Frequency of self-reported sexual aggression and victimization in Brazil : a literature review

The lack of official data on rape has been a challenge for researchers in Brazil. Two recently published studies were based on law enforcement and medical records. Although these studies represent important progress in research on rape in the country, they have several limitations. In order to obtain more realistic rates, the current article reviews Brazilian studies on self-reported sexual aggression and victimization in individuals over 14 years of age. Forty-one studies were identified through electronic searches and reference verification. From 1% to 40% of women and 1% to 35% of men reported some form of victimization in the previous year. The male perpetration incidence ranged from 2% to 44%. Despite the wide variability, these rates were much higher than those provided by official data. The results suggest that sexual orientation is associated with vulnerability. Mixed findings were found concerning race. Most studies were based on convenience samples and focused on female victimization. Male victimization has received increasing attention, but studies on self-reported perpetration are still limited. Sexual Violence; Rape; Crime Victims Correspondence L. Winzer Institute for Population and Social Research, Mahidol University. Salaya, Nakhon Pathon, Bangkok – 73170, Thailand. lylla.winzer@yahoo.com 1 Institute for Population and Social Research, Mahidol University, Bangkok, Thailand.

For decades, the lack of official data on rape has posed a challenge for public health researchers in Brazil.Two official documents published in 2013 and 2014 had great repercussions in the media and social networks, fueling the discussion on rape in the country.The first report was published by the Brazilian National Information System for Public Safety (SINESP) 1 and provided data on rape based on law enforcement records in municipalities with more than 100,000 inhabitants.The second document was prepared by the Brazilian Institute of Applied Economic Research (IPEA) 2 and provides data on rape cases assisted by public health services.Based on law enforcement and medical records, they were among the first documents to report official nationwide data on the occurrence of rape, thereby facilitating comparisons among the country's five major regions and identification of increasing or decreasing trends in the preceding years.However, the reports showed several limitations.From a public health perspective, they should be seen neither as the yardstick for studying rape nor as the only source of information for policymakers in Brazil.
The first limitation is that both reports only recorded victims that had turned to the police 1 or the public health care system 2 .According to self-report studies in Brazil, only 2.4% of female and 0% of male sexual assault victims reported their cases to law enforcement 3 .Victims' access to the health system is also limited 4 , and there is evidence that many health professionals are unqualified to identify and appropriately refer victims to law enforcement 5,6 .This explains why the cases identified and officially recorded by public health services are much fewer than those obtained through self-report questionnaires in the same services 7 .Self-report studies also include other forms of sexual aggression (e.g., sexual contact, sexual coercion) in addition to rape, which partly explains why self-report studies provide higher rates than official records.However, studies suggest that rape is highly underreported, especially in middle and low-income countries, and that "official" data on rape victims represent only a small minority of actual cases 8 .According to a recent report in the United States, the number of rape incidents identified by the National Intimate Partner and Sexual Violence Survey in 2010 (1,270,000 cases) was nearly 15 times higher than the figures reported to the Federal Bureau of Investigation (FBI) in the same year (85,593) 9 .Middle-income countries like South Africa and India show similar patterns 10,11 .Even if underreporting occurs worldwide, the problem may prove even more critical in middle and low-income countries, due to lack of appropri-ate infrastructure to support crime reporting 10 and lack of trust in the justice system.The Brazilian Ministry of Health estimates that less than 10% of rape cases are reported to the police 12 .In fact, according to the SINESP report 1 , rape incidence in 2012 ranged from 0.27 to 61.0 per 100,000 inhabitants across Brazil's states.Absolute rape figures ranged from 2 in Acre State to 12,888 in São Paulo State, rates that appear to be grossly underestimated.In addition, the IPEA report suggests that at least among adults, the more the act resembles the "common sense" definition of rape (i.e., an act perpetrated by a male stranger against a female victim in a public space, at night and using force), the more likely the victim seeks medical assistance and becomes part of official medical records.According to the same report 2 , among adult victims, 97% of aggressors were male and 98% of victims were female.The perpetrator was unknown to the victim in 61% of cases.Of those, 61% involved physical force and at least 50% occurred on the street.Nearly 70% of adult rape cases occurred between 6:00 PM and 6:00 AM.These data differ sharply from those provided by studies in the general population or convenience samples, where both women and men report sexual victimization by intimate partners, occurring in domestic settings, and where other forms of coercion were applied (e.g., taking advantage of victim's incapacitated state) 13,14 .Besides the underreporting problem, the IPEA report did not included victims treated at private clinics, which mainly serve the middle and upper classes in Brazil.
A second limitation is the way sexual aggression is defined in the IPEA and SINESP reports.Both reports adopt the legal definition of rape.Article 213 of the Brazilian Penal Code 15 defines rape as "the use of violence or serious threat to force someone to have sexual intercourse or to commit or force someone to engage in other lewd and lascivious acts".Cases that involve alcohol or drug intoxication by the victim are addressed by an additional category called "rape of a vulnerable person" (Section 217-A, § 1), defined as "to have sexual intercourse or commit other lewd acts with someone who lacks the necessary insight into such acts, or who cannot resist for any other reason".However, the law enforcement system provides limited information for researchers.Public health professionals are usually interested in a broad range of nonconsensual sexual activities, defined as sexual aggression, that threaten the reproductive health of women and men, even when rape is not consummated.Sexual aggression constitutes unwanted sexual contact (e.g.kissing, rubbing against another person's private parts) up to any form of penetration (e.g.into a woman's vagina or someone's anus or mouth).Such acts can be perpetrated through different aggressive strategies, such as verbal coercion, the use or threat of physical force, or exploitation of the victim's incapacitated state 16 .This broader definition results in higher rates of sexual aggression than the official records, but offers some advantages.It enables researchers to identify a continuum from less to more severe forms of sexual aggression in a society.It also challenges the common sense belief that rape is limited to violent sexual assault by a stranger in a dark place 17 and shows that it can be perpetrated by anyone, including intimate partners, through more subtle coercive strategies rather than physical force 18 .
Therefore, two important challenges for public health researchers are the detection of "hidden" rape cases and identification of other forms of sexual aggression.Self-report and more inclusive definitions of sexual aggression are strategies to deal with these challenges.Self-report interviews or questionnaires are simple and convenient to apply and are one of the few available methods to assess private behaviors such as sexual behavior 19 .Although data based on selfreport can be subject to recall bias, social desirability bias, shame, and denial 20 , self-reporting is still an effective method for assessing "obscure figures", providing a more trustworthy picture of rape levels 18 .The method's limitations can be reduced through an atmosphere of trust and respect, and the assurance of voluntary participation and anonymity.As a result, studies based on self-reporting have identified alarming rates of sexual aggression, confirming the huge gap between official data and the real scope of rape.

