The effects of method of free report interview on the accuracy and completeness of children\u27s recall of information about sexual abuse

The purpose of this study was to determine if there were differences between three methods of interviewing children: using anatomically correct dolls, drawings, or verbal descriptions. The above techniques were examined as they relate to the accuracy and completeness of children\u27s recall of information about sexual abuse in a free report interview format.A non-clinical population of 45 female subjects (ages nine and ten) were drawn from a local public school.The target information was a commercially available video-tape designed to educate children about prevention of sexual abuse. Subjects viewed the film in groups and were interviewed individually in one of the three interviewing conditions. Interviews were conducted by trained, independent interviewers between four and six hours later. A standard protocol was utilized for the instructions and for the introduction of the dolls and drawing materials. After a warm-up period, the subjects were asked to tell the interviewer everything they could remember that happened in the film. Using video-tapes of the interviewers, trained, independent raters calculated the number of correct and incorrect statements made by each subject.A 3 x 3 factorial MANOVA was conducted with method of interview and interviewer effect as the two independent variables, and number of correct and number of incorrect responses as the two dependent variables.Results indicated that both the dolls and drawing conditions yielded higher scores than the verbal condition for number of correct responses. However, the differences in group means were not great enough to establish statistical significance. Incidental findings were that over one third of the subjects avoided dealing with information regarding intra-familial sexual abuse, that the method of play in which subjects performed best was associated with longer interview periods, and individual differences among subjects\u27 scores were extremely large