Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Pacific L (Hyatt Regency San Francisco)

Evaluating Community Knowledge of a Registered Sex Offender in the Neighborhood

Sarah Craun, PhD, University of Tennessee, Knoxville.

Purpose: While sex offender registries remain a publicly popular policy, little academic research has been conducted to determine if the public is actually using the registries and specifically, if residents are aware of the sex offenders in their neighborhoods. This study's purpose was to determine how many residents who live close to a registered sex offender knew about the offender. Furthermore, the aim also included determining if there are any individual or neighborhood level predictors of knowledge of a registered sex offender. Social Disorganization Theory provided the basis for the selection of the neighborhood variables.

Methodology: Respondents in an urban county of North Carolina completed a self-administered, mailed survey on community safety. Respondents were asked if anyone in their neighborhood had been arrested for certain crimes, one of which was a sexual offense. Respondents were either in an investigational group, where they were randomly selected from residential parcels within one-tenth of a mile of a sex offender, or a control group, where respondents were living at least one mile away from any registered sex offender. There were a total of N=631 respondents from both groups, which constituted a 45% response rate. Neighborhood level data was obtained for each respondent from the US Census at the census tract level. Neighborhood level data was reduced using factor analysis with oblique rotation. Hierarchical Linear Modeling with a Bernoulli distribution was used to simultaneously consider the neighborhood level factors and individual level variables on knowledge of a registered sex offender.

Results: The research demonstrated that 31% of those who live within one-tenth of a mile of a registered sex offender are aware of a sex offender in the area. To ensure, that this is not representative of the population as a whole, a comparison between the control group and investigational group was made. Only 2.3% of respondents in the control group indicated that someone in the neighborhood had been arrested for a sex crime, which represents a statistically significant difference (chi-squared (1, N=603) = 93.52, p<.000). An analysis using hierarchical linear modeling found that both individual and neighborhood variables could predict knowledge of a sex offender. Specifically at the individual level, knowledge about Megan's Law, knowledge of crime in the community, knowledge of a fictitious crime policy and the number of sex offenders within one-tenth of a mile of a respondent were all related to an individual's knowledge of a sex offender. At the neighborhood level, the factor representing Hispanic immigrants demonstrated a statistically significant inverse relationship with knowledge of a sex offender. No other neighborhood level factors were significant.

Implications for Social Work Policy and Practice: Having less than one-third of residents who live close to a registered sex offender aware indicates that the policy is still struggling to meet its objective. Policy makers need to consider the implementation of this policy if the goal is increased community awareness. The predictors also demonstrated areas for possible intervention points for community organizers or other macro level social workers.