An exploration of racial and ethnic differences in paraphilic interests and behaviours among a nonoffending sample

ABSTRACT

Paraphilic interests are often associated with sexual offending and recidivism, and therefore, accurately identifying paraphilias is crucial. Recently, there has been increased recognition of diversity as it relates to sexual offending; however, few studies have examined race/ethnicity as it pertains to paraphilias, and those that have are often limited to forensic samples. Given that important decisions such as post-incarceration civil confinement are, in part, based upon paraphilia diagnoses, information concerning population base rates of paraphilic interests and behaviours, and how they may vary by race/ethnicity, is needed. The current study explored racial/ethnic differences in paraphilic interests/behaviour among a non-offending U.S. sample of 2,292 participants over age 18, using a measure adherent to the DSM-5-TR paraphilia criteria. The findings revealed group differences, highlighting the need for further research examining these differences.

PRACTICE IMPACT STATEMENT

The results provide information regarding potential racial/ethnic differences in self-reported paraphilic interests and behaviour among a general population, non-offending sample. This information will assist clinicians who evaluate and diagnose paraphilic interests and disorders and represents a contribution to the larger goal of informing best practices for culturally competent risk assessment and evaluation, treatment, and prevention of sexual abuse.

KEYWORDS:

• Paraphilic interests
• paraphilic behaviours
• general population
• diversity
• ethnic differences
• racial differences

Diagnostic manuals, such as the Diagnostic and Statistical Manual of Mental Disorders 5th Edition Text Revision [DSM-5-TR; American Psychiatric Association (APA), 2022] and the International Classification of Diseases 11th Revision (ICD-11; World Health Organization, 2021), contain a category of disorders related to non-normative or atypical sexual interests, though it should also be noted that what is considered normal and abnormal can vary across time and culture and therefore requires ongoing examination (Ayonrinde & Bhugra, 2015; Bhugra et al., 2010). According to the DSM-5, these so-called paraphilic interests (or paraphilias), refer to “an intense and persistent sexual interest other than sexual interest in genital stimulation or preparatory fondling with phenotypically normal, physically mature, consenting human partners” (APA, 2022, p. 779). Paraphilic interests are known to be associated with some sexual offences and sexual recidivism; in fact, research has shown that paraphilic arousal is one of the strongest predictors of sexual reoffending (Akerman & Beech, 2012; Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2005; Mann et al., 2010; Seto, 2019; Seto et al., 2023). Therefore, the effective identification, evaluation, and treatment of paraphilic disorders is of utmost importance.

Recently, there has also been increased recognition of and interest in the role that racial and ethnic diversity may play when it comes to understanding criminal behaviour, including the overrepresentation of ethnic minorities in the criminal justice system (Bonczar, 2003; Fix et al., 2017; Piquero, 2008). Race refers to physical differences that groups and cultures consider socially significant (e.g. African American or Black, Asian, Native American, European American or White), while ethnicity refers to shared cultural characteristics such as language, beliefs, and traditions (e.g. Hispanic, Latinx; American Psychological Association, 2020).

Specific to sexual offences, research has suggested that there may be racial and ethnic differences in the characteristics of sexual offending behaviour (Ikomi et al., 2009; Lee et al., 2020) and that the predictive validity of risk assessment tools may vary among different racial and ethnic groups (Martinez et al., 2007). Yet, to date, limited studies have empirically examined the role of race and ethnicity as they pertain specifically to paraphilic interests (Burton & Ginsburg, 2012; Fix et al., 2017; Lee et al., 2020). The identification and understanding of potential racial and ethnic differences in the presentation of paraphilic interests would serve to inform best practices for culturally competent assessment and treatment, and prevention of sexual abuse. Thus, the present study aims to examine differences between racial and ethnic groups in self-reported paraphilic interests and behaviour using a large, non-offending, United States (U.S.) sample.

DSM-5-TR paraphilic interests and disorders

Within the DSM-5-TR, there are eight non-normophilic or “paraphilic” categories, which are divided into “anomalous activity” and “anomalous target” preferences. Anomalous activity preferences are further subdivided into courtship disorders, which involve distorted human courtship behaviour (voyeurism, exhibitionism, frotteurism), and algolagnic disorders, which involve pain and suffering (sexual masochism, sexual sadism). Anomalous target preferences include those directed at other humans (pedophilia) and other-than humans (fetishism, transvestism). Generally speaking, to meet diagnostic criteria for a paraphilic disorder, there must be evidence that an individual experiences “recurrent” and “intense” sexual arousal lasting for at least six months, and that they have either acted upon the sexual urges with a non-consenting person, or that the sexual urges or fantasies cause clinically significant distress or impairment in at least one domain of the individual’s daily functioning (American Psychiatric Association, 2022 ). Thus, the mere presence of a paraphilic interest alone does not satisfy the criteria for a paraphilic disorder.

Prevalence of paraphilic interests and behaviours

Previous studies that have sought to examine the prevalence of paraphilic interests and behaviours reflect limitations in that they utilise paraphilia definitions not directly in line with the language of the DSM-5-TR, do not assess for all eight listed paraphilias, and/or they assess for paraphilic interests/behaviours but not disorders. Nonetheless, the extant literature has broadly supported the notion that paraphilic interests are relatively common amongst the general population (see Winters et al., 2023 for a review; see also Holvoet et al., 2017; Joyal, 2015; Wurtele et al., 2013). Joyal and Carpentier (2017) found among a sample of 1,040 adults in the general Quebec population that nearly half (45.6%) expressed a desire to engage in a paraphilic behaviour, and approximately one-third (33.9%) had engaged in a paraphilic behaviour. Bártová and colleagues (2021) found that 31.3% of men and 13.6% of women from a large representative online sample of the general Czech population endorsed at least one paraphilic interest, with 15.5% of men and 5% of women reporting multiple paraphilic interests; furthermore, 23.3% of men and 10.1% of women reported engaging in a paraphilic behaviour at least ten times in their life.

