Family Group Decision Making

Nixon, P., Burford, G., & Quinn, A. (with Edelbaum, J.). (2005, May). A survey of international practices, policy & research on family group conferencing and related practices. Retrieved December 9, 2008, from http://www.americanhumane.org/assets/docs/protecting-children/PC-FGDM-practices-survey.pdf 

This report summarizes responses from an online survey of international practices, policy and research efforts on the use of family group conferencing and related practices. Carried out in early 2004, it summarizes in thematic form the responses from 225 respondents in 16 countries.

Keywords: family participation, child welfare, children’s mental health, education, youth justice, delivery system, qualitative data, quantitative data

Northwest Institute for Children and Families, & Catalyst for Kids. (2007, May). Finding our roots: Family group conferencing in Washington. Seattle, WA: Author.

The Finding Our Roots program gave family group conferences to children ages 10-18, as well as quarterly permanency team meetings. The study gave priority to children of color, as they are disproportionately represented in the child welfare system. For instance, 8 percent of the children in King County, Wash., are African-American or Native American, but they make up 33 percent of children removed from their homes. The 550 meetings held averaged 9.8 participants, with greater participation from the maternal family. Service provider attendance averaged at 3.5 per meeting. Through the conferences, nearly all children and youth gained closer and more positive relationships with their families and communities and an increase in supports.

Keywords: United States, U.S., Washington, child welfare, child permanency, program evaluation, process evaluation, qualitative data

O’Sullivan, B., McKinney, A., & Gallagher, S. (2002). Family group conference pilot project. Belfast, Northern Ireland: North Western Health Board.

This evaluation of an Irish pilot project assesses the applicability of family group conferencing as a partnership-based model to promote and enhance child-centered and family-focused practice. The authors were particularly interested in assessing what modifications to the family group conferencing model were necessary for application in their region. The evaluation draws on cross-sectional postal questionnaire surveys and semi-structured interviews with family members, children and professionals who participated in the first six of 10 family welfare conferences. High levels of satisfaction are reported from family and professionals about the experience, including the plans for the children, the provision of information to the family, the degree to which the meetings were child-focused and the fit between the practice principles and other services employing partnership and inclusive practices.

Keywords: Republic of Ireland, Ireland, qualitative, quantitative, process evaluation, implementation stage

Pennell, J. (2006). Restorative practices and child welfare: Toward an inclusive civil society. In B. Morrison & E. Ahmed (Eds.), Restorative justice and civil society [Special issue]. Journal of Social Issues, 62(2), 257-277.

Given that federal reviews of child welfare in the United States have found that involving families in decision making is related to stabilizing children’s placements and promoting children’s well-being, a North Carolina study examined the extent to which inclusive participation was achieved at 27 family group conferences held to address child welfare issues. Three indicators of participatory decision making were assessed: 1) greater attendance of family group members than service providers was assumed to reflect a stronger family voice at the deliberations; 2) family group members’ satisfaction with the process and outcome was assumed to reflect consensual decision making; and 3) democratic decision processes used during the family private time were assumed to reflect inclusive participation. The study found that all three indicators supported that family group conferencing promotes participatory decision making: 1) the 221 family group members outnumbered the 115 service providers; 2) the family group members were largely satisfied with the conference process and its resulting decision; and 3) the most influential decision processes during the family’s private time were consensus, inspiring (following a trusted leader) and bargaining. Far less common were ordering, voting, avoiding and manipulating. Satisfaction with the resulting plan, but not the process, was lowered when bargaining was employed. Conferences with inadequate preparations were more likely to result in manipulation taking place during the family’s private time.

Keywords: U.S., North Carolina, child welfare, family participation, developmental stage, program evaluation, process evaluation, qualitative data, quantitative data

Pennell, J., & Burford, G. (2000). Family group decision making: Protecting children and women. Child Welfare, 79(2), 131-158.

This study of family group conferencing in Newfoundland and Labrador was conducted in 1993-1996, in three culturally diverse regions: Inuit, rural and capital city. Over an approximately one-year period, family group conferences were convened for 32 families, with 472 participants at their conferences, of whom, 384 were family group members and 88 were service providers. An average of one year after their conferences, 115 out of the 384 family group members were interviewed. Whether the plans were completed in their entirety or in part, the majority of the interviewees reported that the family was better off because of the family group conference. In particular, interviewees thought the family was better off because the conference strengthened positive relationships and enhanced their sense of being family. In addition, interviewees reported that children from the project families suffered less maltreatment and had better parenting. A review of child welfare files found a reduction in indicators of child maltreatment and domestic violence for the 32 project families and a moderate increase for the 31 comparison families. After the conference, child protection workers were less likely to make emergency visits to project families than to comparison families because they knew the project families better. The number of child placements stayed relatively constant for the comparison families and fell for the project families. Small in number but nevertheless troubling, children and youth abusing mothers appeared to persist for project families and rose for comparison families. Mother abuse was linked with child self-harm (i.e., self-mutilation and attempted suicide) for both groups. Family group conferencing plans became quickly dated in especially chaotic family situations.

Keywords: Canada, Newfoundland and Labrador, child welfare, domestic violence, family violence, developmental stage, program evaluation, quasi-experimental design, qualitative data, quantitative data

Sawyer, R. Q., & Lohrbach, S. (2008, January). Olmsted County Child and Family Services: Family involvement strategies. Rochester, MN: Olmsted County Child and Family Services.

