Southern African Social Work Students ’ Acceptance of Rape Myths

Despite numerous interventions to promote gender equality, sub-Saharan Africa has one of the highest prevalence rates of non-partner sexual assault in the world, thus constituting a major social and public health issue in the region. As social workers frequently provide services to this population, an exploratory cross-sectional study was conducted to explore rape myth acceptance among undergraduate social work students studying in Namibia. Findings revealed the positive influence of social work education in reducing rape myth acceptance, as well as highlighting the influence of age, gender, country of origin, self-identification as a feminist, and religiosity on rape myth acceptance among this population.


Introduction
Rape myths suggest that women are responsible for any acts of sexual assault because of their words, actions, clothing choices, or alcohol or drug consumption (Lonsway and Fitzgerald 1994). These beliefs shift blame from the perpetrator of sexual assaults to those who are victims. These myths commonly blame victims, absolve the perpetrator of responsibility, and downplay the inherent violence and violation that occurs during a rape (Bohner et al. 2006). When individuals in a society are willing to accept these rape myths, it is more likely that a culture of rape develops. This term describes an environment in which rape is pervasive and normalized due to social attitudes towards gender, misogyny, and gender inequality (Buchwald et al. 2005). Many aspects of a rape culture are evident in Southern Africa, including pervasive and persistent patriarchal worldviews, sexual objectification of women, failure to acknowledge the true incidence of rape, as well as lack of awareness and acknowledgement of the multifaceted nature of the harm caused by sexual violence (Buchwald et al. 2005). These cultures provide an enabling environment for the support of rape myths, both subtle and overt, and ultimately contribute to the persistence of sexual assault as a major social and public health issue in the region.

Gender-Based Violence in Southern Africa
Gender-based violence (GBV) refers to violence associated with normative gender role expectations, as well as unequal power dynamics within the context of a specific society (Bloom 2008, p. 14). While GBV affects males, the primary targets of GBV worldwide are women and girls, who often are perceived as often having lower socioeconomic status, less power, fewer options, and access to fewer resources than their male counterparts (United Nations Population Fund 2008). GBV includes not only physical violence but also emotional and sexual violence.
It is estimated that approximately 35% of women worldwide have experienced physical or sexual violence committed by either a partner or non-partner (World Health Organization 2017).
More specifically, the World Health Organization (WHO) estimates that 36.5% of Sub-Saharan African women have experienced some form of violence in their lifetime. Regionally, Africa has the third highest rate of GBV, lower than only Southeast Asia and the Middle East/North Africa region. However, when examining only sexual violence committed by non-partners, Africa has the highest regional prevalence rate in the world, with an estimated prevalence rate of 11.9% (Confidence Interval (CI): 8.5%-15.3%) (WHO). Other recent research provides even more insight into regional differences, and concrete data related to Sub-Saharan Africa. In central sub-Saharan Africa, 21% of respondents reported lifetime non-partner sexual violence, and in southern sub-Saharan Africa, 17.4% reported the same experience (Abrahams et al. 2014).
A 2010 study that examined the prevalence of adverse childhood events among rural South African youth revealed that 39.1% of women and 16.7% of men had experienced childhood sexual abuse (Jewkes et al. 2010). Similarly, 24% of females in Namibia who reported sexual debut between 15-24 years indicated that their first sexual encounter involved forced sex. Among those respondents who reported sexual debut between 10-14 years of age, 42% reported that their first encounter involved force (United Nations Children's Fund 2011). However, it is not just adolescent and young adult females who are at-risk of sexual assault given that 33% of married women in Namibia report having experienced physical, emotional, and/or sexual violence, with 50% of the perpetrators being the husband or partner (United Nations Children's Fund 2007).
In 2016, the police recorded 939 cases of rape and 222 of attempted rape in Namibia (Ministry of Gender Equality and Child Welfare 2016). This is consistent with recent data from the Namibian Police identifying rape as the most prevalent crime in the country, with approximately 1075 reported cases of rape nationwide annually from 2009-2012(Namibian Police 2013. Notably, approximately one-third of all rape complaints during this period were reported by females under the age of 18. However, rape and all forms of GBV remain underreported because of the widespread cultural acceptance of violence perpetrated on the basis of gender. This is supported by research which revealed that approximately one in three Namibian women feel that it is justifiable that male partners beat their female partners (Ministry of Gender Equality and Child Welfare 2009).
A review of the literature revealed evidence of association between several independent variables, including gender, religiosity, age, education level, and rape myth acceptance. For example, previous research has explored the attitudes of police officers and found that male officers were more likely to endorse rape myths than their female counterparts (Brown and King 1998). Further, a recent study from Africa found that sex, age, marital status, and level of education were predictors of rape myth acceptance (Abeid et al. 2015). Finally, in a recent study of university students in the USA, religiosity was found to be associated with rape myth acceptance. Specifically, participants who identified as Roman Catholic or Protestant endorsed rape myths at significantly higher levels than their agnostic or atheist counterparts (Barnett et al. 2018). It was based on the review of existing literature that the independent variables in the current study were selected.

