Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 May 1.
Published in final edited form as: J Interpers Violence. 2012 Dec 21;28(7):1519–1536. doi: 10.1177/0886260512468237

Self-Reported Sexual Assault in Convicted Sex Offenders and Community Men

Laura Widman 1, Michael A Olson 2, Rebecca M Bolen 3
PMCID: PMC3624064  NIHMSID: NIHMS435464  PMID: 23262829

Abstract

Although self-reported sexual assault perpetrated by men against women has been well documented among college men, less is known about self-reported perpetration among convicted sex offenders and community men. This study provides unique descriptive and comparative information on sexual assaults in these understudied populations. Participants were 40 convicted sex offenders and 49 demographically-comparable community men who completed the Sexual Experiences Survey (SES; Abbey, Parkhill, & Koss, 2005; Koss, Gidycz, & Wisniewski, 1987) and other surveys to capture the promiscous sex and hostile masculinity pathways posited by the confluence model (Malamuth, 2003). We found notably few differences between sex offenders and community men in the rate and severity of sexual assault perpetration and the tactics used to obtain unwanted sexual contact. Specifically, 68% of sex offenders and 59% of community men acknowledged they had perpetrated sexual assault. Both groups used guilt and anger as the most frequent tactics to obtain unwanted sexual activity from their female victims. Consistent with the confluence model, an impersonal orientation towards sexual relationships was associated with sexual assault for both sex offenders and community men. Future directions for research on sexual assault perpetration and violence prevention efforts are discussed in light of these findings.

Keywords: sexual violence, rape, Sexual Experiences Survey, confluence model

Sexual assault is a prevalent social problem. Between five and ten percent of college men admit to forcing a woman into sexual activity that meets the definition of rape (e.g., Abbey, McAuslan, Zwacki, Clinton, & Buck, 2001; Koss et al., 1987; Kosson, Kelly, & White, 1997; Loh, Gidycz, Lobo, & Luthra, 2005; Spitzberg, 1999). Further, between one quarter and one half of college men admit to verbally or physically coercing a woman into other unwanted sexual acts, such as unwanted sexual contact and sexual coercion (e.g., Abbey et al., 2001; Koss et al., 1987; Loh et al., 2005).

Although much is known about sexually aggressive behavior among college men (for reviews, see Koss, 1993; Schewe, 2002), much less is known about this behavior among men who are not in college. To date, only a small handful of studies have examined sexual assault perpetrated against adolescent and adult women by men in community samples (Abbey et al., 2006; Calhoun, Benat, Clum, & Fame, 1997; Merrill, Thomsen, Gold, & Milner, 2001). For example, in one community study, Abbey et al. (2006) surveyed a racially diverse group of 163 men from Detroit (median age = 29) and found 63% of men had committed at least one form of sexual assault, and 24% had committed rape or attempted rape against a woman over the age of 14. Self-reported sexual assault perpetrated against adolescent and adult women has also been infrequently studied in samples of convicted sex offenders (for exceptions, see Freeman-Longo, 1985; Scully, 1990; Weinrott et al., 1991). In fact, to our knowledge, although the Sexual Experiences Survey (SES; Abbey et al., 2005; Koss et al., 1987) is very commonly used to measure sexual assault perpetrated against adolescent and adult women in research settings, no known studies have utilized the SES in a sample of convicted sex offenders.

The limited research on these populations leaves us with at least two important unanswered questions. First, do rates of self-reported sexual assault differ between known sex offenders and men in a community sample? While we posit that the frequency and severity of self-reported sexual assault will be higher among known offenders than community men, a direct comparison of these groups using the SES has not been previously documented. Second, do the factors that are known to predict sexual assault among college men also predict sexual assault among known offenders and community men? The confluence model (Malamuth, Linz, Heavy, Barner, & Acker, 1995; Malamuth, Sockloskie, Koss, & Tanaka, 1991) is a robust theory of sexual assault that has identified two factors that consistently predict sexual assault in college samples: hostile masculinity and sexual promiscuity. According to Malamuth et al. (1995, pg. 354), men high in hostile masculinity have “an insecure, defensive, hypersensitive, and hostile-distrustful orientation,” and men high in sexual promiscuity have a “non-committal, game-playing orientation in sexual relations.” Research with college men has consistently shown these two factors are individually associated with the perpetration of sexual assault and also work in a synergistic fashion, such that men high in hostile masculinity and promiscuous sex are the most likely to commit sexual assault (Dean & Malamuth, 1997; Hall, Teten, DeGarmo, Sue, & Stevens, 2005; Malamuth et al., 1991; Vega & Malamuth, 2007; Wheeler, George, & Dahl, 2002). Yet, to date, there is a relative paucity of research that uses the confluence model to predict sexual offending in samples of sex offenders and community men.