The current study
Although the SINESP and IPEA reports represent important strides in research on rape in Brazil, both documents have some limitations.They are unable to identify the "obscure figures" of rape and other forms of nonconsensual sexual activities.In order to obtain more realistic incidence rates for sexual aggression, researchers often use self-report methods and focus on a broad spectrum of coercive sexual acts rather than rape only.The current study thus aims to investigate the frequency of sexual aggression in Brazil through self-report by victims and perpetrators.A literature review was conducted on studies that identified the prevalence or incidence of self-reported sexual aggression and victimization among women and men over age 14 (the age of consent in Brazil).

Method
Databases and key word search strategy Electronic searches were conducted in SciELO, MEDLINE via PubMed, Scopus, PsycINFO, and Web of Knowledge.Google Scholar and reference checking were also used to identify potential articles in the gray literature.Target descriptors were intimate partner violence OR sexual aggression OR sexual violence OR sexual coercion OR sexual assault OR rape AND Brazil in both English and Portuguese.To keep the review manageable, PhD dissertations were not included.No time limit was specified for the search.
The review included: (1) studies on sexual aggression in females and males over age 14 and (2) studies that included any form of sexual aggression (e.g., from unwanted sexual touching to rape) and any form of coercive strategy (e.g., verbal coercion, physical force or threat, taking advantage of the victim's incapacitated state) committed by anyone known or unknown to the victim.Publications in English, French, German, Portuguese, and Spanish were potentially considered for the review.Exclusion criteria were: (1) studies on sexual aggression/rape in mixed samples of children, adolescents, and adults, where it was unfeasible to separate victimization of adults, adolescents, and children (except for two studies that included pregnant women older than 13 years 21,22 ); (2) when rates of sexual aggression were presented in combination with other forms of aggression (e.g., physical and psychological), thus preventing the separate identification of rates of sexual aggression individually; and (3) studies on sexual harassment, forced marriage or cohabitation, forced abortion, genital mutilation, obligatory inspection for virginity, forced prostitution, and sex trafficking.According to the World Health Organization (WHO), the latter terms fall under the broad definition of sexual violence 8 .However, the present study focused on nonconsensual sexual activities from unwanted sexual contact to any form of penetration, previously defined as sexual aggression.
The search was conducted by one researcher and repeated by the same individual before submission to verify accuracy.All studies that met the inclusion criteria were selected for the review, regardless of their methodological rigor.Methodological limitations will be addressed later in the discussion.

