Psychometric Evidence That Paraphilia Is a Natural Kind

Is the category paraphilia a natural kind? That is, do different paraphilias share anything scientifically interesting or are they classified together because they are unusual and sometimes problematic? We investigated this question systematically in 11 samples of paraphilic males (N = 4,617) and 11 samples of control males (N = 1,494). Primary data consisted of responses to the 11-item Paraphilic Interests Scale. Contrary to our initial hypothesis, the scale mean was similar for paraphilic and control samples. Using logistic regression and the same items, we derived three highly correlated measures that robustly discriminated paraphilic and control samples (ds ranging from 0.86 to 0.92). These successful measures capitalized on the unanticipated fact that some items (especially those assessing transvestism and masochistic humiliation) were positively associated with membership in paraphilic samples, while others (especially those assessing voyeurism) were negatively associated with such membership. Subsequent analyses focused on one of the measures, the Paraphilic Interests Scale Contrast (PISC). Consistent with prior findings distinguishing paraphilias and homosexual orientation, PISC was not elevated among homosexual males compared with heterosexual males among the control groups. Within four paraphilic samples, PISC was positively associated with additional paraphilic phenomena. Results provide tentative support for both the proposition that paraphilia is a natural kind and the usefulness of PISC as a measure of paraphilia.