The Role of Social Workers in the GBV Response System
The social work profession as defined by the International Federation of Social Workers (IFSW) promotes social change, problem-solving in human relationships, and the empowerment and liberation of people to enhance well-being of individuals, groups, and communities through the pursuit of support human rights and social justice (Nicholas et al. 2010, p. 5). In Namibia, social workers perform a variety of functions in the GBV response system, including prevention education, advocacy, crisis intervention, psychosocial support, victim support, and case management services (Matthews and Hase 2013).
The effects of sexual violence are multi-dimensional and far-reaching. People who experience sexual assault may experience physical impacts such as sexually transmitted infections, as well as emotional impacts, such as depression, anxiety, and post-traumatic stress disorder (Matthews and Hase 2013). As such, these impairments in functioning may be the impetus for first contacts with social workers. This typically occurs through the social workers employed by the Ministry of Gender Equality and Child Welfare and the Ministry of Health and Social Services, or through one of the domestic or international organizations focusing on GBV in the country.
As social workers are often the front line of prevention, treatment, care, and support for victims of sexual violence in many countries, including Namibia, it is imperative to understand their attitudes towards sexual violence and endorsement of subtle rape myths. As service providers to vulnerable individuals, having a positive affirming attitude as well as advocating for the development of social welfare and health systems that work to prevent and effectively treat cases of sexual violence is of paramount importance as is the exploration of how these views evolve throughout student years of study in a social work education program.
It is hypothesised that individuals possessing higher levels of rape myth acceptance would demonstrate less empathy when engaging with clients who report having had this experience. As a value-based profession, the Code of Ethics for Social Workers mandates that professional social workers engage with clients from a non-judgmental perspective. However, it stands to reason that if a professional social worker actively endorses rape myths, they will be unable to fulfil this expectation for professional social work practice. As a result, they may be more likely to engage in victim blaming, or minimizing threats, real or perceived, as well as downplaying the impact of sexual assault. Even if these attitudes were not overtly discussed with clients, this does not mean they will go unnoticed. Service users (clients) may pick up on subtle messages during their interactions with social workers, especially at the initial stage of a professional relationship. It is at this crucial period of the professional relationship that service users are trying to determine whether the helping professional is knowledgeable, genuinely cares, and has a positive view of her (the service user) (Robbins and Cook 2017). If a service user does not perceive that the professional helper meets these criteria, they will be less likely to tentatively trust the worker enough to open to them. Without an open relationship based on acceptance and mutual understanding, it is much less likely that the professional social worker will be successful in helping the client to address the challenges they are facing (Robbins and Cook 2017).

Research Questions
This exploratory cross-sectional inquiry utilized a survey design to explore social work students' acceptance of rape myths. Two research questions guided the inquiry: (1) "To what degree do undergraduate students studying social work in Namibia accept rape myths?" and (2) "Are there significant differences in the levels of rape myth acceptance based on sociodemographic factors, such as students' age, sex, year of study, religiosity, or country of origin, or self-identification as a feminist?".

Procedures
After obtaining approval from the university's ethics review board, questionnaires were administered in a classroom setting at the end of a regular course meeting for each year group. Given the lack of exposure to participating in research projects, the administering investigator explained the informed consent document and study procedures, as well as reviewed the concepts of anonymity and voluntary participation. Participants were provided with copies of the instrument as well as copies of the informed consent document. Consistent with the general ethical guidelines related to social science research, no inducements or incentives were offered to respondents.

Participants
The study sample consisted of 153 respondents who represented full-time students enrolled in all four-year groups in the Bachelor of Social Work programme at a public tertiary institution in Namibia. Females accounted for 81.0% (n = 124) of those reporting gender in the sample, and males represented Soc. Sci. 2018, 7, 152 4 of 11 15.7% (n = 24). The age of respondents ranged from 18-52, with a mean of 25.73. Respondents included citizens of six different African countries who speak 17 different languages as their mother tongue.
Gender: Consistent with enrolment in most social work degree programmes, females represented an overwhelming majority of respondents (81%, n = 124). Males represented the remaining 15.7% (n = 24) of respondents.
Age: Respondents' ages range from 18-52, with the average age being 25.7. The distribution of age was consistent when considering the mode (M = 25) and the median (Md = 24).