Results
A total of 595 studies were initially selected from the literature databases.Two additional articles were found through reference checking and three through the Google search engine.Figure 1 summarizes the selection of studies.Forty-one articles met the inclusion criteria and were selected for the review.All studies were cross-sectional, and most were conducted after the year 2000.In order to understand possible variation between studies in the rates of sexual aggression, the respective methods were also analyzed for time frame, target population, study location, sampling methods, and definitions.
Self-reported sexual victimization in women and men

• Time window
Of the 41 selected studies, 27 provided lifetime prevalence data on sexual victimization, and six provided incidence data for the previous year (Table 1).Other studies collected information on sexual victimization using different time frames, such as after age 14 years, after age 15, after age 16, after age 18, during pregnancy, since starting university, and during the intimate relationship.Table 1 shows the rates of sexual victimization, defined here as the number of victims divided by the total sample.Lifetime prevalence of sexual victimization ranged from 7% to 39% in women and 5% to 16% in men.Incidence rates for the previous year varied widely.Rates in women ranged from 1.3% for forced sex to 40% for any form of sexual victimization (including partner's refusal to use a condom).The corresponding rates for men varied from 1.4% to 35%.

• Region
More than a half of the studies were conducted exclusively in the Southeast of Brazil, especially in São Paulo State.Nine studies included samples of the five macro-regions of Brazil, and two other studies compared data from the Northeast with the South or Southeast (Table 2).Of those 11, only four provided data on differences between regions.One study found that female and male prevalence rates of victimization were higher in the North and Northeast, but the difference was only statistically significant in men 13 .A second study showed that the total rate of 16.7% for sexual coercion among young people in Salvador (Northeast Brazil) was significantly higher than the 10.9% in Porto Alegre (South) and 13.1% in Rio de Janeiro (Southeast) 3 .Two other publications also found that the South showed the lowest victimization rates, but the authors did not test the difference for statistical significance 23,24 .

• Population
Although the majority of the studies had relatively large samples, most were based on convenience samples (Table 2).Three studies used randomized samples from health service units 25,26,27 and one drew a sample from a vulnerable neighborhood 28 .Of the nine studies held in Brazil's five major regions, only one drew on a representative sample of the Brazilian population 13 , a second on a representative sample of high school sophomores 29 , and a third on a stratified sampling of psychiatric patients 30 .The other six studies, despite their large samples, were based on quota sampling (in which participants are selected according to fixed quotas based on age and rural versus urban areas) 23,24 .** About 24% of the sample were adolescents, but the authors did not specify whether they were younger or older than 14 years; *** The authors did not specify how many women were 13 years old (probably a minority).
convenience sampling 31,32,33 , and respondentdriven sampling (a variation of the snowball technique in which the sample is weighted to compensate for the fact that participants are not randomly assigned) 34 , thus not meeting the criteria for statistical representativeness.
Nineteen studies included clinical samples (e.g., pregnant women, postpartum women, patients with sexually transmitted diseases, psychiatric patients, and HIV-positive men who have sex with women).Lifetime prevalence for clinical groups ranged from 7.1% to 39.1% in women and 6.1% to 24.1% in men.In pregnant women, rates of sexual victimization during pregnancy varied from 0.003 to 1.3.Six studies included samples of high school or college students.Twenty-six studies included women only, four studies men only, and 11 studies both women and men (Table 2).Five studies addressed same-sex sexual aggression 13,14,29,34,35 , and 12 studies provided information on race issues 3,13,21,23,24,26,30,32,34,35,36,37 .
The highest victimization rates were reported by persons with unconventional sexual behavior, such as fetishism, voyeurism, incest, threesome, exhibitionism, sadomasochism, and group sex (56.4% for women and 85.5% for men) 31 , but this study did not specify the time window.The highest previous-year incidence rate was reported by a sample of university students (38.5% in women and 34.5% in men) 38 .Three other studies also reported high victimization rates in university students 14,39,40 .
Regarding sexual orientation, one study found that sexual victimization rates were almost eight times higher in female and male homosexuals than in heterosexuals 29 .Two other studies also observed that self-reported victimization in female and male homo-and bisexuals was slightly more frequent than in heterosexuals.However, in the first study the difference was only statistically significant in males 13 .In the second, the samples of homo-and bisexuals were too small to test for statistical significance 14 .
The results on race issues were mixed.One group of studies found that white individuals were significantly less likely to report sexual victimization 3,13,26 .A second group of studies found no significant difference in victimization rates according to skin color 21,30,34,35,36 .A third group (four studies) did not test for statistical significance and found contradictory differences in rates.For example, one study found that Asian-descendant and indigenous women reported higher rates of victimization than white, brown, or black women 24 .Another study found that white and brown individuals reported higher rates than black, Asian-descendant, or indigenous persons 32 .Two other studies found that white women reported victimization less frequently than the other groups 23,37 .

