This article presents the results of an evaluation of child protection mediation in five California counties. These counties used a variety of different mediation approaches, yet achieved similar results. Data retrieved for the study included a combination of court records, mediation files, parent surveys, and interviews with attorneys, child protection workers and other involved professionals. Analysis of this data found that parents felt "heard" at mediations and preferred mediation to a court hearing, that some form of settlement was reached in 90% of all mediated cases, that mediated treatment plans were more likely to reference specific services to be provided to the child, that mediation was more likely to lead to the parent acknowledging the need to cooperate with the service plan, and that compliance was greater for mediated cases.

Description of Study: Comparative study of child protection mediation in Los Angeles, Santa Clara, Tulare, Contra Costa, and Sacramento Counties in California.

Method: Reviewed court files for both mediated and non-mediated cases in all five counties. The control groups in Tulare and Santa Clara counties were made up of cases comparable to those sent to mediation and that were scheduled for hearings on days when mediation was not available; in Contra Costa and Sacramento counties, the control group was made up of cases comparable to those sent to mediation, but which were filed in the year prior to the start of the mediation program. There was no control group for Los Angeles County.

Data was also collected from questionnaires completed by parents and forms completed by mediators.

Comparative: Yes

Comparison Groups: Some programs matched case characteristics of those cases sent to mediation to those that were not; other programs matched cases of those filed prior to the program to those mediated.

Sample Size: 499 parents; 968 mediator reports. Court files examined for 606 mediated cases and 223 non-mediated cases.

Variables Examined: Settlement rate, outcomes, compliance

Program Variables: In two counties, mediation occurred at the initial stages of case processing; two others referred cases to mediation at almost every stage of case processing. There was no information on mediation timing for the other county. Referral to mediation was made by the judge. The average mediation length was 90 minutes, but sessions of more than 2 hours were common. Attorneys were present for at least part of the session in all but one county, in which they were excluded.

Findings: 88% of mediated cases did not require a contested 6-month review hearing, compared to 53% of control cases. More than 90% of parents felt they had a chance to talk about issues important to them. Most parents preferred mediation to court. At 6 months post-disposition, mediated cases showed better compliance with the treatment plan than the control group More than 90% of cases at each site reached some form of settlement; 60-80% were full agreements.