Thus, the aim of this study is to fill an important gap in research on sexual assault by examining rates of self-reported perpetration in two non-college samples: 1) convicted sex offenders and 2) demographically-comparable men recruited from the community. This study will also examine if the factors that are strongly associated with sexual assault in college men are also apparent for the current populations under study.

Method

Participants

Sex Offenders

Participants in the sex offender sample were 44 men between the ages of 18 and 63 (M = 39, SD = 11.6) recruited from an outpatient sex offender treatment program in the Southeastern United States for a larger study of sexual attitudes (reference removed for blind review). Participants were recruited through study announcements and flyers distributed at the treatment center that advertised a study of men’s sexual attitudes. One participant arrived at the research session but did not complete the study due to severe hand tremors that did not allow him to use a computer. Data from a second participant were excluded because he identified as exclusively homosexual, and he only perpetrated sexual offenses against boys (data from this participant was removed because the focus of the current paper was to assess sexual assault perpetrated against women). Finally, two participants were dropped because they did not complete the sexual assault scale. Thus, the final sample included 40 sex offenders with useable data. The racial identification in this sample was primarily Caucasian (88%), and also included Native American (5%), African-American (2%), and other races (3%). The highest level of education completed included: high school graduate or less (47%), some college (29%), Associates/Bachelors degree (16%), or graduate degree (5%). With respect to economic status, participants reported their income in the past year was: $0–10,000 (31%), $10,001–20,000 (14%), or $20,001–40,000 (53%).

Community Men

We recruited a convenience sample of community men from the same geographical area. The recruitment goal was to match the sex offender sample on basic demographic characteristics such as age, race, socioeconomic status, and education level. To recruit this sample, we posted study flyers advertising a study of men’s sexual attitudes in targeted neighborhoods throughout the community. Participants were 51 men between the ages of 19 and 79 (M = 36, SD = 13.9), two of whom were omitted from analyses because they identified as exclusively homosexual. Thus, the final community sample included 49 men. The racial identification in this sample was primarily Caucasian (82%), and also included African-American (6%) and other races (10%). The highest level of education completed included: high school graduate or less (39%), some college (29%), Associates/Bachelors degree (29%), or graduate degree (4%). Finally, with respect to economic status, participants reported their income in the past year was: $0–10,000 (45%), $10,001–20,000 (29%), $20,001–40,000 (22%), or $40,001 or more (4%).

General Procedure

Participants first read and signed informed consent forms. All data were collected via laptop computers, one to two participants at a time, each in a private room. For sex offenders, data were collected at the outpatient sex offender treatment office; for community men, data were collected in our research lab. To encourage honest responding, data were collected anonymously (i.e., participants initialed the informed consent document and their names were never collected as part of the study). The study duration was approximately one hour, and all participants were compensated with $30. At the conclusion of the study, participants were fully debriefed utilizing a procedure modified from Malamuth and Check (1994) aimed at reducing rape myth acceptance. The University Institutional Review Board approved all procedures.

Demographic Information Form

Several questions were included to capture basic demographic information, including age, race, sexual orientation, education, work status, and socioeconomic status.

Sexual Offending Questionnaire

We gathered information about legal charges and convictions for a sexual offense, including the type of conviction, victim age and gender, and the offender’s relationship to the victim(s).

Promiscuous Sex

In line with the assessment proposed by Malamuth et al. (1991), the promiscuous sex variable was assessed with two items: 1) number of consensual lifetime sexual intercourse partners (responses were coded on an 6-point scale from “0” to “50 or more”) and 2) age at first intercourse (ages were recoded on a 6-point scale from 12 or younger to 19 or older). Both variables were standardized and then summed to form the index of Promiscuous Sex.

Hostile Masculinity

Hostile masculinity was assessed by a composite of three scales: the 10-item Revised Hostility toward Women Scale (RHTW; Lonsway & Fitzgerald, 1995; Cronbach’s α for current study = .83), the 20-item Illinois Rape Myth Acceptance Scale, Short-Form (IRMA-SF; Payne, Lonsway, & Fitzgerald, 1999; Cronbach’s α for current study = .90), and the 8-item Sexual Dominance Scale (Nelson, 1979; Cronbach’s α for current study = .91). Scores from each scale were standardized and then summed to form the index of Hostile Masculinity.