• Definition of sexual victimization
The definition of sexual victimization varied widely (Table 3).The number and wording of questions potentially affected the resulting rates.Several studies proposed their own operationalization, using from one to five screening questions.The questionnaire adapted from the World Health Organization Multicentre Study was the most frequently used measure for assessing sexual victimization in Brazilian studies.The WHO instrument includes three questions on episodes of physically forced sexual intercourse, sexual intercourse under threat, and forced degrading sexual acts (Table 3).Two other internationally used instruments, the Revised Conflict Tactics Scales (CTS2) and the Sexual Experiences Survey (SES), were also applied to Brazilian samples.
In general, studies that considered the same time frame and used the same definition of sexual aggression found comparable data.For example, using the same instrument to assess sexual victimization, two studies in the five major regions of Brazil found similar rates of lifetime female sexual victimization (13% in 2004 23 and 10% in 2010 24 ).Studies that used CTS2 in university students also found similar previous-year incidence rates, ranging from 39% to 40% in women and 30% to 37% in men 38,39,40 .Of the twelve studies that adopted the WHO questionnaire, three found similar lifetime prevalence rates for sexual victimization, varying from 9.8% to 11.8% in São Paulo State 13,26,41 .However, the rates were slightly lower in the South (8% in Porto Alegre 42 ) and slightly higher in the Northeast (14.3% in the Zona da Mata region of Pernambuco State 41 ).Rates were also considerably higher in economically vulnerable areas, e.g., 28.8% in a community sample in Brasília (Central) 28 and 26% in a large clinical sample in São Paulo 7 .

Self-reported perpetration of sexual aggression by women and men
Available data on perpetration of sexual aggression in Brazil are still very limited.A total of nine studies addressed the incidence or prevalence of self-reported sexual aggression (Table 4).Of these, two investigated lifetime prevalence of sexual aggression.The other seven studies investigated sexual aggression using different time frames, such as in the previous year, after age 14, after age 18, during pregnancy, since starting university, and during an intimate relationship.
Regarding study population, five studies involved community samples and four involved clinical samples.Two studies included female respondents only, three studies included men and women, and four studies included men only.Specific social groups were addressed by seven studies: university students 14,38,40,43 , pregnant women 44 , men who have sex with men 35 , and HIV-positive men who have sex with women 45 .
Rates varied from 2% to 44% in men and from 3% to 32% in women.However, differences in time frames and whether the aggression was   ** These studies addressed both victimization and perpetration.In order to assess self-reported perpetration, they used parallel versions for perpetration from the items described in this table ; *** This paper also addressed perpetration.For the perpetration version, two questions were asked: (1) Have you ever forced your partner to have sexual relations when she did not want to? and (2) Have you ever forced your partner to do certain sexual practices that she did not like?Table 4 Prevalence and incidence of self-reported perpetration of sexual aggression in Brazilian samples.

Author(s) Time window Sexual perpetration (%) Victims
Women Men  43 Since the beginning of college -

Any
Note: all studies addressed both sexual victimization and perpetration.More information on the operationalization of perpetration of sexual aggression, region and characteristics of the sample can be found in Tables 2 and 3.
* Studies with clinical samples of healthcare services.
committed against intimate partners or against any persons limit the data's comparability.Differences in perpetration rates according to sexual orientation and skin color were also addressed.One study found that female and male bisexuals and male homosexuals were more likely to report perpetration than heterosexuals.However, the authors did not test for statistical significance 14 .Two other studies found that skin color was not significantly associated with perpetration 43,45 .