Dependent
The primary data collection instrument utilized in the current study was the Updated Illinois Rape Myth Acceptance Scale (IRMA). The original instrument was developed by Payne et al. (1999) and is reported as being one of the most reliable and psychometrically validated instruments to measure acceptance of rape myths (McMahon and Farmer 2011). The IRMA was updated in 2011 and contains 22 items that measure an individual's endorsement of a variety of common myths about rape (McMahon and Farmer 2011). Previous research identified that the 2011 version of the IRMA contains four subscales including: 'she asked for it', 'he didn't mean to', 'it wasn't really rape', and 'she lied'. Sample statements from the updated instrument include: "Women tend to exaggerate how much rape affects them" and "If both people are drunk, it can't be rape". Responses to these 22 statements on the updated IRMA are provided using a Likert-type scale, with responses ranging from 1 to 5, with 1 indicating strong agreement with the statement and 5 indicating strong disagreement. Higher scores indicate greater rejection of rape myths, while lower scores suggest greater acceptance of rape myths.
The IRMA has a history of successful application in a variety of cultural and linguistic contexts, including research conducted in Asia, Europe, Africa, and the US (Chiroro et al. 2004;Xue et al. 2016). Available evidence from almost 20 years of use suggests that the IRMA is a cross-culturally reliable and a valid instrument that can be used to assess respondents' acceptance of rape myths (Chiroro et al. 2002).

Independent
A sociodemographic questionnaire containing nine questions was utilised to collect data concerning respondents' gender, age, year of study, primary language, country of origin, religious affiliation, the frequency of religious participation, and self-identification as a feminist. Self-identification as a feminist was categorized as a nominal variable, with only "yes" and "no" as options.

Data Analysis
Completed questionnaires were assigned a unique identification number and entered into SPSS version 24.0 for analysis. Data analysis procedures included univariate and bivariate analyses (Analysis of Variance (ANOVA) and t-Test) that explored the dimensions of the respondents' rape myth acceptance levels based on the socio-demographic variables included in the demographic questionnaire.

Sample
Year of Study: Students from the first to fourth year of study in the Bachelor of Social Work programme are represented in the total sample (N = 153). Thirty-four first-year students responded to the survey, which represented 23.4% of the total sample. Among the second-year students, 22 (or 15.2% of the total sample) responded to the survey. The third-year students comprised 26.2% of the study sample, given that 38 students completed the survey. Finally, in the fourth year, 51 students (or 35.2%) completed the measurement instrument. Frequency of Religious Participation: A four-point Likert scale was used to assess the level of religious participation, ranging from never to daily. However, for data analysis purposes, data were collapsed into a dichotomous variable with 'daily' or 'weekly' being coded as religious and all other categories were coded as 'not religious'. Analysis indicated that 63.6% (n = 91) of the sample participated in religious services daily or weekly, whereas 36.4% (n = 52) of the sample reported monthly, holiday, or never participating in religious activities.
Country of Origin: Large numbers of international students (n = 4000) are studying in all faculties at the University of Namibia (A. Sam, personal communication, 4 January 2018). In the current study, social work student respondents reported six different countries of origin, with Namibia representing approximately 70.6% (n = 108) of all respondents. Zimbabweans represented 20.3 (n = 31), while 5.2% (n = 14) were from other African nations, including Zambia, Nigeria, Botswana, and the Democratic Republic of the Congo.
Updated Illinois Rape Myth Acceptance Scale (IRMA) scores were tabulated for all participants. On the 22-item measure, four subscale scores and one total score were calculated. Lower scores indicate a greater acceptance of rape myths; subscale and total scores are summarized in Table 1.