Sexual Assault

Sexual assault perpetration was assessed with an updated “tactics first” version of the Sexual Experiences Survey (SES; Abbey et al., 2005; Koss et al., 1987). The SES captures sexual assault perpetrated by men against women since the age of 14. The wording of some items was altered slightly to accommodate the reading level of this sample (e.g., we changed the word “intoxicated” to “drunk”). There were 35-items on the SES; however, in line with the suggestion from Abbey et al. (2005), we combined the 5 items that query about giving alcohol with the 5 items that query about giving drugs so that the final scale included 30 items. Specifically, participants reported whether or not they had used six different tactics – 1) arguments/pressure, 2) lies/promises, 3) guilt/anger, 4) alcohol or drugs, 5) taking advantage of an intoxicated woman, and 6) using physical force – to engage in five different types of unwanted sexual behavior: 1) fondling/kissing, 2) attempted sex, 3) oral sex, 4) sexual intercourse, and 5) anal sex/insertion of objects. For each item, participants were asked to indicate whether or not they had engaged in the behavior since the age of 14. We used the SES to calculate the total acts of sexual assaults (possible range 0–30) as well as an index of sexual assault severity: 1) no sexual assault; 2) sexual contact; 3) sexual coercion; 4) attempted rape; or 5) completed rape (See Koss et al., 1987 for definition of each category). Extensive reliability and validity information on the SES can be found elsewhere (Abbey et al., 2005; Koss & Gidycz, 1985; Koss et al., 2007). In the current sample, scale reliability was excellent (Cronbach’s α for sex offenders = .92, Cronbach’s α for community men = .91).

Analytic Plan

To address our research questions, we conducted a series of analyses. First, we conducted independent samples t-tests (for continuous variables) and chi-square tests (for dichotomous variables) to determine if there were any significant differences between sex offenders and community men on basic demographic factors. Next, we conducted a series of descriptive analyses to characterize the overall percentage, frequency, and severity of sexual assaults reported by sex offenders and community men. Then, we conducted an additional set of chi-square tests to determine if the frequency or severity of sexual assaults varied between sex offenders and community men. Finally, using negative binomial regression analyses, we examined the predictors of sexual assault among all participants. We used a dummy coded “group” variable (0 = community men, 1 = sex offenders) to determine if any differences existed between groups. Negative binomial regressions were used to account for the skewed count data (see Atkins & Gallop, 2007, for a thorough discussion of non-normal distributions and corrective statistical techniques). Age was entered as a control variable in all regression analyses.

Results

Sample Comparisons

Independent samples t-tests and Chi-square tests revealed that the sex offender group and community group did not significantly differ on main demographic variables, including: age [t(81) = −1.09, p = .28], race (Caucasian/not Caucasian; χ2 = 2.23, p = .14), yearly income (less than $10,000/more than $10,000; χ2 = 2.07, p = .15), education (high school or less/some college or more; χ2 = 0.69, p = .41), or work status (working/not working; χ2 = 1.00, p = .32).

Sexual Assault among Convicted Sex Offenders

Among sex offenders, the average number of charges for a sexual offense was 1.15 (SD = .56), and the average number of convictions for a sexual offense was 1.14 (SD = .48). Regarding the type of offense, 17% (n = 7) of sex offenders were charged or convicted solely of a non-contact offense (e.g., receipt of child pornography, indecent exposure) and 73% (n = 29) were charged or convicted of a contact offense (e.g., rape, attempted rape, aggravated sexual battery). Four offenders did not report the nature of their offenses. A breakdown by the age and gender of victim revealed that 60% of offenders (n = 24) were charged/convicted of a sexual offense against adolescent girl or adult woman and 33% (n = 13) were charged/convicted of an offense against a child or adolescent boy. Three offenders did not provide information on victim age.

When sexual assault was assessed via the SES, 68% (n = 27) of the sex offender sample acknowledged they had committed a sexual assault since the age of 14. The number of acts of self-reported sexual assaults ranged from 0 to 20 (M = 4.00, SD = 5.41). Further, 53% (n = 21) of sex offenders reported perpetrating more than one act of sexual assault and 25% (n = 10) reported perpetrating five or more acts. Further, 10% (n = 4) of sex offenders had perpetrated a form of sexual contact against a woman’s will, 35% (n = 14) had perpetrated an act of sexual coercion, no participants had attempted rape, and 18% (n = 7) had raped a woman. As shown in Table 1, the tactics used most frequently by sex offenders included guilt and anger (n = 17, 43%), followed by arguments and pressure (n = 15, 38%), and then lies and false promises (n = 14, 35%). One sex offender admitted to using physical force to obtain sexual activity.

Table 1.

