Functioning of Standardized Self-Report Measures for Caregivers with Active Child Welfare Service Cases

Abstract

The purpose of this dissertation was to assist in the development of an assessment system that supports caseworkers in Child Welfare Service (CWS) in making informed decision regarding the families they are serving. Caseworkers CWS have a difficult job of protecting and promoting the well-being of vulnerable children. To accomplish this task, they need assessment tools that both predict future maltreatment and provide guidance on family functioning, service need, and treatment progress. This dissertation presents two studies that examine the results of seven standardized self-report assessments obtained from 318 caregivers involved with CWS (235 families) whose children are in their care, and attempts to use these data to address some of the assessment needs in CWS. Study 1 examined the standardized assessment results and used survival analysis to determine whether the assessments were independently predictive of future child maltreatment and whether they added predictive value to the Structured Decision Making (SDM) tool used by CWS. Due to the coercive nature of caregivers involvement with CWS, the data were analyzed separately for caregivers who were categorized as reporting non-defensively and defensively on the Parent Stress Index. The assessment results for caregivers who were categorized as non-defensive provided insights into the families' struggles and added predictive value both independently and additionally when included in survival analysis with the SDM. For defensive caregivers, the assessments did not appear to provide insights into their struggles, and scores were not predictive of future child maltreatment. The results indicated no significant difference in survival time between the defensive and non-defensive caregivers. Furthermore, caseworkers appeared to struggle with the assessment of the defensive responders, as indicated by the SDMs lack of predictive value for these families. Study 2 examined how the data from the non-defensive responders could be used to develop a self-report multidimensional assessment of caregivers involved with CWS. Item Response Theory (IRT) was used to examine the functioning of items on two of the administered assessments. The results indicated both adequate item difficulty and item discrimination parameters. IRT was used to reduce the assessment length, resulting in shorter scales with similar predictive validity. Lastly, the study investigated how IRT can assist in the development of assessment tools that address assessment needs of CWS including potentially being resistant to the defensive responding of caregivers.