The scientist-practitioner model is presented as a format for conducti
ng child sexual abuse evaluations. This model bases conclusions on emp
irically established relationships between data and the behavior of in
terest, rather than on subjective opinions. Using empirically derived
evidence, the scientist-practitioner defines child sexual abuse as a l
ife event rather than a clinical syndrome, relies on base rates of beh
avior for distinguishing and understanding differences between nonsexu
ally abused and sexually abused children, and considers issues of inst
rument sensitivity and specificity when utilizing assessment tools in
child sexual abuse evaluations. This model enlists safeguards to keep
the evaluator from inappropriately forming cause and effect associatio
ns between a child's single response (e.g., behavioral and emotional s
ymptoms, interactions with anatomically detailed dolls, drawing genita
lia) and the occurrence of an event (e.g., sexual abuse). (C) 1998 Joh
n Wiley & Sons, Ltd.