The DSM-5-TR has incorporated some information regarding the prevalence of paraphilic interests and behaviours, although these tend to reflect a range of research studies from specific countries (e.g. Canada, Australia), different populations (e.g. sexual offenders court-mandated to post-incarceration civil confinement, general population), and using a variety of outcome variables (e.g. interests, behaviours, disorders). Moreover, for most paraphilic diagnoses, the population prevalence of the disorder is listed in the DSM-5-TR as “unknown”. See Table 1 for descriptions of the paraphilia categories and a review of the available prevalence estimates as listed in the DSM-5-TR.

Table 1. DSM-5-TR paraphilia descriptions and estimated prevalence rates.

Paraphilia category	Description	Disorder prevalence	Interests/behaviours prevalence
Voyeurism	Observing an unsuspecting person who is naked, in the process of disrobing, or engaging in sexual activity	Unknown	In a Quebec Internet and phone survey sample, the lifetime prevalence of voyeuristic behaviours was reported to be as high as 34.5% (50.3% in men, 21.2% in women). A study of 1,346 incarcerated sex offenders from Austria found a prevalence of voyeuristic disorder of 3.7%.
Exhibitionism	Exposure of one’s genitals to an unsuspecting person	Unknown	In a Quebec Internet and phone survey sample, lifetime prevalence of exhibitionistic behaviours was reported to be 30.9% (32.6% in men, 29.4% in women).
Frotteurism	Touching or rubbing against a nonconsenting person	Unknown	Frotteuristic acts  … may occur in up to 30% of adult men in the U.S. and Canadian general population.
Sexual masochism	Being humiliated, beaten, bound, or otherwise made to suffer	Unknown	In Australia, it has been estimated that 2.2% of men and 1.3% of women had been involved in BDSM behaviour in the past 12 months.
Sexual sadism	Physical or psychological suffering of another person	Unknown	Among civilly committed sexual offenders in the U.S., less than 10% have sexual sadism disorder. Among a representative sample of the population in Australia 2.2% of men and 1.3% of women reported they had been involved in BDSM behaviour in the previous year. In a population-based sample in Finland, the lifetime prevalence for sexually sadistic behaviour was 2.7% among men and 2.3% among women.
Pedophilia	Sexual activity with a prepubescent child or children (generally age 13 or younger)	Unknown “but likely less than 3% among men in international studies”	None
Fetishism	Use of nonliving objects or a highly specific focus on nongenital body part(s)	No section	No section
Transvestism	Cross-dressing	Unknown	Fewer than 3% of Swedish men report having ever been sexually aroused by dressing in women’s attire.

Relationship between paraphilic interests and sexual offending

As noted previously, it has been well established in the literature that paraphilic interests are associated with sexual offences, as well as with sexual recidivism risk. While the relationship between paraphilias and sexual offending is not one-to-one (i.e. not all those with a paraphilia will commit a sexual offence, and perpetrators of sexual offences do not always have a paraphilic disorder), paraphilias have been identified as one of the primary motivations to commit sexual offences, and sexual offenders differ from other types of offenders and from non-offenders with respect to the presence of paraphilic interests (Seto, 2019). Two landmark meta-analyses, the first including 61 studies and 28,972 individuals convicted of sex offences, and the second including 82 studies and 29,450 individuals convicted of sex offences, both identified paraphilia indicators, also referred to as “sexual deviance” indicators, as the strongest predictors of sexual reoffending, with pedophilic interests being the single strongest predictor of sexual recidivism (Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2005). Subsequent research has continued to corroborate the link between paraphilia indicators and the risk for sexual recidivism (Hawes et al., 2013; Mann et al., 2010; Olver & Wong, 2006; Seto et al., 2023). For good reason, these studies have been conducted using samples of known offenders; however, this also means that the extent to which paraphilic interests apply to the general population (offending but undetected samples, as well as non-offending samples) is not well known.

Racial and ethnic differences in sexual interests and behaviours

To date, few studies have empirically examined the role of race and ethnicity with regard to sexual interests and behaviour in non-offending populations, but the existing studies do suggest possible differences in sexual norms. For example, Black adolescents report an earlier onset of sexual activity and of sexual maturity (development of secondary sex characteristics) than White adolescents (Fix et al., 2019; Herman-Giddens, 2006), whereas White adolescents have been shown to be more likely to engage in “sexting” (transmitting sexually explicit images or messages electronically) than Black adolescents (Temple et al., 2012). However, apart from these findings, much of the literature on racial and ethnic differences in sexual interests and behaviour has focused on known offender populations, and often with a focus only on comparisons between Black and White individuals.

