The Influence of Survey Confidentiality and Construct Measurement in Estimating Rates of Childhood Victimization Among Navy Recruits

Surveys that directly ask participants whether they have been abused are likely to derive different results than studies that ask about specific childhood sexual experiences or about experiences with specific types of parental aggression that participants may or may not personally define as abusive. Rates are also likely to be affected by the level of confidentiality of the survey. We compared results from the nonconfidential military Sailor's Health Inventory Program (SHIP; Mittelman, Plunkett, & Bayer, 1998) with the results of the Survey of Recruits' Behaviors (SRB; Merrill, Newell, Hervig, Booth-Kewley, Patriarca, & Gilman, 1998). The SRB included both confidential and anonymous survey conditions. Rates of self-reported childhood abuse varied markedly. SHIP rates were lowest, with 5% of the participants identified as victims of abuse. Using SRB behavioral definitions under anonymous survey conditions resulted in the highest rate of 63%. Data derived using self-definitions and behavioral definitions independently accounted for variability in participants' symptoms of depression, anxiety, and sexual difficulties.