position to come into contact with potential victims and that some places may draw offenders and targets together. *Methods This research was spawned from a larger project that examined the characteristics of all sex offenders in the state of Arkansas. For that project, the records of all sex offenders registered with the state on February 1, 1999, (a total of 1,646) were col- lected. These data were augmented by the criminal history records of the offenders. For the project at hand, all the sex offenders in that database who had a residency address in one metropolitan county (Pulaski County) were selected. This county was chosen because it had the largest number of sex offenders of any county in the state and, for the purposes of this project, it was the only county that had been mapped sufficiently to be able to relate sex offender residences with potential targets. Of the 1,646 sex offenders in the database, 252 were from this county. Because of the nature of the targets selected for this analysis (areas where a concentration of children were likely), this data set was further broken down into offenders who had a child as a victim and those who did not. A total of 170 offenders had a child as a victim. The second set of data to be collected was the locations of the potential target areas. We selected day cares, schools, and parks as the potential targets based on legislation at the state and federal level that specifically identify schools, playgrounds (parks), and day cares as areas of concern for criminal activities such as gun possession, drug use, rehabilitation centers, and sex offenders. Fur- thermore, we decided to use these targets because they represent areas where typically there is a concentration of potential victims. School districts within the county supplied the address of each primary and secondary school. A total of 83 schools were identified in the county. The loca- Sex offenders and victims • 21 tion of parks in the county (a total of 43 parks) was obtained from the Parks and Recreation Department. Phone records were used to obtain a listing of all day cares in the county. After identifying all day cares, we confirmed the addresses by contacting the day care by telephone. A total of 171 day cares were identified. After all of the addresses for the sex offenders and targets were obtained, they were geocoded and mapped using ArcView. A number of geocoding prob- lems had to be overcome to proceed with the project. These problems resulted from changes in street names and streets that had been added in the county after release of the maps used in the analysis. Ultimately 90% of the addresses for offenders, 88% of the addresses for schools, 95% of the addresses for parks, and 85% of the addresses for day cares were included in the analysis. We do not see this as a problem, however, because of the nature of the project. The analyses here are mapping the relationship of the location of targets to where child sex offenders live. Problems with the geocoding resulted from inadequate mapping points for geocoding the addresses. As a result, if a sex offender could not be properly mapped to a particular street segment, then neither could the potential target (if there was no street segment mapped to geocode an offender, then there would be no street segment mapped for any of the targets). About the same percentage of each of the data sets could not be geocoded (from 5% to 15% of addresses). Again, since neither target nor sex offender could be mapped, a re- port of the relationship between the two should not be affected. After the data on offenders were broken down into those who offended against children and those who did not, three sets of analyses were conducted. The first was a point location analysis whereby the point locations of all child sex offend- ers and all of the potential targets were geocoded. The maps were then examined for any concentrations or abnormalities that might be important in determining the spatial relationship between child sex offenders and potential victims. In the second analysis, 1,000-foot buffer zones were drawn around each of the targets in the county. Each of the buffered potential targets was then exam- ined to determine how many child sex offenders lived within them. In addition, the number and percentage of child sex offenders living in the buffer zones were compared to the number and percentage of non-child sex offenders living in those buffer zones. The concept of a 1,000-foot buffer zone came from examining federal regu- lations and the Arkansas code, which prohibit certain activities near schools. For example, drug-free zones are established by 21 U.S.C. § 860, which enhances penalties for distributing illicit substances within 1,000 feet of any school, uni- versity, or playground (park). Arkansas Code 5-64-411 provides for a 1,000- foot drug-free zone around schools. Furthermore, 18 U.S.C. § 921 defines a “school zone” as any place “in or on the ground of a public, parochial, or private 22 • Justice Research and Policy school or within a distance of 1,000 feet of a public, parochial, or private school.” In the third analysis, Spatial Analyst, a tool that measures the physical dis- tance between objects and how densely they are concentrated, was used to examine the relationship between each of the target types and where the offenders lived. *Study Limitations As with any research, this study has certain limitations. The primary limitation is the assumption that child sex offenders are choosing to live within close proxim- ity to potential targets. The premise going into this research was that if there was a predominance of child sex offenders living close to potential victims, at least part of the reason was to take advantage of these potential targets. This initial premise was supported by our findings, but