Although research shows that Black individuals tend to be disproportionately overrepresented in the criminal justice system overall, and for a variety of violent and non-violent offences (Bonczar, 2003; FBI, 2018; Fix et al., 2017; Piquero, 2008), this pattern has not always held for sexual offences, particularly among adolescents who offend (Burton & Ginsburg, 2012; Fix et al., 2017; Leguizamo et al., 2010; Craun & Kernsmith, 2006), and ethnic minorities overall have been underrepresented in the literature. There is also evidence of possible racial and ethnic differences in the specific types of sexual offences committed (e.g. research showing White individuals may have younger victims and are more often charged with sexual abuse, attempted sexual abuse, and/or online sex offences, while Black individuals may be less likely to have child victims and are more often charged with rape or attempted rape; Babchishin et al., 2011; DeHart et al., 2016; Fix et al., 2017; Lee et al., 2020), as well as in other demographic variables (e.g. education level, history of victimisation, psychiatric problems, medical conditions, suicide attempts; Burton & Ginsburg, 2012; Burton & Meehan, 2007; Dominguez et al., 2018; Fix et al., 2019; Murphy et al., 2001; Schaaf et al., 2019). In an Australian sample, Indigenous youth were overrepresented with regard to sexual offending; they represented 12%–14% of known youth who sexually offended, while comprising only 5% of the youth population (Allan et al., 2002). A recent study also revealed that Australian Indigenous youth who committed sexual offences also differed from their non-Indigenous counterparts in a number of other ways, with respect to systemic issues (e.g. familial antisocial attitudes/incarceration, engagement with antisocial peers, low socioeconomic status), as well as with respect to the circumstances and context of their sexual offences (e.g. substance use, age and relationship to victim, use of force; Adams et al., 2020). Research has also suggested that the predictive validity of risk assessment tools may vary among different racial/ethnic groups (Chu et al., 2012; Lee et al., 2020; Lee & Hanson, 2017; Molnar et al., 2022).

With regard to racial and ethnic differences in paraphilic interests specifically, research suggests there may be higher levels of paraphilic interests among White individuals who commit sex offences, compared to their Black counterparts. For instance, Leguizamo and colleagues (2010) found among a sample of individuals incarcerated for sex offences that White individuals were more likely to have child victims, a history of non-contact and fondling offences, and to have used pornography in the context of a sexual offence. Similarly, Lee and colleagues (2020) found among a sample of adult males convicted of sexual offences that Black individuals demonstrated fewer paraphilic indicators (e.g. significantly less likely to be diagnosed with pedophilia, to have had minor or male victims, or to have been involved in non-contact offences such as exhibitionism or voyeurism) and more antisociality (e.g. general criminality, nonsexual violence, adolescent antisocial behaviour) than White individuals. Similar patterns have been observed among samples of adolescent offenders; Burton and Ginsburg (2012) found that White youth scored significantly higher than Black youth on an overall deviance composite score, as well as in the following deviant interest categories: females under age 12, masochism, masturbating in public (exhibitionism), voyeurism, sadism, and fetishism (specifically, underwear fetish).

Utilising a penile plethysmography (PPG) method of assessing sexual arousal (which has limitations; see Abel et al., 2004; Akerman & Beech, 2012; Kalmus & Beech, 2005), Murphy and colleagues (2001) found that Black adolescents with a history of sexual offending displayed lower levels of erectile responding overall (to all stimuli), compared to their White counterparts. Regarding paraphilic interests, White adolescents with male victims displayed significantly higher levels of erectile responding to male child rape and male peer rape stimuli than Black adolescents with male victims; White adolescents also showed higher levels of erectile responding to female peer rape stimuli than their Black counterparts (Murphy et al., 2001).

Current study

While both “normative” and paraphilic sexual interests may vary by race and/or they may be influenced by culture, few studies have examined racial and ethnic differences in paraphilic interests among non-offending adult samples. Ethnic and cultural differences in how people experience and report or do not report paraphilic interests is important to examine, as understanding these differences may help guide clinicians by reducing potential cultural bias in assessment, arriving at the most accurate diagnoses of paraphilic disorders, and most accurately identify those individuals in greatest need of intervention. Understanding base-rates is also integral to accurate decision-making, especially with respect to high stakes psycho-legal issues such as sentencing and post-sentence civil commitment evaluations (Bar-Hillel, 1980; Yung, 2013). Therefore, the current study aimed to examine racial and ethnic differences in self-reported paraphilic interests as defined by the DSM-5-TR, utilising the Paraphilia Interests and Disorders Scale (PIDS) among a large, non-offending, non-clinical sample drawn from the general population. To that end, there were four primary goals, comprised of identifying potential racial and ethnic differences in: (1) the presence of self-reported paraphilic interests and behaviours, (2) paraphilic interests being the preferred or primary sexual interest, (3) the age of onset of the paraphilic interest, and (4) distress resulting from the paraphilic interest.

Method

Participants and procedure

The sample included 2,292 adult participants who lived in the U.S. and were able to speak and write in English. Participants completed the study online using Prolific, a survey platform that allows individuals to complete research studies in exchange for monetary compensation. Compared to similar crowdsourcing platforms (e.g. university pools, MTurk, Crowdflower), Prolific has been found to reach a more diverse participant population (e.g. geographical location, race/ethnicity; Palan & Schitter, 2018). In order to ensure confidentiality and privacy, no identifying information including IP address was collected from participants during the course of the study.

The study was advertised to potential participants through postings on the Prolific website or email invitations, which included a brief synopsis as well as information regarding inclusion criteria, confidentiality, and compensation. Those who were interested in participating and met inclusion criteria (i.e. over 18 years old living in the U.S., could read and write in English) read an informed consent form and electronically indicated their agreement to participate. Participants then responded to demographic questions (e.g. age, sex at birth, self-identified gender, race/ethnicity, highest education level, employment status, income, sexual orientation, relationship status) followed by the Paraphilic Interests and Disorders Scale (PIDS), a measure of self-reported paraphilic interests/behaviours (described further below). With respect to race and ethnicity, participants selected from the following pre-populated response options: White, Black/African-American, Hispanic/Latino, Asian/Pacific Islander, Middle Eastern/North African, American Indian/Alaskan Native, Mixed Race, and Other. Based upon low rates of endorsement, the categories Middle Eastern/North African, American Indian/Alaskan Native, Mixed Race, and Other were later combined together to comprise the “Other” category.

