Guidelines developed by the American Professional Society on the Abuse of Children are also considered. Issues considered include confidentiality, protection from harm, and bias. The discussion focuses on conjoint interviews between a child and an accused parent, the evaluation of a child for the attorney representing the accused, and agreeing to perform an evauation for one side of a custody/visitation dispute. The analysis concludes that ethical child sexual abuse evaluations do not guarantee that the truth can be discovered. Evaluation methods may be flawed. Data sources may be limited. Inconsistencies may resist resolution. No scientific or legal method exists to guarantee that the truth will be known. All that an ethical evaluation can promise is that the evaluation will be objective, unbiased, and as fair as possible to all parties. This approach will have the best chance of protecting child victims. Case examples, tables, notes, and 37 references (Author abstract modified)