The characteristics of sexual abuse in sport: A multidimensional scaling analysis of events described in media reports

Abstract
Most research on sexual abuse has been conducted within family settings (Fergusson & Mullen, 1999). In recent years, following several high profile convictions and scandals, research into sexual abuse has also encompassed institutional and community settings, such as sport and the church (Gallagher, 2000; Woolfe, Jaffe, Jette, & Poisson, 2003). Research into sexual abuse in sport, for example, began with both prevalence studies (Kirby & Greaves, 1996; Leahy, Pretty, & Tenenbaum, 2002) and qualitative analyses of the processes and experiences of athlete sexual abuse (Brackenridge, 1997; Cense & Brackenridge, 2001, Toftegaard‐Nielsen, 2001). From such work, descriptions of the modus operandi of abusers in sport, and the experiences and consequences for athlete victims, have been provided, informing both abuse prevention work and coach education. To date, however, no study has provided empirical support for multiple associations or identified patterns of sex offending in sport in ways that might allow comparisons with research‐generated models of offending outside sport. This paper reports on an analysis of 159 cases of criminally defined sexual abuse, reported in the print media over a period of 15 years. The main aim of the study was to identify the nature of sex offending in sport, focusing on the methods and locations of offenses. The data were analysed using multidimensional scaling (MDS) as a data reduction method in order to identify the underlying themes within the abuse and explore the inter‐relationships of behavior, victim, and context variables. The findings indicate that there are specific themes that can be identified within the perpetrator strategies that include “intimate,” “aggressive,” and “dominant” modes of interaction. The same patterns that are described here within the specific context of sport are consistent with themes that emerge from similar behavioral analyses of rapist (Bishopp, 2003; Canter & Heritage, 1990) and child molester groups (Canter, Hughes, & Kirby, 1998). These patterns show a correspondence to a broader behavioral model—the interpersonal circumplex (e.g., Leary, 1957). Implications for accreditation and continuing professional education of sport psychologists are noted