Following completion of the study, participants were provided a debriefing form, thanked for their time, and provided with compensation. Consistent with U.S. federal employment rates (U.S Department of Labor, 2020 ), compensation in the amount of $4.00 was provided for the completion of the 30-minute survey, which has been found to produce valid data (Horton & Chilton, 2010). This study was approved by Farleigh Dickinson University’s Institutional Review Board.

Data cleaning

A total of 2,430 responses were collected. Responses were removed from the final dataset if: (a) the survey was incomplete (n = 112), (b) two or more of the four attention checks were failed (n = 2), or (c) the survey completion time was two standard deviations outside the mean (n = 8). Responses provided by individuals who submitted two response sets (n = 10), and by those who provided a Prolific ID number that did not correspond with the ID number in Prolific system (n = 6), were also removed.

The final sample consisted of 2,292 participants and included representation from all 50 states. Demographic information for the sample is provided in Table 2. Males (53.0%; n = 1,214), and females (46.7%; n = 1,071) were represented at nearly equal rates. A majority of participants identified as White (68.9%; n = 1,580) and heterosexual (79.9%; n = 1,831), with most reporting their relationship status as married (39.9%; n = 914) or single (32.4%; n = 743). Regarding education, almost all (99.2%; n = 2,275) of respondents had a high school diploma or higher. Slightly over half (55.8%; n = 1,280) were employed full-time.

Table 2. Demographic characteristics by race/ethnicity (N = 2,292).

Variable	Total		White		Black		Hispanic		Asian		Other
	N	%	N	%	N	%	N	%	N	%	N	%
Total	2,292	–	1,580	68.94	262	11.43	133	5.8	224	9.77	93	4.06
Sex
Male	1,214	53.0	816	35.6	160	6.98	71	3.1	125	5.45	42	1.83
Female	1,071	46.7	759	33.12	102	4.45	62	2.71	98	4.28	50	2.18
Other	7	0.3	5	0.22	–	–	–	–	1	0.04	1	0.04
Gender
Man	1,191	52.0	802	34.99	159	6.94	69	3.01	122	5.32	39	1.70
Woman	1,023	44.6	720	31.41	101	4.41	59	2.57	95	4.14	48	2.09
Transgender M-to-F	6	0.3	5	0.22	–	–	–	–	1	0.04	–	–
Transgender F-to-M	9	0.4	7	0.31	–	–	1	0.04	–	–	1	0.04
Non-binary	39	1.7	28	1.22	1	0.04	2	0.09	4	0.17	4	0.17
Questioning	9	0.4	6	0.26	1	0.04	1	0.04	–	–	1	0.04
Other	10	0.4	9	0.39	–	–	1	0.04	–	–	–	–
Prefer not to Answer	5	0.2	3	0.13	–	–	–	–	2	0.09	–	–
Education
8th grade	2	0.1	1	0.04	–	–	–	–	–	–	1	0.04
Some high school	15	0.7	10	0.44	2	0.09	2	0.09	1	0.04	–	–
GED	40	1.7	29	1.27	9	0.39	1	0.04	1	0.04	–	–
High school graduate	219	9.6	163	7.11	18	0.79	15	0.65	13	0.57	10	0.44
Some college	565	24.7	379	16.54	56	2.44	49	2.14	53	2.31	28	1.22
College graduate	824	36.0	566	24.69	79	3.45	48	2.09	102	4.45	29	1.27
Some graduate school	103	4.5	64	2.79	15	0.65	3	0.13	13	0.57	8	0.35
Completed graduate school	507	22.1	359	15.66	81	3.53	15	0.65	39	1.7	13	0.57
Other	14	0.6	1	0.04	1	0.04	–	–	1	0.04	–	–
Prefer not to answer	3	0.1	8	0.35	1	0.04	–	–	1	0.04	4	0.17
Employment
Full-time	1,280	55.8	905	39.49	174	7.59	63	2.75	101	4.41	37	1.61
Part-time	271	11.8	179	7,81	31	1.35	11	0.48	32	1.4	18	0.79
Unemployed	294	12.8	196	8.55	25	1.09	27	1.18	27	1.18	19	0.83
Student	211	9.2	99	4.32	21	0.92	25	1.09	55	2.4	11	0.48
Self-employed	62	2.7	83	3.62	6	0.26	1	0.04	–	–	3	0.13
Disabled	22	1.0	12	0.52	–	–	–	–	3	0.13	1	0.04
Retired	93	4.1	36	1.57	2	0.09	3	0.13	1	0.04	1	0.04
Other	43	1.9	50	2.18	3	0.13	2	0.09	5	0.22	2	0.09
Prefer not to answer	16	0.7	20	0.87	–	–	1	0.04	–	–	1	0.04
Sexual Orientation
Heterosexual	1,831	79.9	1250	54.54	238	10.38	93	4.06	183	7.98	67	2.92
Homosexual	100	4.4	79	3.45	7	0.31	2	0.09	11	0.48	1	0.04
Bisexual	233	10.2	162	7.07	11	0.48	30	1.31	19	0.83	11	0.48
Pansexual	63	2.7	46	2.01	3	0.13	5	.22	1	0.04	8	0.35
Asexual	32	1.4	23	1.00	1	0.04	1	0.04	6	0.26	1	0.04
Other	25	1.1	16	0.70	2	0.09	1	0.04	2	0.09	4	0.17
Prefer not to answer	8	0.3	4	0.17	–	–	1	0.04	2	0.09	1	0.04
Relationship Status
Single	743	32.4	431	18.8	98	4.28	53	2.31	121	5.28	40	1.75
Dating/cohabiting/engaged	501	21.9	345	15.05	40	1.75	38	1.66	52	2.27	26	1.13
Married	914	39.9	696	30.27	115	5.02	37	1.61	47	2.05	19	0.83
Divorced	84	3.7	66	2.88	6	0.26	3	0.13	3	0.13	6	0.26
Widowed	16	0.7	16	0.70	–	–	–	–	–	–	–	–
Separated	20	0.9	17	0.74	3	0.13	–	–	–	–	–	–
Other	7	0.3	5	0.22	–	–	–	–	1	0.04	1	0.04
Prefer not to answer	7	0.3	4	0.17	–	–	2	0.09	–	–	1	0.04