as with any research, rejection of the null hypothesis (that there is no relationship between where child sex offenders live and the geographic location of potential targets) does not imply absolute acceptance of the research hypothesis (that sex offenders choose to reside in close proximity to potential targets). Alternative hypotheses are also possible. For ex- ample, as discussed below, sex offenders could be residing in these areas because of a social or political inability to exclude undesirable residents from living there; or sex offenders may choose to live in these areas because they are more afford- able. It may be more accurate to say, then, that there is a relationship between where child sex offenders live and the geographical location of potential targets irrespective of choice. Since most people do make choices about where they live, however, it is also plausible (and is supported by the co-location of child sex offenders and targets in this research) that sex offenders are choosing to live in close proximity to potential victims. A second and related limitation of the study is that it does not make com- parisons with residents in these areas who had not been convicted of a sex of- fense. As one reviewer suggested, information should have been collected on the relative similarity of age, race, and income between sex offenders and other resi- dents. While this kind of information would have been helpful, we do not believe that it substantially changes the conclusions herein. First of all, this kind of infor- mation was not available for the sex offenders in this research (and neither was information we desired, such as whether the offender had children or not). Addi- tionally, this information is not particularly accurate for the general population. The 2000 Census was not out at the time of this research, and we believed that 1990 Census data were probably inaccurate given the changes in the characteris- tics of neighborhoods. Finally, we argue that it makes little difference to policymakers whether sex offenders are overrepresented on mean income in an Sex offenders and victims • 23 area or whether one area has 1.7 sex offenders per 1,000 population while another has 1.5 sex offenders per 1,000 population. We believe that what is important is how many child sex offenders are living within 1,000 feet of a potential target. *Analyses The analyses for this project were broken into three parts: The first was an ex- amination of the point locations of child sex offenders and targets within the county, the second was an examination of buffer areas around the potential tar- gets as they related to where sex offenders lived, and the third was an examina- tion of the spatial density of the relationship between the child sex offenders and potential targets. Point Location Analysis Although the map in Figure 1 is quite crowded, it is easy to see the substan- tial overlap between where child sex offenders live and the locations of potential targets. Obviously, a more detailed analysis was needed. Through a visual ex- *Figure 1 Pulaski County Sex Offenders - Overview (Portion enlarged to show detail) 0816Miles N Daycares Offenders Parks Schools 24 • Justice Research and Policy amination of the maps, we identified 36 areas where a number of child sex offenders were living in close proximity to one or more potential targets—areas where there appeared to be a higher concentration of child sex offenders, or where the physical relationship between child sex offenders and targets seemed to be abnormal. Figure 2 details one of the 36 areas identified as containing a concentration of child sex offenders in close proximity to potential targets. The map has two particular areas of interest. The first is in the upper right-hand corner, where there were twelve child sex offenders living within a few blocks of four day cares and a park. Also interesting is the area in the map’s center, which shows a close interspersing of sex offenders around a number of day cares, schools, and parks. In one of these cases, three child sex offenders lived within five blocks of a school, day care, and park, and one of the sex offenders lived on the same block as the park and only two blocks from a school and day care. The three sex offenders essentially surrounded one day care, and one of the sex offenders lived on the same block as a day care. Only a short distance away, three other child sex of- fenders surrounded another day care, and in another place, one child sex of- fender lived one block away and between a school and a day care. N 0 0.3 0.6 Miles Daycares Offenders Parks Schools *Figure 2 Pulaski County Sex Offenders - Cluster of Offenders and Targets Sex offenders and victims • 25 Each of the other areas within the 36 identified as concentrations exhibited patterns similar to that shown in Figure 2. Some areas had a greater concentra- tion of child sex offenders around day cares and schools than the area shown in Figure 2, but the latter area was used as an example because the multiple sites within the concentration area demonstrated the likelihood of several concentra- tions even within a single area. Figure 3 also illustrates the co-location of child sex offenders and targets. In both areas the sex offenders were living in very close proximity to the day care. What is particularly interesting, though, is that each day care was in a relatively rural area, not particularly close to anything, and yet a child sex offender (or in one case two child sex offenders) lived in very close proximity. What we determined from analyzing the 36 concentration areas is that in some areas of the city, a large number of child sex offenders lived in close prox- imity to concentrations of children that may serve as potential targets. The point location relationship between child sex offenders