Items Endorsed on the Sexual Experiences Survey by Sex Offenders and Community Men

Sex Offenders Community Men
SES Tactic and Items Total N = 40

N (%)		 Total N = 49

N (%)
Tactic 1: Overwhelmed a woman with continual arguments and pressure in order to:
      Fondle, kiss, or sexually touch her 8 (20%) 14 (29%)
      Attempt sexual intercourse (not completed) 9 (23%) 11 (22%)
      Make her have oral sex 5 (13%) 7 (14%)
      Make her have sexual intercourse 7 (18%) 10 (20%)
      Make her have anal sex or insert an object in her 3 (8%) 4 (8%)
Any use of arguments/pressure tactic: 15 (38%) 18 (37%)
Tactic 2: Told a woman lies or made promises that were untrue in order to:
      Fondle, kiss, or sexually touch her 13 (33%) 12 (25%)
      Attempt sexual intercourse (not completed) 9 (23%) 9 (18%)
      Make her have oral sex 8 (20%) 9 (18%)
      Make her have sexual intercourse 11 (28%) 13 (27%)
      Make her have anal sex or insert an object in her 4 (10%) 3 (6%)
Any use of the lies/promises tactic: 14 (35%) 18 (37%)
Tactic 3: Shown you were not happy by making a woman feel guilty, swearing, sulking, or getting angry in order to:
      Fondle, kiss, or sexually touch her 16 (40%) 19 (39%)
      Attempt sexual intercourse (not completed) 13 (33%) 16 (33%)
      Make her have oral sex 8 (20%) 9 (18%)
      Make her have sexual intercourse 14 (35%) 14 (29%)
      Make her have anal sex or insert an object in her 2 (5%) 2 (4%)
Any use of the guilt tactic: 17 (43%) 21 (43%)
Tactic 4: Gave a woman drugs or alcohol without her knowledge or consent in order to:
      Fondle, kiss, or sexually touch her 2 (5%) 1 (2%)
      Attempt sexual intercourse (not completed) 1 (3%) 2 (4%)
      Make her have oral sex 2 (5%) 1 (2%)
      Make her have sexual intercourse 3 (8%) 2 (4%)
      Make her have anal sex or insert an object in her 1 (3%) 0
Any use of the alcohol/drugs tactic: 4 (10%) 3 (6%)
Tactic 5: When a woman was passed out or too intoxicated to give consent or stop what was happening, have you ever:
      Fondled, kissed, or sexually touched her 8 (20%) 5 (10%)
      Attempted sexual intercourse (not completed) 2 (5%) 3 (6%)
      Made her have oral sex 2 (5%) 1 (2%)
      Made her have sexual intercourse 6 (15%) 1 (2%)
      Made her have anal sex or inserted an object in her 2 (5%) 1 (2%)
Any use of the intoxication tactic: 9 (23%) 6 (12%)
Tactic 6: Used some degree of physical force or in any other way held down or physically hurt a woman in order to:
      Fondle, kiss, or sexually touch her 1 (3%) 1 (2%)
      Attempt sexual intercourse (not completed) 1 (3%) 1 (2%)
      Make her have oral sex 1 (3%) 1 (2%)
      Make her have sexual intercourse 1 (3%) 1 (2%)
      Make her have anal sex or insert an object in her 1 (3%) 0
Any use of the physical force tactic: 1 (3%) 1 (2%)
Open in a new tab

Note. Participants were instructed to answer the items about experiences that happened since the age of 14. See Abbey et al., 2005 for complete item wording on the SES.

Further examination of SES scores by contact/no-contact offending group revealed that 18 of the 29 men who were convicted of a contact offense reported some sexual assault on the SES (frequency range = 1–20 acts), while 11 of the 29 contact offenders reported nothing on the SES. The legal convictions of these 11 men included statutory rape/rape of a child (n = 5), attempted rape (n = 3), and sexual battery/attempted sexual battery (n = 3). Also, five of the seven participants who were convicted of a non-contact offense reported some sexual assault on the SES (frequency range = 1–16 acts), while the other two non-contact offenders reported nothing on the SES. Both of these men were convicted of possession of child pornography. Finally, all four of the men missing offense contact information reported some sexual assault on the SES (frequency range = 2–11 acts).

Additionally, examination of SES scores by victim type revealed that 15 of the 24 men who offended against adolescent girls or adult women reported some sexual assault on the SES (frequency range = 1–20 acts). Further, 9 of the 13 men who offended against children or adolescent boys reported something on the SES (frequency range = 1–16 acts). Finally, all three of the men who were missing information on victim type reported some sexual assault on the SES (frequency range = 4–11 acts).

Sexual Assault among Community Men1

Since the age of 14, 59% (n = 29) of the men in the community sample reported they had committed at least one act of sexual assault, with the number of acts of sexual assaults ranging from 0 to 19 (M = 3.53, SD = 4.74). Further, 47% (n = 23) of community men reported perpetrating more than one act of sexual assault and 28% (n = 14) reported perpetrating five or more acts. A breakdown by the severity of sexual assault indicated that 16% (n = 8) had perpetrated a form of sexual contact against a woman’s will, 29% (n = 14) had perpetrated an act of sexual coercion, 6% (n = 3) had perpetrated an act of attempted rape, and 8% (n = 4) had raped a woman. As shown in Table 1, the tactics used most frequently by the community men included guilt and anger (n = 21, 43%), followed equally by arguments and pressure (n = 18, 37%) and lies and false promises (n = 18, 37%). One community man admitted to using physical force to obtain sexual activity.