Materials

Paraphilic interests and disorders scale (PIDS)

The Paraphilic Interests and Disorders Scale (PIDS; Winters et al., 2022) is a self-report tool developed to assess for the DSM-5/DSM-5-TR paraphilic interests and disorders (i.e. voyeuristic, exhibitionistic, frotteuristic, sexual masochism, sexual sadism, pedophilic, fetishistic, transvestic). The measure was created using the DSM-5 language for the paraphilias, as well as prior paraphilic interest measures (e.g. Bártová et al., 2021; Seto et al., 2012), and consultation with a psychiatrist who also consulted for the DSM-5 and has been deemed an expert on the diagnostic category. Additional information on the development and pilot testing for the PIDS can be found in Winters et al. (2022). Despite the recent adaptation of the DSM-5-TR, the PIDS remains relevant, as the diagnostic criteria for paraphilic interests/disorders were not changed in the text revision.

The PIDS assesses for “current” (i.e. over the last six months), and “lifetime” (i.e. any time in the past) paraphilic interests. When completing the PIDS, respondents are asked to indicate if, at any point in their lives, they experienced each of the paraphilic interests. For each interest that is endorsed, they are asked if that is their primary sexual interest, their age when they first experienced the interest, and the longest period of time during which they had that interest (0–5 months, 6–11 months, 1–2 years, 2–5 years, more than 5 years). Consistent with the DSM-5-TR criteria, they then complete follow-up questions related to duration, intensity, and impairment. First, in relation to Criterion A, they are asked if the interest was ever present for at least six months. For any time period of six months in which the interest was present, they are asked how often the interest occurred (the specific number) and to rate the intensity of the interest (1 = low intensity, 2–3 = moderate intensity, 4–5 = high intensity). Next, for Criterion B, respondents indicate how many times (specific number) and in what ways (i.e. viewing pornography, masturbation to the idea or fantasy, masturbating to pornography, engaging in acts with another person, other) they acted on the interest. They are also asked questions regarding the level of distress associated with the interest (1 = low distress, 2–3 = moderate distress, 4–5 = high distress), and if any impairment resulted (Yes/No). For each question, participants are also able to respond by indicating they would “Prefer not to Answer,” given the sensitive nature of the subject matter (see Winters et al., 2023 for additional details regarding the PIDS scoring).

Using feedback from experts (n = 22) who specialise in paraphilia diagnoses, content validity for the PIDS was established by Winters and colleagues (2022). The measure has also demonstrated convergent validity and test-retest reliability (Winters et al., 2022 in submission).

Results

A series of chi-square tests using ANOVA and post-hoc configural frequency analysis were conducted to examine differences among racial/ethnic groups (i.e. White, Black, Hispanic, Asian, Other) in self-reported paraphilic interests and behaviour, as well as whether the paraphilic interest was the primary sexual interest and whether it caused distress. See Table 3 for descriptive statistics regarding the frequency and percentage of individuals in each racial/ethnic group based on the aforementioned interests, behaviours, primary interest, and distress. Table 4 presents the chi-square results for the group comparisons for each dependent variable.

Table 3. Descriptives for PIDS interests, behaviours, primary interest, and distress by race/ethnicity.