Discussion
Sexual aggression has attracted increasing political attention and scientific research in Brazil since 2000.Using self-report methods, research-ers have identified sexual victimization in the general population and in community and clinical samples.A review of such studies is important for obtaining more realistic rates, showing that law enforcement and medical records are limited and may have misdirected public policies addressing the problem of sexual aggression.
The current article reviewed studies on the incidence and prevalence of self-reported sexual aggression and victimization among women and men in Brazil.Despite the heterogeneity of methods and limited comparability of data, the review showed that sexual aggression is a pervasive problem in young adulthood 3,14,29,38,39,40,43 and is much more prevalent than identified by official data.Studies that used the same definition of sexual victimization and same time interval showed comparable rates, suggesting consistency of the instrument across samples.Differences in rates among studies that used the same instrument could be attributed to other factors, such as time frame, location, and socioeconomic vulnerability.For example, people from the North and Northeast of Brazil or who lived in economically deprived areas were more likely to report sexual victimization 7,13,28 .
Concerning gender, most studies focused on female victimization.Nevertheless, an increasing number of studies have addressed male victimization in the last decade.Studies on self-reported perpetration are still few, especially among women.Despite the myth that men can only be sexually coerced by other men 46 , the majority of men who report victimization are heterosexuals and mostly report that their aggressors are women 3,14 .However, the percentage of victimization was higher in male homosexuals and bisexuals than in heterosexuals 13,14,29 .Still, regardless of sexual orientation, victimization in men has been ignored for some time by researchers and authorities.Male victimization does not minimize or justify male aggression against women, which is highly prevalent and appears to have more severe consequences than the opposite case 47 , but this shows that public health professionals should be aware of male victimization and that health services should be prepared to receive and treat male victims.
The findings were inconclusive concerning an association between sexual aggression and skin color.The studies were not clear on their definition of skin color, especially in a country like Brazil with a strong history of miscegenation.Some studies only adopted a dichotomy (white vs. non-white; or black vs. non-black), making impossible to identify other groups such as brown, indigenous, or Asian-descendant.The review's findings also contrast with those provided by official medical records 2 , according to which, in Brazil, white and brown women are the most likely to be rape victims.
High incidence rates for sexual victimization in university students corroborate the international literature that this group may be at increased risk of sexual aggression 48 .Brazilian studies found the highest incidence rates of victimization in this group.However, it was not possible to say with certainty that this group is at greater risk than the general population, for two reasons: the two samples were not compared to each other, and the rates were obtained by different measures.Higher victimization rates in university students could be attributed to the instruments used, like CTS2 and SES 14,38,39,40 .These instruments consist of multiple items with behaviorally specific descriptions of unwanted sexual acts, which prevents vagueness.For example, instead of asking, "Have you ever been raped?", the SES applies thorough descriptions of behaviors that correspond to the legal definition of rape (e.g."A man put his penis, or someone put their fingers or objects into my vagina without my consent", "A man put his penis, or someone put their fingers or objects into my anus without my consent?", and "Someone had oral sex with me or made me have oral sex with them without my consent") 16 .Another advantage of these two instruments is that they assessed coercive acts that were usually neglected by other studies, such as verbal coercion (e.g., insistence, threat to spread rumors about the victim's sexuality), taking advantage of victims incapacitated state (e.g., when the victim was intoxicated and unable to defend her/himself) 16 , and refusal by the partner to use a condom during sexual intercourse 38 .The more inclusive and meticulous definitions of sexual aggression may partly explain the inflated rates in university students.
The results of the present review allow identifying some gaps in Brazilian research on sexual aggression.First, the majority of studies in Brazil relied on convenience samples, which provide results that cannot be generalized to the overall Brazilian population.Random selection of participants from a health service or from a vulnerable neighborhood also does not make the sample representative of all health services or vulnerable neighborhoods, much less of the general population.Randomly chosen respondents from a single health service makes up a representative sample of the users of that service only.Besides, more than a half of the studies were conducted exclusively in the Southeast, the richest region of the country and where research institutes are concentrated, revealing the imbalance of available information on self-reported sexual aggression and victimization across the five major geographic regions.Second, sexual orientation and race were occasionally addressed, but it is still unclear how they might be related to vulnerability.Third, many studies were unable to differentiate between sexual abuse (before age 14) and sexual aggression in adulthood (after age 14).Although childhood sexual abuse may act as a risk factor for later sexual victimization 49 , the two phenomena relate to distinct underlying mechanisms.Fourth, in addition to the lack of information on time frame, location, and sample age, some studies failed to provide clear information on how they defined sexual aggression.Other studies applied only a single or a few screening questions.These short instruments tend to use generic terms like "being forced" or "being raped", which leave room for interpretation and may elicit lower rates 50 .Respondents might have disregarded other forms of aggressive sexual acts (e.g., taking advantage of a victim's inebriation), although such behaviors would legally qualify as rape.The use of behaviorally specific descriptions of coercive acts prevents ambiguity and ensures more accurate data 16,50 .
To address those gaps and better orient public policies, I outline some suggestions for future research in Brazil: (1) self-report studies on sexual aggression and victimization with representative samples of the Brazilian population, identifying the differences across the five major geographic regions; (2) studies that address sexual aggression from a developmental and longitudinal perspective, from young age until late adulthood, seeking to understand the onset, severity, and persistence of the problem over the years; few Brazilian studies have addressed sexual victimization and perpetration prospectively 49,51 ; (3) studies that differentiate between childhood sexual abuse and sexual aggression i.e., sexual victimization before and after age 14, respectively; (4) studies that address sexual aggression and victimization in both men and women, taking sexual orientation and skin color into consideration; (5) studies that address incidence of sexual aggression in the previous year, using a behaviorally specific description of the legal definition of rape in Brazil, such as proposed by SES 16 , so that rates can be directly compared to the official data on rape in Brazil; (6) studies that confirm if, and explain why, specific social groups (e.g., homosexuals and bisexuals, and university students) may have higher vulnerability for sexual victimization and perpetration.
Rape and other forms of sexual aggression involve enormous costs, including medical (pain, HIV, pregnancy), psychological (fear, distress), and social and economic (absenteeism, reduced earning capacity) 52 .From a public health perspective, sexual aggression should be understood as a continuum of violence that threatens the well-being of victims.Police and medical records in Brazil fail to account for the victims that are of special public health interest: those who do not or cannot access medical care and the justice system, but who may be in need of help.If sexual aggression is underreported, numerous victims remain subjected to physical, mental, and social trauma without receiving any form of medical, psychological, or legal support.Public health should address the huge gap between official data and the problem's real scope in Brazil and advocate for population-based policies that facilitate victims' access to the health and legal system.