and potential targets supports the argument that at least some child sex offenders are selecting where they live based on potential targets in the area. This was particularly clear when the maps and relationships were examined in detail, especially when the child sex of- fender and target were not co-located with other people. What must be avoided, however, is assigning a purposeful nexus between offenders and targets simply Daycare Offenders 0400Feet Offenders Daycare 600 400 200 0 Feet *Figure 3 Pulaski County Sex Offenders - Proximity of Offender and Daycare Area 1 Area 2 26 • Justice Research and Policy because of their co-location. This relationship may exist because these areas are more accepting of others, or because the residents do not have the social or po- litical power to prevent sex offenders from living in the area. When these condi- tions exist in areas where there are a number of potential targets, however, the likelihood of child sex offenders choosing to live close to potential targets ap- pears to increase. Visual examination of a map does not represent a complete analysis of this relationship, however. A second analysis was undertaken, there- fore, that used buffer zones around the potential targets to see how many child sex offenders were living in those areas. Buffer Areas By counting the number of sex offenders living in the 1,000-foot buffer zones around potential targets, an estimation of the relationship between the sex of- fenders and targets could be made. We found that just under half (48%; 82 of 170) of all sex offenders in the data set who had children as victims lived within the buffer zone of one of the potential targets, and over one third lived within the buffer zones of multiple targets. An example of the analysis and results of buffering potential targets is shown in Figure 4. Here, several of the schools and day cares did not have a child sex offender living within the buffer zone. Two areas are illustrative, however. The first is in the center left of the map, where two day cares and two schools heavily 0 0.5 1 Miles Daycare Schools Offenders N *Figure 4 Pulaski County Sex Offenders - Offenders and Buffer Areas Sex offenders and victims • 27 overlapped one another. Also in this buffered area was a child sex offender who lived immediately adjacent to a day care. A more explicit example of child sex offenders living in close proximity to potential targets is the area just to the right. Four day cares and a school were all buffered together. There were also four child sex offenders living within this buffered area, and four more who lived just outside the buffered area. As a part of this analysis, sex offenders in the data set who had children as victims were mapped against sex offenders in the data set who did not have children as victims to determine if there were differences in their residence pat- terns. As discussed above, 48% of all child sex offenders lived within the buffer zones. In comparison, 26% (19) of non-child sex offenders lived in these buffer zones. This lends support to the argument that child sex offenders are choosing to live in areas where there are concentrations of potential victims. At the very least, this finding is troubling for the criminal justice system and families alike: that there is a larger percentage and number of sex offenders who had children as victims living in close proximity to concentrations of children than there are sex offenders who did not have children as victims. The analysis of 1,000-foot buffers around potential targets provides some quantification for the argument that child sex offenders are choosing to live in areas where they have access to potential victims. In this analysis, almost half of all child sex offenders lived within 1,000 feet of a day care, school, or park, and a substantially higher percentage of child sex offenders were living in these areas than offenders who did not have children as their victims. Certainly some of the variation can be attributed to the distribution of residential areas within the county. This is not a total explanation, however, and the assumption must be made that some of the child sex offenders were choosing to live in areas where they had access to potential victims. This relationship was further explored in the final analysis of this research, which examined the spatial density of child sex offend- ers and targets. Spatial Relationships In the third and final analysis, Spatial Analyst was used to examine the rela- tionship between each of the target types and where the offenders lived. Although this is a qualitative measure, it does provide additional evidence that can be used to evaluate the assumptions of this research. As shown in Figure 5, there was a substantial relationship between where the child sex offenders lived and the locations of schools. The light area through- out this graph indicates a high co-location, with at least one sex offender living within one mile of at least one potential target throughout the area. There were a few areas with a lesser amount of overlap between the two, as indicated by the 28 • Justice Research and Policy darker spots in the map. Only a small portion of the graph was darker, however, which is the least amount of overlap, where sex offenders live more than two miles from a potential target. Contrast this with the map showing the relationship between offenders and parks (Figure 6), where there is obviously a lesser amount of co-location. Part of the reason for this pattern is that there were fewer parks and they were more spread out (73 schools were geocoded in the analysis, and only 41 parks). Much of the difference, however, is that there did not seem to be the attraction between child sex offenders and parks that there was with schools (discussed above) and day cares (which follows). The