![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
Abstract
Guided by an environmental justice perspective of public policy, this study combines child sex offender registry information with US Census and geospatial data to examine the extent to which differences in the physical structure of Chicago neighborhoods (N = 872) affect the efficacy and fairness of the city's sex offender residence restrictions. Findings from mapping and regression analyses show that, because of differences in neighborhood size and the spatial distribution of prohibited sites (i.e., parks, schools, and daycares), Chicago's 500‐foot rule restricts the available residential space for child sex offenders to a much greater extent in disadvantaged neighborhoods (n = 175) than in affluent neighborhoods (n =130). The data also indicate, however, that despite legal proscriptions, child sex offenders continue several years after residence restrictions were first imposed to concentrate in disadvantaged neighborhoods and in relatively close proximity to prohibited sites. Implications for the offenders and communities are discussed.
Acknowledgments
This paper was strengthened by helpful comments from Katya Botchkovar, Mary Evans, Richard Tewksbury, Jim Short, Eric Silver, and four anonymous reviewers. Assistance with data collection was provided by Patrick Donnelly and Kenneth Williams.
Notes
1. Important exceptions include the research literatures that have developed around Newman's (Citation1973) Defensible Space, Wilson and Kelling's (Citation1982) “Broken Windows,” and Cohen and Felson's (Citation1979) Routine Activities Theory and the related notion of criminal hot spots (see Eck & Weisburd, Citation1995).
2. Referring to residence restrictions as “Jessica's Law” is somewhat of a misnomer, as the latter typically incorporates a range of additional provisions (e.g., registration, enhanced penalties for repeat child sex offenders, electronic and GPS monitoring for sex offenders on parole, etc.) at both the federal and state level.
3. For a review of journey‐to‐crime research on various types of sex offenders, see Beauregard, Proulx, and Rossmo (Citation2005).
4. Contrary findings are reported by Walker, Golden, and VanHouten (Citation2001), whose analysis of registry data from an Arkansas county indicated that sex offenders with child victims are significantly more likely than those with adult victims to live near child‐related facilities, such as schools, parks, and day care centers. Although the authors regard this finding as at least partially reflective of child sex offenders “putting themselves in close proximity to potential targets” (p. 30), the lack of data on characteristics of offenders, specific offenses, and neighborhoods precludes definitive conclusions regarding the nature of choice involved in the residential locations of child sex offenders versus those with adult victims.
5. This study suffers from several limitations, most notably the omission of key explanatory variables and the possibility of regression to the mean in their logistic regression models predicting upward or downward mobility.
6. When a sex offender moved into the home of one of the author's neighbors, several residents of this largely White, middle‐class neighborhood began collecting signatures to petition the neighborhood association to hold a special meeting in which the bylaws would be amended to state that sex offenders would no longer be permitted to live in the area. This proposed amendment is in addition to a city ordinance requiring sex offenders to live at least 500 feet from schools.
7. In Illinois, all sex offenders must register. If the offender is sentenced to probation, the duration of registration is 10 years from the date of conviction; if paroled or otherwise released from prison, the duration is 10 years from the date of release. Sex offenders deemed “sexually dangerous” or “sexually violent” must register for life.
8. The vast majority of these cases were located immediately outside of city boundaries; the remaining could not be properly geocoded due to a missing street suffix.
9. Of these child sex offenders, only six were residing in tracts that could not be matched to census boundary data.
10. A shapefile is a digital format for storing geospatial data, including geographic location and related attributes (e.g., area).
11. Unfortunately, the data do not permit exclusion of airports, commercial properties (e.g., shopping centers), and all non‐residential locations.
12. We exclude all cases for which no match could be found, as well as those involving “fuzzy matches,” place‐based ZIP matches, closest address matches, and direction corrected matches.
13. Nine parks could not be associated with a specific census tract and are excluded from all analyses.
14. Zero cells are addressed by adding a constant of 1 prior to taking the log.
15. While our measure of the 500‐foot buffer zones around parks and schools begins at the outer boundary of these objects, we were unable to obtain area data for daycares and thus use the centroids of these objects to construct buffer zones. Although all estimates of protected and available living space should therefore be viewed as conservative, daycares often are established in private and other small residences and are unlikely to occupy as much physical space as parks and schools.
16. Not surprisingly, excluded tracts tend to be smaller, less populated, and relatively privileged, with lower means for all indicators of neighborhood disadvantage. While excluded tracts generally have lower levels of residential mobility, there is virtually no difference in the average percent foreign‐born.
17. The size of parks also tends to be larger in disadvantaged neighborhoods (total area = 1165.53 acres; mean area = 10.41 acres) than in affluent ones (total area = 557.41 acres; mean area = 6.06 acres). Although the total land area of schools is larger in disadvantaged areas (554.21 acres v. 370.51 acres), their mean area is slightly larger in affluent neighborhoods (2.87 acres v. 2.83 acres).
18. The ratio of facilities in disadvantaged versus affluent neighborhoods is 6.1 for daycares, compared to just 1.5 for pars and 1.2 for schools. Based on census data and our review of the names of licensed owners, we suspect that the disproportionate representation of daycares in disadvantaged neighborhoods reflects both greater need for such services (i.e., more children, fewer available adults) and the business opportunities they provide to unskilled men and women in these communities. Rent for private daycare providers also may be less expensive in these areas.
19. Although the higher density of housing in disadvantaged neighborhoods may imply relatively more housing options available to sex offenders per acre, the proximity of dense housing to restricted sites (i.e., parks, schools, daycares, victim residence, and other sex offenders) likely would result in even greater neighborhood differences in housing availability than we report based on land area calculations.
20. Analyses based on the entire sample of registered sex offenders, regardless of victim age, yielded virtually identical results (data not presented but available from the authors).
21. Sex offender residence restrictions did not exist in Arkansas until 2003; hence, Walker et al. (Citation2001) selected a 1,000‐foot distance requirement based on state laws mandating school safety and drug‐free zones. Currently, Arkansas law bars all sex offenders who are determined to be dangerous from living within 2,000 feet of schools and day care centers (Ark. Code Ann § 5–14–128; see also Weems v. Little Rock Police Dep't, Citation2006).
22. The Lagrange Multiplier‐Error test value is 1.34 (1 df; not significant), providing evidence that little spatial autocorrelation remains in the residuals after the spatial lag term is introduced into the model.
23. Note that our analyses cannot speak to environmental justice arguments that focus more on the quality and cultural specificity of built environments than on basic access (see Frumkin, Citation2005).
24. The Illinois Department of Corrections lists only 233 child sex offenders in Chicago as being under parole or mandatory supervised release. Assuming the average duration of parole is 3 years, we may conclude that the remaining offenders in our sample either were sentenced to probation or have completed parole and thus were released from prison prior to January 2006, when parks and daycares were added to the long‐standing residence restrictions around schools.