PIDS scale	Interests	Behaviours	Primary Interest	Distress
	n (%)	n (%)	n (%)	n (%)
Voyeurism
White	598 (37.8)	160 (10.1)	70 (4.4)	39 (2.5)
Black	125 (47.7)	60 (22.9)	30 (11.5)	18 (6.9)
Hispanic	31 (23.3)	7 (5.3)	2 (1.5)	3 (2.3)
Asian	79 (35.3)	20 (8.9)	5 (2.2)	4 (1.8)
Other	33 (35.5)	8 (8.6)	2 (2.2)	1 (1.1)
Exhibitionism
White	86 (5.4)	39 (2.5)	20 (1.3)	16 (1.0)
Black	10 (3.8)	5 (1.9)	3 (1.1)	1 (0.4)
Hispanic	1 (0.8)	1 (0.8)	1 (0.8)	1 (0.8)
Asian	6 (2.7)	0 (0)	0 (0)	0 (0)
Other	5 (5.4)	3 (3.2)	0 (0)	1 (1.1)
Frotteurism
White	116 (7.3)	43 (2.7)	9 (0.6)	11 (0.7)
Black	30 (11.5)	16 (6.1)	4 (1.5)	4 (1.5)
Hispanic	9 (6.8)	3 (2.3)	2 (1.5)	1 (0.8)
Asian	19 (8.5)	4 (1.8)	1 (0.4)	2 (0.9)
Other	10 (10.8)	0 (0)	1 (1.1)	1 (1.1)
Sexual Masochism
White	318 (20.1)	186 (11.8)	51 (3.2)	11 (0.7)
Black	24 (9.2)	9 (3.4)	6 (2.3)	2 (0.8)
Hispanic	26 (19.5)	18 (13.5)	8 (6.0)	1 (0.8)
Asian	56 (25.0)	18 (8)	8 (3.6)	2 (0.9)
Other	21 (22.6)	12 (12.9)	4 (4.3)	1 (1.1)
Sexual Sadism
White	117 (7.4)	61 (3.9)	20 (1.3)	13 (0.8)
Black	16 (6.1)	8 (3.1)	3 (1.1)	1 (0.4)
Hispanic	12 (9.0)	2 (1.5)	0 (0)	0 (0)
Asian	22 (9.8)	3 (1.3)	1 (0.4)	3 (1.3)
Other	15 (16.1)	6 (6.5)	3 (3.2)	1 (1.1)
Pedophilic
White	29 (1.8)	11 (0.7)	2 (0.1)	4 (0.3)
Black	3 (1.2)	2 (0.8)	0 (0)	1 (0.4)
Hispanic	2 (1.5)	0 (0)	1 (0.8)	1 (0.8)
Asian	3 (1.3)	0 (0)	0 (0)	0 (0)
Other	1 (1.1)	0 (0)	1 (1.1)	1 (1.1)
Fetishistic
White	197 (12.5)	112 (7.1)	14 (0.9)	11 (0.7)
Black	26 (9.9)	15 (5.7)	3 (1.1)	4 (1.5)
Hispanic	21 (15.8)	12 (9.0)	4 (3.0)	2 (1.5)
Asian	38 (17.0)	14 (6.3)	3 (1.3)	0 (0)
Other	14 (15.1)	9 (9.7)	0 (0)	0 (0)
Transvestic
White	74 (4.7)	38 (2.4)	11 (0.7)	10 (0.6)
Black	9 (3.4)	5 (1.9)	2 (0.8)	1 (0.4)
Hispanic	4 (3.0)	2 (1.5)	0 (0)	0 (0)
Asian	19 (8.5)	4 (1.8)	3 (1.3)	1 (0.4)
Other	6 (6.5)	5 (5.4)	2 (2.2)	2 (2.2)

Table 4. Chi-square results for PIDS interests, behaviours, primary interest, and distress by race/ethnicity.

PIDS scale	chi-sq	p	V
Interests
Voyeurism	23.85	<.001	.07
Exhibitionism	9.01	.06	.04
Frotteurism	6.99	.14	.04
Sexual Masochism	23.54	<.001	.07
Sexual Sadism	11.71	.02	.05
Pedophilic	1.08	.9	.02
Fetishistic	7.1	.13	.04
Transvestic	9.39	.05	.05
Behaviours
Voyeurism	23.2	<.001	.07
Exhibitionism	4.36	.36	.03
Frotteurism	9.67	.05	.05
Sexual Masochism	9.39	.05	.05
Sexual Sadism	12.08	.02	.05
Pedophilic	7.78	.1	.04
Fetishistic	9.36	.05	.05
Transvestic	8.29	.08	.04
Primary interest
Voyeurism	23.2	<.001	.07
Exhibitionism	7.05	.13	.04
Frotteurism	3.25	.52	.03
Sexual Masochism	5.28	.26	.03
Sexual Sadism	4.99	.29	.03
Pedophilic	11.77	.02	.05
Fetishistic	5.52	.24	.03
Transvestic	2.32	.68	.02
Distress
Voyeurism	10.26	.04	.05
Exhibitionism	3.46	.48	.03
Frotteurism	1.03	.91	.01
Sexual Masochism	1.68	.79	.02
Sexual Sadism	1.72	.79	.02
Pedophilic	1.86	.6	.02
Fetishistic	11.04	.03	.05
Transvestic	2.86	.58	.03

Note: Bold values indicate results with significant p values (<.05).

For paraphilic interests, the results revealed significant differences among groups for voyeurism (χ² [4, N = 2,266] = 23.85, p < .001, V = .07), sexual masochism (χ² [4, N = 2,278] = 23.54, p < .001, V = .07), and sexual sadism (χ² [4, N = 2,276] = 11.71, p = .02, V = .05); all effect sizes were considered small. Post hoc analyses showed that proportionally, more Black respondents endorsed voyeuristic interests compared to the other racial/ethnic groups (χ²= 2.66, p_corrected= .004), and fewer Hispanic respondents endorsed that interest (χ²= −2.69, p_corrected= .004) compared to the other groups. Proportionally fewer Black respondents endorsed sexual masochism (χ²= −3.78, p_corrected< .001) compared to the other groups, whereas proportionally more respondents who self-identified their ethnicity as “Other” endorsed sexual sadism (χ²= 2.77, p_corrected= .003).

With regard to paraphilic behaviours, significant group differences were found only for the voyeuristic behaviour category (χ² [4, N = 591] = 23.2, p < .001, V = .07) after Bonferroni corrections. Post hoc analyses indicated that proportionally, Black respondents were more likely to endorse voyeuristic behaviours (χ²= −2.81, p_corrected= .002), and were less likely not to endorse voyeuristic behaviours (χ²= 3.22, p_corrected= .001), compared to the other groups.

Per one-way ANOVAs, race/ethnicity was not significantly associated with the age of onset for any of the paraphilic interests (p’s range from .098 to .70; see Table 5).

Table 5. Age of onset of first PIDS interest by race/ethnicity.

