“…Aithough some empirical research comparing the play of abused and nonreferred children has supported the assumption that abused children are more likely to engage in sexualized behavior with AD dolls than are nonabused children (e.g., August & Forman, 1989; Jampole & Weber, 1987; White, Strom, Santilli, & Halpin, 1986), other studies have not found differences between these groups (e.g., Cohn, 1991;Kenyon-Jump, Bumette, & Robertson, 1991). Unfortunately, some of this research suffers from methodological constraints, such as the lack of information regarding the reliability of behavioral coding, small sample sizes, and lack of experimenter unawareness of group (e.g., abused vs. nonabused) membership (for methodological criticisms, see Koocher et al, 1995;Levy, 1989;Maan, 1991;Sagatun, 1991).…”