Project Lead: Andrew Brankley
Collaborators: Karl Hanson, Raymond Knight, Calvin Cressman

This study examined whether pedophilia is best understood as a dimension or a discrete class (taxon). We considered pedophilia to include both physical sexual attraction to children as well as emotional attraction to children and childhood. Archival data from 901 adult males from a U.S. civil commitment treatment center for sexually dangerous persons were used. Three indicators were submitted to Meehl and Ruscio’s taxometric analyses: (a) a screening scale for pedophilic interests based on criminal history, (b) a composite indicator of pedophilic physical attraction, and (c) a composite indicator of pedophilic emotional attraction. All analyses supported a taxon. Compared with nontaxon members, taxon members had more interpersonal deficits, more neurodevelopmental perturbations, more sex-crime specific problems, and fewer problems related to criminality overall. Although previous research showed mixed results (some dimensional, some taxonic), the current study moves the balance of evidence toward considering pedophilia as a distinct disorder. 

Project Lead: Calvin Cressman
Collaborators: Andrew Brankley, Cory Gerritsen

The efficient and reliable identification of a problem increases the likelihood that it will receive clinical attention. Measures, such as the Revised Screening Scale for Pedophilic Interest (SSPI-2; Seto et al., 2017), provide a rapid screen for physical attraction to children because they only require readily available file information and can be quickly scored. The instruments commonly used to measure emotional identification with children usually require interview or self-report data to score. The purpose of this study is to develop a plausible series of efficient indicators of emotional identification with children for screening purposes and that can increase the ability of the SSPI-2 in identifying pedophilia. An item pool was constructed based upon existing measures of emotional identification that could reasonably be scored using file information (e.g., has history of child-oriented jobs and hobbies). File information was coded from the SSPI-2 development sample of 1,900 adult males convicted of a sexually-motivated offence against a child was recoded. Data will be fit to a three-level measurement model and redundant variables will be removed or combined utilizing Spearman’s Rho correlations, theory, and consensus. The remaining candidate variables will be dichotomized and used with the original SSPI-2 items to predict existing results from volumetric penile plethysmograph through logistic regression. Models will be compared based on AIC and BIC values. Success of a model including emotional identification would provide further evidence for the inclusion of the construct as part of pedophilia and guide future revisions of the SSPI-2.

Project Lead: Emma Holmes

Collaborators: Kelly Babchishin, Raymond Knight

This study examined…