1 in 4 SA men surveyed admitted to rape - MRC report

High proportion of men interviewed in Eastern Cape and KZN admit to having forced woman or girl into sex

Understanding men's health and use of violence: interface of rape and HIV in South Africa

Executive Summary


South Africa has one of the highest rates of rape reported to the police in the world and the largest number of people living with HIV. The rate of rape perpetration is not known because only a small proportion of rapes are reported to the police. There is considerable concern about the links between these two problems. Obviously HIV can be transmitted in the course of rape and this compounds the human rights violation of the rape. Research has established that men who rape and are physically violent towards partners are at more likely to engage in sexual risk taking than other men and this has raised a concern that they are more likely to be infected with HIV. The aim of this research was to understand the prevalence of rape perpetration in a random sample community-based adult men, to understand factors associated with rape perpetration, and to describe intersections between rape, physical intimate partner violence and HIV.


The study was conducted in three districts in the Eastern Cape and KwaZulu Natal Provinces - spanning geographical areas: rural, urban and city. It was a crosssectional with a two stage random sample. The sample was drawn by Statistics South Africa. Following a cluster design, 222 enumeration areas (ea) were selected and 20 households approached per ea for interview. One man aged 18-49 years interviewed per household. Interviews followed a questionnaire and were administered via APDAs (Audio-enhanced Personal Digital Assistants). A finger prick specimen of blood was requested for HIV testing and collected as a blood spot which was dried.

Blood was tested for HIV in the laboratory of the National Institute for Communicable Diseases in Johannesburg, using ELISA. Ethics approval was given by the Medical Research Council's ethics committee. We completed interviews in 215 of 220 eligible eas (97.7%) and we completed interviews in 1,738 of 2,298 (75.6% ) of the enumerated and eligible households.


The sample included men of all racial groups and of a range of different socioeconomic backgrounds. Half of the men were under 25 years of age and 70% were under 30. The population was some what younger than men in the general population.

Rape prevalence

Rape of a woman or girl had been perpetrated by 27.6% of the men interviewed and 4.6% of men had raped in the past year. Rape of a current or ex-girlfriend was disclosed by 14.3% of men. Since many men had raped more than once, rape of a woman or girl who was not a partner was actually more often reported than rape of partners. In all only 4.6% of men had raped a partner and not raped a woman who was not a partner (i.e. an acquaintance or stranger). 11.7% of men had raped an acquaintance or stranger (but not a partner) and 9.7% had raped both. In total, 8.9% said they had raped with one or more other perpetrators when a woman didn't consent to sex, was forced or when she was too drunk to stop them. Rape of men and boys was also reported, 2.9% said they had done this. Attempted rape was reported by 16.8% of men and 5.3% of men said they had done so in the previous 12 months.

Patterns of rape

Nearly one in two of the men who raped (46.3%) said they had raped more than one woman or girl. In all, 23.2% of men said they had raped 2-3 women, 8.4% had raped 4-5 women, 7.1% said they had raped 6-10 and 7.7% said they had raped more than 10 women or girls.

Asked about their age at the first time they had forced a woman or girl into sex, 9.8% said they were under 10 years old, 16.4% were 10-14 years old, 46.5% were 15-19 years old, 18.6% were 20-24 years old, 6.9% were 25-29 years and 1.9% were 30 or older.

Factors associated with raping

Age was significantly associated with the likelihood of having raped, with men aged 20-40 were more likely to have raped than younger or older men. Education was also associated, with men who had raped being significantly better educated, although they were not more likely to have a tertiary qualification. There were significant racial differences in rape reporting, mostly notably men who were Coloured were over represented among those who had raped. Men who had raped were significantly more likely to have earnings of over R500 per month, although they were not more likely to be in the top income bracket, over R10 000. Men who raped were more likely to have occasional work and less likely to have never worked at all.

Parental absence was significantly associated with raping, as was the quality of affective relationships with parents was related to raping. Men who raped perceived both their fathers and mothers to be significantly less kind (p<0.0001). Rape was associated with significantly greater degrees of exposure to trauma in childhood.

Teasing and harassment, or bullying, were reported by many of the men in their childhood. Over half of the men had experienced this themselves (54%) and somewhat fewer (40%) had teased and harassed others. Both experience of bullying and being bullied was much more common among men who raped. Delinquent and criminal behaviour were more common among men who raped. Men who raped were much more likely to have been involved in theft and, with the exception of legal gun ownership, they were very much more likely to have been involved with weapons, gangs and to have been arrested and imprisoned.