Sexual Assault Comparisons Between Groups

Next, to determine if the number of acts of sexual assault varied between the sex offenders and community men, Chi Square tests were performed on each of the 30 SES items using a Bonferroni adjustment to account for multiple comparisons. No significant difference emerged between groups. Chi Square tests were also conducted to determine if there were differences in the severity of sexual assault between groups. No group differences were observed in the amount of unwanted sexual contact (χ2 = .76, p = .39), sexual coercion (χ2 = .42 p = .52), attempted rape (χ2 = 2.53 p = .11), or completed rape (χ2 = 1.77 p = .18) between groups. Further, no significant differences were observed between the mean number of acts of self-reported sexual assaults between sex offenders (M = 4.00, SD = 5.41) and community men (M = 3.53, SD = 4.74), t(87) = .44, p = .66.

Examining Predictors of Sexual Assault among All Participants

Finally, to examine if variables in the confluence model predict sexual assault among sex offenders and community men, we conducted a series of negative binomial regression analyses. First, we examined the main effects of promiscuous sex, hostile masculinity, and group status (0 = community men, 1 = sex offenders) on sexual assault frequency. Then we examined the interaction between promiscuous sex and hostile masculinity. Finally, we examined if the relationship between sexual assault and promiscuous sex or hostile masculinity varied by group membership. All analyses controlled for age. As indicated in Table 2, both promiscuous sex and hostile masculinity were significantly associated with the frequency of self-reported sexual assault perpetration. However, unlike results in typical college samples testing the confluence model (e.g., Malamuth et al., 1991), the interaction between promiscuous sex and hostile masculinity was not associated with sexual assault perpetration. Additionally, the main effect of hostile masculinity was qualified by group membership, such that hostile masculinity was associated with more frequent sexual assault perpetration among community men than sex offenders (Wald x2 = 5.04, Exp(B) = 0.46, 95% CI [0.23, 0.90], p < .05).

Table 2.

Negative Binomial Regression Analyses Predicting Sexual Assault Perpetration

Sexual Assault Perpetration
Wald x2 Exp (B) [95% CI]
Main Effects Modelsa
      Promiscuous Sex 9.83** 1.85 [1.26, 2.71]
      Hostile Masculinity 10.64** 1.61 [1.21, 2.14]
      Group 0.28 1.14 [0.71, 1.82]
Test of Promiscuous Sex x Hostile Masculinity Interactions
      Promiscuous Sex 6.98+ 1.70 [1.15, 2.52]
      Hostile Masculinity 6.99+ 1.50 [1.11, 2.02]
      Promiscuous Sex x Hostile Masculinity 0.29 1.15 [0.70, 1.88]
Test of Promiscuous Sex x Group Interactions
      Promiscuous Sex 6.81** 2.20 [1.22, 3.97]
      Group 0.25 1.13 [0.70, 1.82]
      Promiscuous Sex x Group 0.46 0.77 [0.35, 1.66]
Test of Hostile Masculinity x Group Interactions
      Hostile Masculinity 18.60*** 2.98 [1.81, 4.89]
      Group 7.80** 2.16 [1.26, 3.71]
      Hostile Masculinity x Group 5.04* 0.46 [0.23, 0.90]
Open in a new tab

Note. Age was included as a control variable in all analyses. Group = 0 (community men), 1 (sex offenders).

a

Main effects models were the individual regression models for each independent variable, controlling for age.

+

p < .10,

*

p < .05,

**

p < .01,

***

p < .001.

Discussion

This study contributes to a growing body of literature documenting the markedly high prevalence of male perpetrated sexual assault through analysis of data from two non-college populations: convicted sex offenders and community men. We found rates of sexual assault were very high in both groups, with 68% of sex offenders and 59% of community men acknowledging they had perpetrated sexual assault on the Sexual Experiences Survey (SES; Abbey et al., 2005; Koss et al., 1987). By comparing rates of sexual assault between groups, this study offers validity information on the SES, indicating that under the conditions of this anonymous research study, a large portion of known sexual offenders endorsed a history of sexually aggressive behavior on the SES; however, some men appear to underreport sexual assault. Our results also highlight the role of an impersonal, promiscuous orientation toward sexual relationships as a risk factor for sexual offending in both sex offenders and community men (Malamuth, 2003).