PIDS scale	Mean	SD	Range	F-statistic	p-value
Any
White	16.20	5.40	4–54	1.48	.21
Black	16.66	4.58	5–38
Hispanic	17.15	5.17	5–35
Asian	15.52	4.57	5–36
Other	16.29	4.82	5–35
Voyeurism				1.42	.23
White	15.94	5.23	4–54
Black	16.99	4.64	5–38
Hispanic	16.87	4.90	5–30
Asian	15.87	4.61	6–36
Other	15.64	4.00	5–25
Exhibitionism				0.01	1.00
White	18.31	7.33	5–45
Black	18.00	6.67	7–30
Hispanic	19.00	–	19–19
Asian	18.5	5.21	14–28
Other	18.00	3.67	14–23
Frotteurism				0.74	.57
White	17.13	5.53	4–43
Black	18.07	6.43	6–35
Hispanic	19.11	7.47	9–33
Asian	15.89	2.83	10–20
Other	17.60	3.03	13–24
Sexual Masochism				0.54	.70
White	18.64	6.75	4–50
Black	19.79	6.82	10–40
Hispanic	17.54	5.09	5–30
Asian	17.89	5.44	6–35
Other	18.57	5.35	12–38
Sexual Sadism				1.99	.10
White	20.06	7.27	5–49
Black	18.50	6.23	10–30
Hispanic	19.92	6.10	12–35
Asian	16.27	3.55	8–22
Other	21.67	5.65	12–33
Pedophilic				0.69	.61
White	17.41	6.96	4–33
Black	12.67	6.43	8–20
Hispanic	13.50	2.12	12–15
Asian	15.33	2.08	13–17
Other	11.00	–	11–11
Fetishistic				1.00	.41
White	17.37	5.34	5–35
Black	17.00	5.66	6–27
Hispanic	17.52	4.17	12–30
Asian	15.58	4.04	5–25
Other	17.14	5.63	12–35
Transvestic				1.40	.24
White	20.59	7.99	6–55
Black	17.78	7.56	7–27
Hispanic	19.00	3.46	16–24
Asian	16.42	3.66	8–23
Other	21.00	10.14	13–40

Regarding whether the paraphilic interest was the individual’s primary sexual interest, the results revealed significant group differences for voyeuristic (χ² [4, N = 847] = 23.2, p < .001, V = .07) and pedophilic interests (χ² [4, N = 35] = 11.77, p = .02, V = .05); all effect sizes were considered small. Post hoc analyses indicated that proportionally more Black respondents endorsed voyeuristic interests as their primary sexual interest (χ²= 3.90, p_corrected< .001) compared to the other groups, while more respondents who identified as “Other” endorsed pedophilic interests as their primary sexual interest (χ²= 2.62, p_corrected= .004) compared to the other groups.

With regard to subjective experiences of distress related to paraphilic interests, group differences were found for voyeuristic (χ² [4, N = 398] = 10.26, p = .04, V = .05) and fetishistic (χ² [4, N = 187] = 11.04, p = .03, V = .05) interests; all effect sizes were considered small. Specifically, post hoc analyses showed that proportionally more Black respondents reported distress due to voyeuristic (χ²= 2.62, p_corrected= .004) and fetishistic (χ²= 2.59, p_corrected= .005) interests in comparison to the other groups.

Discussion

There is a dearth of literature exploring racial and ethnic differences in paraphilic interests among non-offending adult samples. Yet, these individual differences are important to explore in order to improve the evaluation and treatment of individuals who present with paraphilic disorders, especially considering the relationship between paraphilias and sexual offending (Akerman & Beech, 2012; Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2005). Using a measure of DSM-5-TR paraphilic interests and behaviours (PIDS), racial and ethnic differences were explored among a large non-offending sample with regard to self-reported paraphilic interests, behaviours, age of onset, and associated distress. Overall, the findings revealed some group differences in paraphilic interests, although these findings all reflected small effect sizes.

Given that the research on race, ethnicity, and their relationship to paraphilias is lacking, and the effect sizes in this study are small, it is difficult to make strong conclusions. The most consistent finding in this study was that proportionally more Black individuals reported voyeuristic interests and behaviours, and that voyeurism was identified as their primary sexual interest. Additionally, Black individuals also reported more distress than other racial and ethnic groups resulting from their voyeuristic interest. One explanation for these findings is that while voyeuristic interests and behaviours are not synonymous with viewing pornography, there may be some overlap. The PIDS uses the DSM-5-TR language which defines voyeurism as “recurrent and intense sexual arousal from observing an unsuspecting person who is naked, in the process of disrobing, or engaging in sexual activity” (APA, 2022, p. 780). Arguably, since the definition in the DSM-5-TR does not specify that the observation must be “in vivo,” this definition could apply to watching actors/models in pornographic images or movies. Recent research has suggested that Black individuals viewed pornography at increased rates compared to White individuals in the U.S., and that Black individuals were increasing their pornography viewership at a higher rate than White individuals (Perry & Schleifer, 2019). Therefore, it is possible that participants misunderstood the PIDS voyeurism prompt to reflect pornography viewership, which may be higher among Black individuals compared to other racial and ethnic groups. Alternatively, it may be that “in-vivo” voyeuristic interests and behaviours are also more common among Black individuals, and that those interests are related to pornography viewership, as pornography viewership has been linked to paraphilic interests (Bártová et al., 2021).