Bivariate Analyses
Next, total scores for IRMA were analysed for differences in means with several demographic variables. Table 2 displays participant gender, age, year of study, and country of origin. Though the mean score for female participants was higher than the male participants' mean score, this difference was not statistically significant (p > 0.05; female: M = 86.0, SD = 12.89; male: M = 80.45, SD = 16.98). As a reminder, this indicates that (while not statistically significant), female participants scored lower in rape myth acceptance than the male participants. To further look for important group differences, independent t-tests were run on the IRMA subscales. Only subscale three ("she lied") showed significant differences (p < 0.005) between men and women. Male participants were more likely to believe this rape myth than their female counterparts. On all other subscales, statistically significant differences were not detected. To analyse the relationship between age and IRMA score, a Pearson's correlation was calculated. The results indicated statistically significant differences (r = 0.22, p < 0.05), with a positive correlation between age and IRMA scores. This suggests that older students had higher scores and thus less acceptance of rape myths.
In an attempt to analyse the impact of education on rape myths, a one-way ANOVA was conducted. As indicated in Table 3, there were significant differences between IRMA scores among students in the various years of their academic programs (F = 13.03, df = 123, p < 0.001). Students in the fourth year, for instance, had a mean IRMA score of 87.55, while the first-year students' mean score was 67.33, suggesting that more advanced students were less likely to accept rape myths. To analyse the relationship between age and IRMA score, a Pearson's correlation was calculated. The results indicated statistically significant differences (r = 0.22, p < 0.05), with a positive correlation between age and IRMA scores. This suggests that older students had higher scores and thus less acceptance of rape myths.
In an attempt to analyse the impact of education on rape myths, a one-way ANOVA was conducted. As indicated in Table 3, there were significant differences between IRMA scores among students in the various years of their academic programs (F = 13.03, df = 123, p < 0.001). Students in the fourth year, for instance, had a mean IRMA score of 87.55, while the first-year students' mean score was 67.33, suggesting that more advanced students were less likely to accept rape myths. An investigation of the country of origin among participants, also using a one-way ANOVA, indicated significant differences between groups (F = 4.55, p < 0.05; see Table 4). Participants from Zimbabwe had the highest mean (91.7), while those from other African countries had a much lower mean score (82.4), albeit with a very small sample size. An investigation of the country of origin among participants, also using a one-way ANOVA, indicated significant differences between groups (F = 4.55, p < 0.05; see Table 4). Participants from Zimbabwe had the highest mean (91.7), while those from other African countries had a much lower mean score (82.4), albeit with a very small sample size.  To investigate whether self-identification as a feminist (yes/no) was a significant predictor of IRMA scores, a t-test for independent samples was conducted (see Table 5). Participants who selfidentified as feminists held significantly less accepting views of rape myths than those who did not label themselves as such (p < 0.05; M (feminist) = 88.0, SD = 11.71; M (non-feminist) = 81.68, SD = 15.47). To investigate whether self-identification as a feminist (yes/no) was a significant predictor of IRMA scores, a t-test for independent samples was conducted (see Table 5). Participants who self-identified as feminists held significantly less accepting views of rape myths than those who did not label themselves as such (p < 0.05; M (feminist) = 88.0, SD = 11.71; M (non-feminist) = 81.68, SD = 15.47).
A final bivariate analysis involved the relationship between religiosity and rape myth acceptance. Religiosity was defined as practising religion daily or weekly (yes) or less often/never (no). T-test results are in Table 6. Interestingly, participants who indicated more religious practices also demonstrated lower acceptance of rape myths, though the difference was not statistically significant (p > 0.05). To investigate whether self-identification as a feminist (yes/no) was a significant predictor of IRMA scores, a t-test for independent samples was conducted (see Table 5). Participants who selfidentified as feminists held significantly less accepting views of rape myths than those who did not label themselves as such (p < 0.05; M (feminist) = 88.0, SD = 11.71; M (non-feminist) = 81.68, SD = 15.47). A final bivariate analysis involved the relationship between religiosity and rape myth acceptance. Religiosity was defined as practising religion daily or weekly (yes) or less often/never (no). T-test results are in Table 6. Interestingly, participants who indicated more religious practices also demonstrated lower acceptance of rape myths, though the difference was not statistically significant (p > 0.05).

