Document Type


Degree Name

Doctor of Social Work (DSW)


Social Work


Faculty of Arts

First Advisor

Dr. Lea Caragata

Advisor Role



Objective: In this paper, child welfare decision-making is critically assessed in relation to Beck’s (1992) concept of ‘risk society’. Three key decisions made by child welfare workers during the initial investigation are examined in order to illustrate how risk influences the type of investigative approach used, the determinations about child maltreatment, and the services provided to children and families and to link theory with policy and practice.

Methods: The three exploratory studies all utilize secondary data from several cycles of the Ontario Incidence Study, which collects information directly from frontline child protection workers about incidence of reported maltreatment. Two studies utilize bivariate and multivariate analyses of maltreatment only investigations and risk only investigations respectively, to explore the profile of children and families investigated for various forms of maltreatment and for risk of future maltreatment where no maltreatment has been reported and to explore factors influencing worker decisions to substantiate maltreatment and risk. The third study utilized Chi-squared Automatic Interaction Detector (CHAID) analysis in addition to bivariate analyses to examine the decision about the type of response to use.

Results: The results indicate that despite child welfare policies introduced over the past 10 years in attempts to balance the focus on risk, the overall rates of substantiated maltreatment and risk only investigations have not changed. The research also revealed that when controlling for multiple factors found in previous studies to influence substantiation decisions, the workers’ perception of the future risk of maltreatment emerged as the strongest influence on their decision to substantiate maltreatment. Families investigated for future risk of maltreatment were found to be different in several important ways from families investigated for maltreatment and factors that influenced workers’ decision to substantiate risk and maltreatment were also different. Children who were maltreated received fewer services from child welfare and were referred to outside services less frequently than children who had not been maltreated but were perceived to be at future risk of maltreatment. Only 10% of all investigations in Ontario were cases with significant protection concerns yet only 30% of these cases receive a traditional forensic investigation (most receive a customized response intended for less serious cases). Exposure to intimate partner violence was the factor which had the most significant influence on the workers’ decision regarding the type of investigation but surprisingly severe physical harm and sexual abuse did not emerge as significant factors.

Conclusion: Child protection practice and decision making is complex and risk discourses have had a significant impact on both. Despite policy changes introduced over the past ten years to mediate the negative impact of risk technologies on child welfare decision-making, several risks of risk emerged in the findings including directing attention away from helping children who have already been harmed to a focus on children who are at risk of future maltreatment; contributing to a focus on blaming parents instead of attending to social issues; and reduced opportunities for successful engagement of families. Differential response does not appear to have been successful in addressing the changing needs of families serviced by the Child Welfare system in Ontario. These studies suggest that differential response has not assisted in reducing the number of families subjected to a child welfare investigation nor has it resulted in a tighter more precise classification of reports ensuring that investigations are used when most required. Taken together these studies provide a compelling argument for Ontario to rethink the current approach to both child safety and child and family well-being.

Convocation Year


Convocation Season