|
Session Title: Use of Fidelity Scores in Measuring Outcomes for Children Involved in the Child Welfare System
|
|
Panel Session 299 to be held in INDEPENDENCE on Thursday, Nov 11, 1:40 PM to 3:10 PM
|
|
Sponsored by the Human Services Evaluation TIG
|
| Chair(s): |
| Madeleine Kimmich, Human Services Research Institute, mkimmich@hsri.org
|
| Abstract:
Human Services Research Institute (HSRI) explored the use of model fidelity at the case level, using child fidelity scores as a covariate in outcomes analyses within the context of a five-year evaluation. This evaluation examined three strategies implemented across multiple Ohio child welfare agencies: Family Team Meetings, Supervised Visitation, and Kinship Supports. Fidelity to the strategies was used to help explain individual child outcomes via two statistical approaches: regression and ANOVA. By using these two techniques, evaluators were able to compare means of each fidelity group, as well as use fidelity as an independent variable in predicting child outcomes. In testing whether a very high level of fidelity is actually needed in order to achieve the desired results, evaluators are able to learn about effective practice even when high model fidelity is not evident across all areas. This becomes increasing important in complex service delivery environments, such as child welfare.
|
|
Integrating Child-level Fidelity Scores Into an Outcomes Analysis: An Evaluation of Family Team Meetings
|
| Erin Singer, Human Services Research Institute, esinger@hsri.org
|
|
This presentation examines fidelity measurement in a child welfare intervention, Family Team Meetings (FTM). FTM is a method for engaging family members and other people who can support the family for shared case planning and decision making. Seventeen counties in Ohio implemented this intervention over five years, with varying degrees of model fidelity both across counties, and within counties. Evaluators were still able to learn about effective practice by using child-level fidelity scores as an independent variable in a regression analysis to predict child level outcomes. Outcomes explored include increased permanency, increased reunification, shorter time in placement, and a decrease in subsequent reports of abuse and neglect and/or re-entry into care. An increased understanding of effective FTM practice, as well as how to integrate fidelity into an outcome analysis emerged. An overview of this process and the results will be discussed.
|
|
|
The Predictive Utility of a Child Level Measure of Fidelity: Its Relevance to the Understanding of Outcomes Associated With Enhanced Visitation Practices.
|
| Linda Newton-Curtis, Human Services Research Institute, lnewton@hsri.org
|
|
For those families who are involved with child welfare and for whom reunification is the case plan goal, it is generally understood that regular visits are critical for maintaining and improving the parent-child relationship. Twelve Ohio counties developed a visitation strategy in which supervised weekly visits, lasting a minimum of one hour were enhanced with the introduction of a ‘structured activities’ component. This component was expected to provide a mechanism through which visits might become more therapeutic, such that parenting skills could be improved thus potentially optimizing short and long-term family outcomes. An overview of the steps taken to measure the fidelity of this strategy at the child level is described followed by a discussion of the predictive utility of a child-level measure of visit fidelity in order to assess outcomes such as exit to reunification, length of stay in out-of-home care, and number of subsequent abuse and neglect reports.
| |
|
Kinship Support Index: Do More Intensive Programmatic Efforts Result in Better Outcomes for Children?
|
| Kim Firth, Human Services Research institute, kfirth@hsri.org
|
|
This presentation explores use of an index categorizing the degree of kinship support practices in county child welfare agencies participating in ProtectOhio. During this waiver, six counties focused on enhancing kinship placement supports via a practice model including increased staffing, recruitment of kin caregivers, and provision of hard goods and services. Because many of these kinship support efforts were also being utilized in other project counties, evaluators chose to index all participating counties on key kinship support elements, providing a measure of fidelity to the kinship model as well as an assessment of closeness to the model for non-strategy counties. These county-level index rankings were utilized as an additional variable in outcomes analysis for a sample of children in kinship care during the project period. Methodological challenges and outcomes findings will be discussed, providing lessons for the field on kinship support measurement and application of fidelity measures to outcomes analysis.
| |