Discussion
Though the study is exploratory in nature, the findings have important implications for social work education in Namibia. To begin, students who had progressed further in their social work programme were significantly less likely to endorse subtle rape myths, with significant differences in the "year of study" analysis. In fact, there was a 20-point difference between first-year and fourth-year students in the overall IRMA score. This positive change between year one and year four provides strong support for education about rape as a helpful method to improve attitudes and knowledge among African social work students.
As in many other studies, the participants' demographic characteristics and life experiences played an important, and in some cases, statistically significant, role in their attitudes toward rape and rape myths. In particular, a somewhat surprising finding related to religiosity. Those who practised religion daily or weekly were less likely to believe rape myths than those who reported attending church less regularly. Though the difference was not statistically significant, this result differs from other studies finding more religious participants to hold less favorable attitudes about rape and rape myths (Freymeyer 1997;Barnett et al. 2018). Van Klinken (2013 points to the need to critically analyse and understand the role of religion in many African societies, and warns of the need to carefully interpret the politics of human rights in Africa; rape myth acceptance can broadly be categorized in this context. Further, the lack of a statistically significant difference between male and female participants related to the acceptance of rape myths may also be cause for cautious optimism. Previous studies (Lonsway and Fitzgerald 1994) have found gender to be the strongest and most consistent predictor of rape myth acceptance, with men being more likely than women in nearly every study to believe rape myths. While female participants in the current study were indeed less likely to believe rape myths, this difference was not significantly different. Given that they had opted to elect social work as an educational programme, the male participants in this study may have demonstrated more favorable attitudes than most male students. Further, the majority of male respondents were in their third or fourth years of study, thus suggesting that social work education has a positive impact on changing attitudes towards sexual assault and the acceptance of rape myths. Finally, educational and psycho-educational interventions have been shown to help reduce rape myths in other contexts (Gilbert et al. 1991) and that may have been the case here.
Similar to participants in other studies (Krause et al. 2017;Boakye 2009;Rebeiz and Harb 2010), participants in the current study who labelled themselves as feminists were less likely to accept rape myths. In Namibia, the holding of feminist beliefs is not a widely-discussed topic, except perhaps among university students, especially those studying social work.
More specifically related to the unique African contexts being studied, the participant's country of origin was also related to differential attitudes. Participants from Zimbabwe had the highest mean IRMA score, demonstrating less acceptance of rape myths than their Namibian counterparts. Those from other African countries (such as Nigeria, Zambia, and the Democratic Republic of Congo) had a much lower mean score, albeit with a very small sample size. Differences among Zimbabwe, Namibia, and other African countries may merit further study. Only one published study to date (Matthews et al. 2017) has examined these beliefs among specific sub-groups of African participants, which calls attention to the need for further research with larger groups from each nation.

Implications for Social Work Education
As stated, a positive trend in the study relates to the participants who have had further years of social work education, which is directly connected to education strategies and classroom activities promoting human rights and advocating for social justice issues, including gender equality. Because more favorable attitudes were reported among those who were more advanced in their studies, it seems that intervention strategies and future research may consider incorporating education about gender equality with a focus on rape and rape myths among future social workers, early in their educational journey.
Whether models of anti-rape education in American classrooms may be fitting in the African context has yet to be investigated. Findings from previous studies of undergraduate students (Fischer 1986) suggest that teaching strategies can impact attitudes toward rape, but questions remain as to the effectiveness of these models in the African context. Future studies aiming to investigate this, particularly with longitudinal data collection, are warranted.

Limitations
The results show a wide range of rape myth acceptance among social work students studying in Namibia. While the analyses reveal some of the correlates, it is not a comprehensive list of all possible factors that may influence participants' attitudes towards rape. Additionally, religious attendance was addressed in terms of frequency of attendance, but the investigators were not able to compare different religions or denominations.
Additionally, the sample size was relatively small and consisted entirely of students who voluntarily chose to enrol in a social work education programme. All attitudes were self-reported and participants were exposed to a curriculum, suggesting that social work ethics promote social justice and human rights. Because the findings related to 'year in school' showed significant positive movement in this domain, further study is merited.
It is important to note the challenges of conducting research such as this in this region. Rape and sexual assault are largely considered a private matter that is not openly discussed in Namibian (and other African) cultures. While the participants who labelled themselves as feminists demonstrated lower levels of rape myth acceptance, it is important to note that this is a relatively rare term in general Namibian society, particularly among older populations. As such, the social and political context of Namibia is likely a factor for the use or avoidance of such terminology (Thomas 2007;Currier 2012). At any rate, each of these factors may have influenced the results of this study.

Directions for Future Research
The current research project has explored social work students' attitudes towards rape myths at one public university in Namibia. Data analysis revealed similarities with previous research, as well as unique factors, such as similar levels of negative attitudes towards rape and rape victims, as well as significantly different levels of negative attitudes when considering participants' country of origin. Researchers looking to extend this line of inquiry in the future may consider a cohort study to track students' attitudes over the course of degree programmes. This suggested study design has the potential to contribute valuable knowledge regarding the ability of the social work curriculum to shape professional social workers' attitudes and practice behaviors in this area.