Perpetration in a Convicted Sex-Offender Sample

This study uniquely contributes to the literature on the prevalence of sexual assault in that it assesses rates of self-reported sexual assault among men previously convicted of a sexual offense. Importantly, there have been few published studies that thoroughly document the self-reported sexual assaults and rapes that convicted offenders perpetrate against adolescent and adult women. In fact, although the SES has been used for nearly three decades with college and community men, this is the first known study to use the SES in a convicted sex offender sample. Results of this descriptive assessment revealed that only 68% of convicted offenders acknowledged they had committed an act of sexual assault via self-report on the SES. There are several possible explanations for the notable discrepancy between SES reports and sexual offense convictions. First, it is possible that the men who did not report an assault on the SES were convicted of a non-contact sexual offense that was not captured by the SES (e.g., indecent exposure). However, results do not fully support this possibility: in follow-up analyses we found that over a third of men who were convicted of a contact sexual offense did not report anything on the SES. Second, because the SES asks about experiences perpetrated against a “woman”, it is possible that some offenders did not report acts they perpetrated against children and this lack of reporting accounted for the discrepancy between reports on the SES and convictions. Again, follow-up analyses suggest the lack of correspondence in wording on the SES cannot fully account for the discrepancy, as 9 of the 24 offenders who perpetrated against adolescent or adult women did not report a sexual assault on the SES. Another less forgiving explanation is that the sex offenders who were convicted of a contact offense but did not report anything on the SES intentionally lied in an effort to deny or minimize their behavior (cf. Ward, 2000). These possibilities remain speculative and subject to investigation in future work.

Regardless of the reason for the discrepancy, the current results suggest that the SES can capture a large portion of self-reported sexual offenses in a convicted sex offender sample but may still be influenced by problems of under-reporting. If the SES is used in future research with known sex offenders, it may be useful to modify the language to “woman or girl” or “person” to reflect a broader range of victim ages and account for perpetration against males. Additional research is needed to further validate the SES in a convicted sample using a large, representative sample of offenders.

Perpetration in a Community Sample

This study adds to the small body of prior research assessing rates of sexual assault in a sample of U.S. community men (Abbey et al., 2006; Calhoun et al., 1997; Merrill et al., 2001). We found that nearly 60% of men in this convenience sample reported they had committed at least one act of sexual assault. This rate is nearly equal to that reported by men convicted of a sexual offense. Further, a full 14% of community men admitted to an act that would meet most legal standards for an attempted or completed rape. These rates are comparable to rates reported by Abbey et al. (2006) who found the prevalence of sexual assault to be near 63% in a sample of community men in Detroit; however, our rates far exceed the prevalence rates around 30% that are typical in college samples (e.g., Abbey et al., 1998; Abbey et al., 2001; Loh et al., 2005). To accurately determine the scope of sexual assault, it would be useful to conduct a national prevalence study among community men.

Of note, there were few differences in the self-reported offending patterns of sex offenders and men recruited from the community. Both samples used guilt and anger as the most frequent tactic to obtain unwanted sexual activity from their victims, followed by arguments and pressure, and then lies and false promises. Further, when we examined the factors that may predict sexual assault, we found that a promiscuous orientation toward sex, defined by an early age of first intercourse and a greater number of sexual partners, was predictive of more sexual assault in both groups. We also found that hostile masculinity, characterized by rape myth acceptance and general hostility toward women, was predictive of sexual assault – though this pattern was more predictive for community men than for sex offenders. Finally, unlike studies with college men (e.g., Dean & Malamuth, 1997; Malamuth et al., 1991; Vega & Malamuth, 2007), the interaction between promiscuous sex and hostile masculinity posited in the confluence model was not significant in either sex offenders of community men. Few prior studies have examined the confluence model in non-college samples, and our results suggest that the factors that robustly predict offending among college men not be universally applicable to men outside the college environment. These results are preliminary and warrant ongoing research attention. Additional studies that include both college and non-college samples should directly test between-group differences using identical measurement procedures. These studies could explore factors such the level and range of hostile masculinity and promiscuous sex between college students and men in the community, as well as the time frame for measuring sex assault.

Limitations

This study must be interpreted with caution due to a few methodological limitations. First, the sample size was relatively small and participants may not be representative of other populations of community men or sex offenders. It is possible that men who volunteered to participate in this study about sexual attitudes were different from men who were not willing to participate, as it has generally been found that people who participate in sexuality research have more liberal attitudes about sex, including more positive attitudes towards sexuality, less sexual guilt, and more sexual experience (Strassber & Love, 1995). A follow-up study with a larger, more representative sample is necessary to confirm and extend the current results. A second limitation is that all sex offenders were receiving outpatient psychological treatment to address their history of offending. Even though results were reported anonymously, because the sex offenders were in treatment, it is possible they felt pressure to under-report prior offending and rape-supportive attitudes due to social desirability pressures. It would be useful to administer the SES to a group of sex offenders who are not receiving treatment and examine if their responses differ significantly from a group in treatment. Future research should also consider how measures of sexual assault and sexual experience that are common in research settings (e.g. SES: Abbey et al., 2005; Koss et al., 1987), differ from the assessment tools used in clinical practice (e.g., clinical interviews, psychosexual screening tools: Hall, 1996; Hartmann, 1991).