Figure 1 Selection
Figure 1 Selection flow diagram.
Zotareli et al. (2012) 43 ** Having been physically forced to have sexual intercourse against their will; Having had sexual intercourse because they were afraid of what the aggressor might do; Having had sexual intercourse despite not wanting to because of a belief that it was the aggressor's right; Having been forced to subject themselves to a degrading or humiliating sexual act; Someone having agreed to use a condom during intercourse but failed to comply with this agreement; Someone having refused to use a condom during sexual intercourse * This questionnaire includes 3 questions: (1) Has he/she physically forced you to have sexual intercourse when you did not want to?; (2) Did you ever have sexual intercourse when you did not want to because you were afraid of what he/she might do?; and (3) Has he/she forced you to do something sexual that you found degrading or humiliating?; Violencia Sexual; Violación; Victimas de Crimen Submitted on 07/Aug/2015 Final version resubmitted on 14/Mar/2016 Approved on 02/May/2016

Table 1
Prevalence and incidence of self-reported sexual victimization among Brazilian samples.

Table 2
Characteristics of studies on self-reported sexual victimization in Brazilian samples.

Table 2
NS: not specified; SES: Sexual Experiences Survey; USB: unconventional sexual behavior.* Studies with clinical samples from healthcare services;

Table 3
Definition of sexual victimization by studies that investigated the frequency of self-reported sexual victimization in Brazilian samples 13e forced you to have sex when you did not want it"; "He forced you into sexual acts that you did not like"; "He 57 Before you were 16 years old (age 15 or younger), did someone in your family try to make you do sexual things or watch sexual things?; Before you were 16 years old (age 15 or younger), did someone other than a family member try to make you do sexual things or watch sexual things?; Since the age of 16 (16 or older), was there a time when someone forced you to engage in sex acts that you really did not want?"Tried to touch my body"; "Touched my body"; "Had forced sexual relations"13Questionnaire adapted from the World Health Organization Multicentre Study * Schraiber et al. (2012) 61 *** Has anyone ever forced you to have sexual intercourse when you did not want to? (continues)