Child maltreatment: The protective effect of neighborhood social cohesion

Date de publication: 

Maguire-Jack, K. & Showalter, K. (2016). The protective effect of neighborhood social cohesion in child abuse and neglect. Child Abuse & Neglect, 52, 29-37.

Revu par: 
Sydney Duder

Earlier studies have found that neighborhood structural factors (e.g., poverty level, population density, crime rate) were significantly related to child abuse and neglect; this present study examined the possible protective role of social cohesion (i.e., strong bonds with neighbours). Data came from the Franklin County Neighborhood Services Study in Ohio; a convenience sample of 1326 parents from participating clinics and childcare centers completed brief paper-and-pencil anonymous questionnaires in clinic waiting rooms while waiting for appointments. Questionnaires with complete responses (N = 896) were retained for analysis.

·         Possible cause: Perceived neighborhood cohesion, measured by the 5-item Social Cohesion and Trust Scale (Sampson et al., 1997). Item responses on 5-point Likert scales were averaged to provide an overall cohesion score.

·         Expected outcomes:  Four subscores from the Conflict Tactics Scale - Parent-to-Child version (CTS-PC, Strauss et al., 1998): two neglect categories (child’s basic needs & caregiver’s mental health or substance abuse) and two physical abuse categories (corporal punishment & severe assault). Responses were coded as recommended by the authors.

·         Covariates: Race (Black, Hispanic or other), married or not, high school education or less, number of children, number of economic hardships experienced in the past year, and Parent Stress index (PSI-SF, Abidin, 1990, 1995).

High neighbourhood social cohesion scores were found to be significantly associated with lower levels of neglect of children’s basic needs; no significant effect was found on neglect due to caregiver’s mental health or substance abuse, or on either of the two physical abuse variables. Significant increases across all maltreatment categories were associated with parenting stress, economic hardship and being black. One unexpected finding was an association between lower education and lower levels of corporal punishment.

The authors argue that these findings have important implications for policy makers; to prevent neglectful behaviors, it would be important to build programming that supports trust and belongingness. They provide a potentially useful list of possible strategies that could be helpful in doing this.

Notes méthodologiques: 

Negative binomial regression was used for the full abuse and neglect scales, and logistic regression was used for the analysis of specific subcategories.  This study appears to have been generally carefully done, but there were methodological limitations, for example cross-sectional design, so effects could not be tracked over time; convenience sample, from one large urban county in Ohio—neighbourly behavior might be different in rural or suburban settings; all subjects were WIC clients so income level was generally low; data were entirely self-report; perceptions of the supportiveness of neighbors could have been influenced by the characteristics of the respondent. All of the above could limit the generalizability of these findings. The authors stress the importance of future longitudinal research, including more geographically diverse populations.