A third limitation is that sexual offense history was gathered via self-reports from sex offenders, which was done to preserve total anonymity. Future research may benefit by conducting a thorough screening of sexual offenses using official court records and treatment notes, and then compare these offense histories to self-reports on the SES.

Conclusions

This study adds to a growing body of literature suggesting that the prevalence of sexual violence remains markedly high. That so many men commit sexual assault suggests that cultural norms, or macrosystem factors, likely contribute to men’s propensity to commit sexual assault (Bolen, 2003; Kalmuss, 2004). Establishing the etiology of sexual assault across ecological levels remains a critical area for future research efforts (Gidycz, Orchowski, & Edwards, 2011; Yeater & O’Donohue, 1999). Such high rates of sexual assault also demonstrate the need for effective primary prevention models. If sexual assault is as high as this and other studies suggest, primary prevention models may be the only way to have a significant impact on reducing this social problem and improving the sexual health of both men and women.

Footnotes

1

When relationship and sexual history were examined in the community sample, one of the 49 community participants revealed a legal history of a sexual offense charge. This participant was charged with a non-contact sexual offense approximately ten years prior to study participation; however, he was not convicted of the crime. Because he was not convicted and never received sex offender treatment, his data were retained in the community sample.

Contributor Information

Laura Widman, University of North Carolina, Chapel Hill, Division of Infectious Diseases.