relationship between day cares and child sex offenders is similar to the relationship between schools and child sex offenders (see Figure 7). There was not as large an area that had the highest co-location between child sex offenders and potential targets; however, as Figure 7 shows, there was a fairly substantial relationship between these two, especially given that there were no darker areas (showing a more dispersed relationship) within the area of highest concentra- tion. This supports the argument that density is not just related to high residen- tial areas, and, conversely, that the relationship between child sex offenders and potential targets is not solely based on the density of the targets. *Figure 5 Pulaski County Sex Offenders - Spatial Density, Offenders to Schools 0816 Miles 0 - 2,499 Feet 2,500 - 4,999 Feet 5,000 - 7,499 Feet 7,500 - 9,999 Feet 10,000 Plus Feet N Sex offenders and victims • 29 *Figure 6 Pulaski County Sex Offenders - Spatial Density, Offenders to Parks *Figure 7 Pulaski County Sex Offenders - Spatial Density, Offenders to Daycares 10,000 Plus Feet 2,500 4,999 Feet 0 - 2,499 Feet N 04812 Miles N 0714 Miles 0 - 2,499 Feet 2,500 - 4,999 Feet 5,000 - 7,499 Feet 7,500 - 9,999 Feet 10,000 Plus Feet 30 • Justice Research and Policy *Discussion Routine activities theory proposes that crime occurs when motivated offenders, potential victims, and the absence of capable guardians merge in time and space. Our analysis of sex offenders and their potential targets in one Arkansas county showed a substantial overlap between those convicted of sex offenses (motivated offenders) and children who may be likely targets (potential victims). Some of the reason for this overlap is no doubt the purposeful actions of the offenders. Our results can be explained using the analogy of a regression model. Certainly not all child sex offenders purposefully move to be close to a school or day care. Part of the variation in where child sex offenders choose to live, therefore, is explained in terms of market conditions, in the inability or apathy on the part of the neighborhood to keep the sex offenders out, and for other reasons. Part of the variation in the residency patterns of child sex offenders, though, is no doubt attributable to them putting themselves in close proximity to potential targets, perhaps in hopes that their routine activities will overlap with the routine activi- ties of potential victims. Like the work of Cohen and Felson (1979), this research was particularly concerned with direct-contact predatory violations. A primary reason for the seriousness of sex crimes is that they represent physical attacks on the person. If, as Cohen and Felson (p. 595) argue, “target suitability influences the occurrence of direct-contact predatory violations,” then the location of potential targets and the proximity of potential offenders is of interest. Any changes in the environ- ment or activities of criminal behavior that can control predatory behavior should certainly be a concern for policymakers. Cohen and Felson (1979) also argue that increases in one of the three ele- ments that change routine activity patterns can increase the crime rate, regard- less of changes in the other elements. Furthermore, an increase may have a mul- tiplicative rather than an additive affect—that is, a small increase in the number of offenders who are seeking to target potential victims may produce an increase in the crime rate much larger than the increase in the number of offenders. If that is true, then a change in the number of child sex offenders living in close proxim- ity to potential targets deserves attention. In this research, we have attempted to establish that two of the elements of routine activities theory may be present in direct-contact predatory violations of child sex offenders. We argue that some child sex offenders will seek out suitable targets. When they find areas with one or more day cares, schools, or parks, they may seek residence there. A high number of child sex offenders living in a par- ticular area may signal that they are purposefully living there because of easy access to potential victims. Sex offenders and victims • 31 What remains to be addressed is the third element of routine activities: the presence of capable guardians. Routine activities theory states that the presence of capable guardians should reduce the likelihood of direct-contact predatory viola- tions. According to the theory, the actions taken by police officers and others to make day cares, schools, and residents of these areas aware of the presence of child sex offenders should increase the number of and/or effectiveness of capable guard- ians. The relationship between the presence of police and their notification of residents and inhabitants of an area was not explored by this research. It can be argued, however, that their actions could have a important impact on the poten- tial level of sex offenses in an area. The police cannot serve as the sole source of capable guardianship in controlling sex offenses, however. They are too few in num- ber, and they cannot always be present when motivated offenders and potential victims converge in time and space. What the police can do, however, is to make as many people in the neighborhood (especially those who are guardians of po- tential victims or may be potential victims themselves) aware of the presence of a potentially motivated offender. This can be done by the very actions that are being taken—registering sex offenders and notifying the community of their pres- ence. If it can be determined that these actions are having the desired effect, then the tripartite relationship central to routine activities theory is supported. *Conclusion The goal of this research was to examine the possibility of a link between where