GAL's were randomly selected from lists in each of the 23 counties and asked to report on two child abuse or neglect cases. In attempting to determine the effectiveness of GAL's, an important issue was the definition of effectiveness. The study adopted a criterion-referenced or procedural definition of effectiveness, as opposed to one that would examine outcome. Effectiveness was defined as "the degree to which the GAL performed the five roles identified as essential to GAL work and the related tasks and activities." The five roles considered are factfinder and investigator, legal representative, mediator and negotiator, case monitor, and resource broker. Findings on GAL backgrounds encompassed recruitment, previous experience, specialized training, experience as a GAL, timeliness of appointment, caseload and hours worked, compensation, supervision and evaluation, and case placement goals. The findings show that no GAL model studied was consistently superior to the others across all five GAL roles. The findings suggest that an optimal approach may involve having a GAL who either has or has access to the expertise and resources of attorneys, lay volunteers, and caseworkers to perform the broad range of functions and services contained in the definition of the child advocate. Appended report on Phase I of the study and data-collection instruments