Drawing on multi-year (1996-2007) secondary statistics and survey results (2004-2006), the authors associate the increased use of a range of family involvement strategies with a variety of positive outcomes, including reduced use of court, timely case resolution, increased use of kin and family placements, reduction in new findings of abuse and neglect, and positive safety outcomes.

Keywords: United States, U.S., Minnesota, Olmsted County, child welfare, child safety, quantitative data, qualitative data, developmental stage, pre/post-comparisons

Sieppert, J., & Unrau, Y. (2003). Revisiting the Calgary project evaluation: A look at children’s participation in family group conferencing. Protecting Children, 18(1 & 2), 113–118.

This brief article re-analyzes children’s participation data from an earlier formative evaluation of a pilot family group conference project in Calgary, Alberta, Canada. Based on nonparticipant observation with 13 children, the study found children’s levels of participation varied greatly between different family group meetings. The authors emphasized that attendance is not equivalent to participation and the meaningful engagement of children in family group conferences and follow-up meetings can meaningfully add to the development of a plan and its monitoring. The study found that some family members and professionals inadvertently mitigated children’s participation in the meetings. The authors concluded by suggesting that how children are included in the family group conference process must be better understood and addressed.

Keywords: Alberta, Canada, qualitative data, child participation, process evaluation

Staples, J. (2007). Knowle West Family Group Conference Project: Evaluation report. Bristol, England: Barnardo’s/Knowle West Neighborhood Renewal

This report looks at the experiences of those participating in family group conferences delivered by the commissioners of the report, during 2005 to 2007. Participants who gave consent were interviewed by telephone post-conference; there were some 70 respondents. The evaluation captured data from the original referral and then set the later outcomes against the original referral data. The evaluation sought to arrive at both findings about the effect of the service and recommendations for service development. Respondents positively rated the service and a majority said that their situation had improved following the family group conference. The respondents cited improved family relationships and improved home-school relationships as outcomes, alongside predicative commentaries that suggested that the family group conference prevented further breakdown of the family situation. Unanticipated outcomes included the surprise of professionals at the capacity within the family for support, enhanced self-esteem of participants and increased knowledge within and about the family.

Keywords: U.K., United Kingdom, Knowle, child welfare, child well-being, delivery system, implementation stage, program evaluation, process evaluation, qualitative data

Sundell, K. (2003). Family group conferences in Sweden -- continuing social services programs for children and parents [English Summary]. Retrieved December 9, 2008, from http://www.sos.se/socialtj/cus/cuse/empstude/famrad99sum.htm 

This summary from Sweden's National Board of Health and Welfare reports on an evaluation of an early trial of family group conferences in 10 local authorities across Sweden, later published by Sundell and Vinnerljung (2004). The summary reports a comparison between children involved in family group conferences (N = 97) and a random sample of children receiving traditional child welfare services (N = 142), following them over a three-year period, post-conference. Children receiving a family group conference were somewhat more frequently rereferred for suspected abuse, but neglect rereferrals were the same for both groups of children. More of the family group conference children were placed with kin, but the majority was nevertheless placed in nonrelative care (79 percent). The children receiving a conference were somewhat more frequently continuing to receive interventions after three years, but interventions more frequently progressed from placements to noncustodial care. The researcher suggests that the type of model applied (i.e., family group conference or conventional models) minimally explained the variance in outcome measurement. The author recognizes both the limitations of the research methodology and the challenges faced by child welfare in resolving issues of child maltreatment. In regards to the quasi-experimental design, Sundell identifies significant difficulties in determining the impact of family group conferencing as the analysis may not have adequately controlled for the conference referrals having more indicators for child maltreatment than the comparison groups. Poor follow-through of the conference plans was also identified as a possible explanation for the results. Overall, the author points to the challenges confronting child welfare in supporting good outcomes for children, noting that the children receiving traditional interventions in the study (the comparison group) demonstrated limited success.

Keywords: Sweden, child welfare, child safety, quasi-experimental design, outcome evaluation, quantitative data

Sundell, K., & Vinnerljung, B. (2004). Outcomes of family group conferencing in Sweden: A 3-year follow-up. Child Abuse & Neglect, 28, 267-287.

This article reports on an evaluation of an early trial of family group conferences in 10 local authorities across Sweden. A comparison was undertaken between children involved in family group conferences (N = 97) and a random sample of children receiving traditional child welfare services (N = 142), following them over a three-year period, post-conference. At the three-year follow-up, 69 percent of the family group conferencing group had been the subject of at least one new child maltreatment report and 60 percent were substantiated. Out of 57 percent of the comparison group with at least one new report, 40 percent were substantiated. Both groups had low levels of rereporting by extended family members (11 percent). On average, the family group conferencing children received more services and were more often placed in foster or residential care. They also had more previous child protection investigations than did the nonconference group (71 percent versus 51 percent) and social workers rated the family group conferencing cases as more serious than the comparison group.

Data analysis found that during the follow-up period, social background and problems had a significant effect on continued child protection contact. Further, the model of family group conferencing as compared to conventional child protection interventions had a significant effect on continued child protection involvement but accounted for only a small part of the variance, from 0 to 7 percent. In terms of the process, the family group conferences were well-attended by family members (75 percent) and children's participation in the conferences was high, with 41 percent of children 9 years and younger and 94 percent of children over 10 years participating, The children reported that their views were respected (77 percent), expressed satisfaction with the plan (89 percent), and preferred family group conferencing as an effective solution-focused method (86 percent).

Keywords: Sweden, child welfare, child safety, outcome evaluation, quantitative data, quasi-experimental design