Another notable finding was that those in the “Other” racial/ethnic category were overrepresented as having sexual sadistic interests in general, as well as identifying pedophilic interests as their primary sexual interest. Both sexually sadistic and pedophilic interests/behaviours have been consistently linked to sexual offending and sexual recidivism (Eher et al., 2016; Kingston et al., 2010; Seto, 2019). In this study, the “Other” group was comprised of Middle Eastern/North African, American Indian/Alaska Native, Mixed Race, and Other. It is possible that those who identified as belonging to the “Other” group were different from those who identified as belonging to the identified racial/ethnic groups examined in this study. For example, it is possible that individuals identifying as “Other” were more recent immigrants or belonged to cultures that have been less extensively studied and have different social and cultural norms pertaining to sexual behaviour, such as marriage at a younger age or more patriarchal societies (Bhugra et al., 2010). Although this question cannot be answered using the current data alone, future research utilising a larger sample overall may result in sufficient sample sizes for additional comparisons among racial/ethnic groups beyond White, Black, Hispanic, and Asian, and may therefore be better equipped to detect meaningful and specific group differences. Future research may also explore respondents’ country of origin and length of time in the U.S. and the potential impact of these factors on paraphilic interests and behaviours.

It is worth noting that there were no significant differences found between racial and ethnic groups on most comparisons across interests, behaviours, and age of onset. As noted, almost all research that has examined racial and ethnic differences in paraphilic interests has been conducted with incarcerated samples, which are disproportionately comprised of racial and ethnic minority individuals, and may represent inequities in how selection criteria are applied, reinforcing the need for an accurate understanding of base rates. Further, it has been suggested that paraphilic interests may be culturally bound (e.g. Bhugra et al., 2010). The current study included only individuals residing in the U.S., and while the U.S. is a multiracial and multiethnic society in many ways, some cultural distinctions in paraphilic interests and behaviours may have been minimised, as all respondents were in fact residing in a Western culture with Western societal norms. Finally, as the PIDS is based upon the DSM-5 TR diagnostic criteria, paraphilic interests and behaviours were measured dichotomously instead of continuously. It is possible that, given the taboo nature of paraphilic interests, asking about the topic outright may render an individual less likely to respond affirmatively – but perhaps asking about the presence of interests and behaviours utilising a continuum such as on the Paraphilia Scale (Seto et al., 2012) or asking indirectly (see Ó Ciardha et al., 2022) may yield different results.

Limitations

There are several limitations of the present study. First, the data was collected via self-report (as opposed to clinical interviews, physiological measures, etc.), which may be impacted by recall and/or social desirability issues. Nonetheless, we believe the self-report method is necessary to obtain large-scale data regarding paraphilic interests and behaviours among the wider, non-offending population, though future studies should explore whether other measures of paraphilic interests may produce similar or different results.

Second, while the large sample size provided relatively large group sizes for each racial/ethnic category, the individuals in the dataset were primarily White. According to the 2022 U.S. Census, 59.3% of the U.S. population identifies as only White, 12.1% identifies as only Black/African American, 18.7% as only Hispanic/Latino and 6.1% as only Asian (U.S. Census Bureau, 2022). Thus, in our sample, White (68.9%) and Asian (9.8%) individuals were overrepresented compared to their prevalence in the general U.S. population, while Latino/Hispanic (5.8%) individuals were underrepresented. The percentage of Black/African American (11.4%) respondents in this sample roughly corresponds to their prevalence in the U.S. population.

In light of the use of overarching racial and ethnic categories, country/region of origin, length of time in the U.S., and generational status were not assessed and may potentially influence paraphilic interests and behaviours in ways that were not directly measured in this study. Moreover, the sample was comprised of individuals who currently reside in the U.S., which as noted previously, may impact the generalisability of the findings to other countries/cultures.

Future replication studies would provide further information regarding the representativeness of the current sample with respect to racial/ethnic differences in paraphilic interests in the larger general population; however, it is worth noting that the overall rates of endorsement of paraphilic interests were generally consistent with that of previous studies (e.g. Bártová et al., 2021; Joyal & Carpentier, 2017; see Winters et al., 2023 for summary).

Implications and future directions

The findings of this study suggest that there may be some racial and ethnic differences in paraphilic interests and behaviour as assessed using the PIDS, which is based upon the DSM-5 criteria. However, given the limitations noted above and the small effect sizes, the conclusions must not be overstated and the findings must be interpreted within the context of these limitations. What the findings do suggest is that it is worthwhile and necessary to conduct more research examining potential racial and ethnic differences in paraphilic interests and behaviours using larger and more representative samples. The findings also provide some support for the need to understand paraphilic interests and disorders within a cultural framework.

Further, given that this study was quantitative in nature, such that participants only reported whether or not they experienced an interest or behaviour, it would be important to contextualise these findings. For example, mixed methods anonymous surveys could allow for individuals to explain the nature of their paraphilic interest, how it developed, and the ways in which it impacts them. While interviews would allow for more nuanced questioning, given the sensitive and potentially illegal nature of some paraphilic interests and behaviours, anonymous surveys may yield more valid responding (Joyal & Carpentier, 2017).

Given the strong link between paraphilic interests and sexual offending (Akerman & Beech, 2012; Hanson & Bussière, 1998; Hanson & Morton-Bourgon, 2005), as well as the fact that important psycho-legal decisions can hinge upon the diagnosis of a paraphilic disorder, it is crucial that more is learned about how paraphilic interests and behaviours may be experienced differently across race and culture. This is particularly salient given the overrepresentation of racial and ethnic minorities within the criminal justice system (See Jeglic & Calkins, 2022 for a review) and sex offender registries (Ackerman & Sacks, 2018).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Fairleigh Dickinson University; New Jersey Association for the Treatment of Sexual Abusers.