child sex offenders choose to live and potential targets. We believe that we have found this link. Our findings support the argument that there is a definite con- vergence of potentially motivated child sex offenders living in close proximity to concentrations of potential victims. The policy implications of this research directly relate to the resources and efforts of sex offender registration and notification. It is not possible for govern- ments or other entities to control the locations of potential targets (day cares, schools, and parks). Furthermore, despite governmental efforts, there is no evi- dence that attempts to limit where sex offenders live have been successful. Rou- tine activities theory argues that crime increases when there is a convergence of motivated offenders and potential targets in the absence of capable guardians. This research has shown that there appears to be a convergence of offenders and potential targets in the county we studied. Even if the argument that child sex offenders are choosing to live in these areas because of potential targets is re- jected, the fact remains that there is a substantial number of child sex offenders living in close proximity to potential targets. 32 • Justice Research and Policy Since an increase in one or both of these polulations may occur, thus poten- tially increasing the rate of sex offenses multiplicatively, and since it appears that neither of these elements can be controlled, the focus must shift to the third element of routine activities theory—capable guardians. In the present instance, this means that governments and the police must work to increase the effective- ness of sex offender registration and notification. Furthermore, it appears im- perative that more attention be given to community notification of the presence of sex offenders. In this way, all who live and work in these areas have the poten- tial to become capable guardians. As stated above, the police alone cannot pro- vide effective guardianship against sex offenses because they cannot be in all places at all times. It is important, then, that the community be enlisted to assist them. Additionally, routine activities theory argues that the more capable guardians there are, the greater the corresponding decrease in potential crime should be. If notification and subsequent community involvement is undertaken effectively, there may even be the multiplicative reduction in the rate of sex offenses, which has been the goal of sex offender registration and notification from its inception. A crucial next step in this line of research, then, is to further examine these findings in relation to potential guardians. Research should examine whether proper notification concerning sex offenders increases the effectiveness of ca- pable guardians in neighborhoods, i.e., whether neighborhoods that are aware of child sex offenders use that knowledge to increase guardianship at times and places where the convergence of motivated offenders and potential victims is likely to occur. If it is true that an increase in capable guardians can reduce crime, and if it is true that community notification concerning sex offenders can in- crease the number of guardians, then two substantial findings could result. First, routine activities theory would be given its strongest support to date, and sec- ond, and perhaps more importantly, the effectiveness of community notification of sex offenders, which has only been supported through belief and opinion to this point, could be validated. Sex offenders and victims • 33 *References Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A rou- tine activity approach. American Sociological Review, 44, 588–608. Colton, M., & Vanstone, M. (1996). Betrayal of trust: Sexual abuse by men who work with children–in their own words. London: Free Association Books. Crewdson, J. (1988). By silence betrayed: Sexual abuse of children in America. Boston: Little, Brown. Felson, M. (1987). Routine activities and crime prevention in the developing metropolis. Criminology, 25, 911–931. Finkelhor, D., Williams, L. M., & Burns, N. (1988). Nursery crimes: Sexual abuse in day care. Sage: London. Goldstein, S. L. (1987). The sexual exploitation of children: Practical guide to assessment, investigation and intervention. Boca Raton, FL: CRC Press. Hillman, D., & Solek-Tefft, J. (1988). Spiders and flies: Help for parents and teachers of sexually abused children. Lexington: Lexington Books. Kelley, S. J., Brant, R., & Waterman, J. (1993). Sexual abuse of children in day care centers. Child Abuse and Neglect 17, 71–87. Lanning, K. V. (1992). Child molesters: A behavioral analysis. Alexandria, VA: National Center for Missing and Exploited Children. McGrath, R. J. (1991). Sex offender risk assessment and disposition planning: A review of empirical and clinical findings. International Journal of Offender Therapy and Comparative Criminology, 35, 328–348. Murray, J. B. (2000). Psychological profile of pedophiles and child molesters. The Journal of Psychology, 134, 211–224. National Clearinghouse on Child Abuse and Neglect Information. (1996). Execu- tive summary of the third national incidence study of child abuse and neglect. Online report: http://www.calib.com/ncccanch/pubs/statsinfo/nis3.cfm. Roncek, D. W., & Maier, P. A. (1991). Bars, blocks, and crimes revisited: Linking the theory of routine activities to the empiricism of “hot spots.” Criminol- ogy, 29, 725–753. Sherman, L. W., Gartin, P. R., & Buerger, M. E. (1989). Hot spots of predatory crime. Criminology, 27, 27–55. Smith, W. R., Frazee, S. G., & Davison, E. L. (2000). Furthering the integration of routine activity and social disorganization theories: Small units of analysis and the study of street robbery as a diffusion process. Criminology, 38, 489-523. 34 • Justice Research and Policy