Michael A. Olson, University of Tennessee, Knoxville, Department of Psychology

Rebecca M. Bolen, University of Tennessee, Knoxville, College of Social Work

References

  1. Abbey A, McAuslan P, Zawacki T, Clinton AM, Buck PO. Attitudinal, experiential, and situational predictors of sexual assault perpetration. Journal of Interpersonal Violence. 2001;16:784–807. doi: 10.1177/088626001016008004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Abbey A, Parkhill M, BeShears R, Cinton-Sherrod A, Zawacki T. Cross-Sectional predictors of sexual assault perpetration in a community sample of single African American and Caucasian men. Aggressive Behavior. 2006;32:54–67. doi: 10.1002/ab.20107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Abbey A, Parkhill MR, Koss MP. The effects of frame of reference on responses to questions about sexual assault victimization and perpetration. Psychology of Women Quarterly. 2005;29:364–373. doi: 10.1111/j.1471-6402.2005.00236.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bolen RM. Child sexual abuse: Prevention or promotion? Social Work. 2003;48:174–185. doi: 10.1093/sw/48.2.174. [DOI] [PubMed] [Google Scholar]
  5. Calhoun KS, Bernat JA, Clum GA, Fame CL. Sexual coercion and attraction to sexual aggression in a community sample of young men. Journal of Interpersonal Violence. 1997;12:392–406. [Google Scholar]
  6. Dean KE, Malamuth NM. Characteristics of men who aggress sexually and of men who imagine aggressing: Risk and moderating variables. Journal of Personality and Social Psychology. 1997;72:449–455. doi: 10.1037//0022-3514.72.2.449. [DOI] [PubMed] [Google Scholar]
  7. Freeman-Longo RE. The adolescent sexual offender: Background and research perspectives. In: Otey EM, Ryan GD, editors. Adolescent sex offenders: Issues in research and treatment. Rockville, MD: U.S. Department of Health & Human Services; 1985. pp. 130–146. Publication No. (ADM) 85-1396. [Google Scholar]
  8. Gidycz CA, Orchowski LM, Edwards M. Primary prevention of sexual violence. In: Koss MP, White JW, Kazdin AE, editors. Violence Against Children. Volume 2: Navigating Solutions. Washington, DC: American Psychological Association; 2011. pp. 159–179. [Google Scholar]
  9. Hall GCN. Theory-Based Assessment, Treatment, and Prevention of Sexual Aggression. New York, NY: Oxford University Press; 1996. [Google Scholar]
  10. Hall GCN, Teten AL, DeGarmo DS, Sue S, Stephens KA. Ethnicity, culture, and sexual aggression: Risk and protective factors. Journal of Consulting and Clinical Psychology. 2005;73:830–840. doi: 10.1037/0022-006X.73.5.830. [DOI] [PubMed] [Google Scholar]
  11. Hartmann U. Psychological evaluation and psychometry. In: Jonas U, Thon WF, Stief GC, editors. Erectile Dysfunction. Berlin: Springer; 1991. pp. 93–103. [Google Scholar]
  12. Kalmuss D. Nonvolitional sex and sexual health. Archives of Sexual Behavior. 2004;33:197–209. doi: 10.1023/B:ASEB.0000026620.99306.64. [DOI] [PubMed] [Google Scholar]
  13. Koss MP. Rape: Scope, impact, interventions, and public policy responses. American Psychologist. 1993;48:1062–1069. [PubMed] [Google Scholar]
  14. Koss MP, Abbey A, Campbell R, Cook S, Norris J, Testa M, Ullman S, West C, White J. Revising the SES: A collaborative process to improve assessment of sexual aggression and victimization. Psychology of Women Quarterly. 2007;31:357–370. [Google Scholar]
  15. Koss MP, Gidycz CA. Sexual experiences survey: Reliability and validity. Journal of Consulting and Clinical Psychology. 1985;53:422–423. doi: 10.1037//0022-006x.53.3.422. [DOI] [PubMed] [Google Scholar]
  16. Koss MP, Gidycz CA, Wisniewski N. The scope of rape: Incidence and prevalence of sexual aggression and victimization in a national sample of higher education students. Journal of Consulting and Clinical Psychology. 1987;55:162–170. doi: 10.1037//0022-006x.55.2.162. [DOI] [PubMed] [Google Scholar]
  17. Kosson DS, Kelly JC, White JW. Psychopathy-related traits predict self-reported sexual aggression among college men. Journal of Interpersonal Violence. 1997;12:241–254. [Google Scholar]
  18. Loh C, Gidycz CA, Lobo TR, Luthra R. A Prospective Analysis of Sexual Assault Perpetration: Risk Factors Related to Perpetrator Characteristics. Journal of Interpersonal Violence. 2005;20:1325–1348. doi: 10.1177/0886260505278528. [DOI] [PubMed] [Google Scholar]
  19. Malamuth NM, Check J. Debriefing effectiveness following exposure to pornographic rape depictions. Journal of Sex Research. 1984;20:1–13. doi: 10.1080/00224498409551203. [DOI] [PubMed] [Google Scholar]
  20. Malamuth NM, Sockloskie JS, Koss MP, Tanaka JS. Characteristics of aggressors against women: Testing a model using a national sample of college students. Journal of Consulting and Clinical Psychology. 1991;59:670–681. doi: 10.1037//0022-006x.59.5.670. [DOI] [PubMed] [Google Scholar]
  21. Malamuth NM, Linz D, Heavy CL, Barner G, Acker M. Using the conflict model of sexual aggression to predict men’s conflict with women: A 10-year follow-up study. Journal of Personality and Social Psychology. 1995;69:353–369. doi: 10.1037//0022-3514.69.2.353. [DOI] [PubMed] [Google Scholar]
  22. Malamuth NM. Criminal and noncriminal sexual aggressors: Integrating psychopathy in a hierarchical-mediational confluence model. Annals of the New York Academy of Sciences. 2003;989:33–58. [PubMed] [Google Scholar]
  23. Merrill LL, Thomsen CJ, Gold SR, Milner JS. Childhood abuse and premilitary sexual assault in male navy recruits. Journal of Consulting and Clinical Psychology. 2001;69:252–261. doi: 10.1037//0022-006x.69.2.252. [DOI] [PubMed] [Google Scholar]
  24. Schewe PA. Preventing violence in relationships: Interventions across the lifespan. Washington, DC: American Psychological Association; 2002. [Google Scholar]
  25. Scully D. Understanding Sexual Violence: A Study of Convicted Rapists. New York: Routledge Press; 1990. [Google Scholar]
  26. Spitzberg BH. An analysis of empirical estimates of sexual aggression victimization and perpetration. Violence and Victims. 1999;14:241–260. [PubMed] [Google Scholar]
  27. U.S. Bureau of the Census. U.S. census bureau news: 2006. Washington D.C.: U.S. Department of Commerce; 2006. [Accessed online April, 2009]. from: http://www.census.gov/PressRelease/www/releases/archives/education/009749.html. [Google Scholar]
  28. Vega V, Malamuth NM. Predicting sexual aggression: The role of pornography in the context of general and specific risk factors. Sexual Aggression. 2007;33:104–117. doi: 10.1002/ab.20172. [DOI] [PubMed] [Google Scholar]
  29. Ward T. Sexual offenders’ cognitive distortions as implicit theories. Aggression and Violent Behavior. 2000;5:491–507. [Google Scholar]
  30. Weinrott MR, Saylor M. Self-report of crimes committed by sex offenders. Journal of Interpersonal Violence. 1991;6:286–300. [Google Scholar]
  31. Wheeler JG, George WH, Dahl BJ. Sexually aggressive college males: Empathy as a moderator in the "Confluence Model" of sexual aggression. Personality and Individual Differences. 2002;33:759–776. [Google Scholar]
  32. Yeater EA, O’Donohue W. Sexual assault prevention programs: Current issues, future directions, and the potential efficacy of interventions with women. Clinical Psychology Review. 1999;13:667–682. doi: 10.1016/s0272-7358(98)00075-0. [DOI] [PubMed] [Google Scholar]

RESOURCES