Men who disclosed having raped were significantly more likely to have engaged with a range of other risky sexual behaviours. They were more likely to have ever had more than 20 sexual partners, transactional sex, sex with a prostitute, heavy alcohol consumption, to have been physically violent towards a partner, raped a man and not to have used a condom consistently in the past year.

Associations between rape and HIV

The HIV prevalence among men who had raped was 19.6% and 18.1% among those who had never raped. This difference was not significant (p=0.53). The HIV prevalence was lower, 12.7%, among those who had raped in the past year. Men who had raped another man, in contrast, had a higher prevalence of HIV (27.8%).

The most striking feature of the age-specific HIV prevalence, when plotted for men who have and have not raped, is the very high prevalence of HIV for all men in this sample. The prevalence among all men aged 25-45 was in excess of 25%, and among those aged 30-39 years, over 40%. When examined by rape perpetration status, however, there was no overall difference between the HIV prevalence of men who had raped women and those who had never raped.

Associations between physical intimate partner violence and HIV

In all 42.4% of men had been physically violent to an intimate partner (current or exgirlfriend or wife). Asked about physical violence in the past year, 14.0% (95%CI 12.4, 15.7) of men disclosed perpetration. Men who disclosed violence were very much more likely to have engaged in a range of risky sexual behaviour, as well as to have raped and been raped.

A logistic regression model of factors associated with having HIV showed that men who had been physically violent to a partner on more than one occasion were significantly more likely to have HIV (OR 1.48 95% CI 1.01, 2.17, p=0.04). Other associated factors were being African, rather than of another race group, being 25 or older, and having had a genital ulcer. Those who had completed matric at school or attended tertiary education and those who were circumcised were less likely to be infected.


The findings highlight the very high prevalence of rape in South Africa and the high prevalence of HIV in the adult population. The prevalence of rape has similarities to that found in other studies in South Africa. The very high prevalence shows that generally rape is far too common, and its origins too deeply embedded in ideas about South African manhood, for the problem which can be predominantly addressed through strategies of apprehension and prosecution of perpetrators.

A much broader approach to rape prevention is required. This must entail intervening on the key drivers of the problem which include ideas of masculinity, predicted on marked gender hierarchy and sexual entitlement of men. Efforts to change these require interventions on structural dimensions of men's lives, notably education and opportunities for employment and advancement. Our study suggests that the pathway which leads to these ideas and the practices of rape and other forms of violence towards women starts in childhood and strengthening families, and protecting children from exposure to adversity in childhood are critical for ensuring that men in the population develop psychologically as pro-social members of society.

A very surprising finding of our study was that men who raped were no more likely to have HIV than men who hadn't raped. Yet one of the very important findings is the very high HIV prevalence found in all the men, but particularly those aged 25-45.

This provides a salient reminder of how likely it is that a man who rapes has HIV, irrespective of whether he has more than another man. Clearly post-exposure prophylaxis for HIV after rape is a very important part of post-rape care for victims who are HIV negative. The fact that so many rapes are gang rapes, or involve multiple acts of sex penetration (30% in cases reported to the police) and the high prevalence of injuries (at least 58% in rapes reported to the police) (Vetten et al 2008) further supports the very considerable risk of exposure to HIV of victims at the time of rape and risk of transmission through rape.

The factors that were shown to be associated with having HIV in the study are in many respects unsurprising. Its well known that the epidemic has disproportionately spread amongst Africans, that the most well educated are relatively more protected, that having genital ulcers increases the likelihood of having HIV and that circumcision is protective. What has previously been suspected, but not shown in research, is that men who are physically violent towards their intimate partners are more likely to have HIV. This finding is completely congruent with the documented association between being violent and sexual risk taking, and indeed the finding that women who experience violence are more likely to have HIV (Dunkle et al 2004).

This is explained by an underlying construction of masculinity which is predicated on use of violent and sexually behaviours. It has been argued that this is a key driver of the HIV epidemic and our finding supports this. HIV prevention needs to embrace and incorporate promoting more gender equitable models of masculinity. The intervention Stepping Stones, has been shown to effectively do this, and should be promoted (Jewkes et al 2008).


1. Rape prevention must focus centrally on changing social norms around masculinity and sexual entitlement, and addressing the structural underpinnings of rape.

2. Post-exposure prophylaxis is a critical dimension of post-rape care, but it is just one dimension and a comprehensive care package needs to be delivered to all victims and should include support for the psychological responses to rape.

3. HIV prevention must embrace and incorporate promoting more gender equitable models of masculinity. Intervention that do this effectively must be promoted as part of HIV prevention

Issued by the Gender and Health Research Unit, South African Medical Research Council, June 18 2009

Click here to sign up to receive our free daily headline email newsletter