The Impact of Sexual Harassment Awareness Training on Attitudes and Norms in India
Sexual harassment can have harmful effects on a woman’s psychological, economic, and social life. The researcher evaluated the impact of a sexual harassment training on the incidence of sexual harassment, opposite sex relationships, and attitudes and beliefs towards sexual harassment in India. Overall, the trainings reduced sexual harassment perpetrated by men, particularly extreme forms in the short run and mostly intermediate forms of harassment in the long run. However, these trainings also reduced opposite sex romantic relationships between classmates both in short and long run. Researcher shows this happens because the training changed men’s beliefs about others attitudes to sexual harassment in particular.
Policy issue
Sexual harassment is prevalent worldwide and can have harmful effects on a woman’s psychological, economic, and social life. Policymakers and academics have advocated for sexual harassment awareness trainings in universities and workplaces as a method of deterring harassment. However, little is known about the effect of these trainings on the incidence of sexual harassment, changes in attitudes towards sexual harassment, and relationships between men and women. Measuring the impact of sexual harassment awareness training is important not only to help deter sexual harassment but also to understand how it affects other behaviors or relationships in environments where social interactions are crucial for performance like schools and workplaces.
Sexual harassment awareness trainings have the potential to affect not only an individual’s own attitude toward sexual harassment, but also their perception of peers’ attitudes which can affect sexual harassment and also relationship between peers. Theoretically, women may try to determine if men are “good” or “bad” partner types when making relationship decisions. Men who intrinsically like to harass women are more likely to be identified as a “bad” type and those that do not like to harass as “good” types. If men change their intrinsic attitude toward sexual harassment to personally disapprove of this act, then harassment may decrease, and women may choose to form more relationships with men. If men perceive greater social disapproval toward harassment among their peers due to the training, then sexual harassment may decrease because men do not want to be seen poorly by their peers. However, in this case, women may be more cautious to form relationships with men because they are unsure if men are genuinely “good” types that disapprove of sexual harassment or are “bad” types only acting like a “good” type to avoid social disapproval. Together, it is unclear if trainings will reduce harassment and if romantic relationships between peers will be impacted.
Context of the evaluation
While there has been a recent proliferation of laws prohibiting sexual harassment around the world, these types of crimes continue to persist throughout India. In Delhi, 80 percent of women reported experiencing sexual harassment over their lifetime. Young women are particularly at risk of sexual violence, making college campuses a key focus of many NGOs in this field.
Safecity is an NGO that specializes in sexual harassment trainings and has been active in urban and rural communities in India since 2013. For this project, the researcher collaborated with Safecity, legal experts, and three colleges in the University of Delhi to develop and provide students with a modified sexual harassment awareness training. Almost half of the women participating in the project reported experiencing low or intermediate forms of sexual harassment and 16 percent reported experiencing extreme forms of sexual harassment in the two months preceding the trainings from men inside and outside the colleges. Results from a survey conducted before the training indicated that men and women from one of the colleges largely underestimated the prevalence of sexual harassment.
Details of the intervention
The researcher used a randomized evaluation to measure the impact of a sexual harassment awareness training conducted with college students in Delhi on the incidence of sexual harassment, relationships between men and women, and attitudes and beliefs towards sexual harassment. Safecity implemented the training between September and October 2019 with men targeted in two colleges (male intervention) and women targeted in one college (female intervention):
- Male intervention: Across two colleges and 93 classes, the researcher randomly assigned all men in half of the classes to receive the training. The men in the remaining classes did not participate in any type of training. However, women in all 93 classes were separately provided with basic information on sexual harassment to avoid under-reporting due to gaps in women’s understanding about what constitutes as sexual harassment although ex-post it did not turn out to be crucial.
- Female intervention: In a third college, the researcher randomly assigned 69 classes to either a high or low intensity group. In the high intensity group, 75% of the women in each class were randomly assigned to receive the training, while in the low intensity group only 25% of women were randomly assigned to receive the training. The female intervention group received an abbreviated form of the training described below.
The sexual harassment awareness training consisted of two main components. First, trained facilitators taught the legal definition of sexual harassment and ways to identify it through discussions, skits, and situation-based exercises in male intervention groups (in female intervention groups this was done using a survey only). Second, participants worked through perspective-taking, empathy-building, and bystander intervention exercises, including discussing anonymous narratives on personal sexual harassment accounts and “courting” behaviors that are legally sexual harassment. This component was only delivered in the male intervention group as earlier research revealed that women grasped the concepts quickly and empathy-building was less relevant. These trainings took a total of 3 to 5 hours and were provided over two sessions over one academic semester.
The researcher conducted follow-up surveys approximately three months after the end of the training. The researcher surveyed women about sexual harassment from men in their class and surveyed students about their romantic partnerships and friends, perceptions of sexual harassment, and intrinsic beliefs using direct and indirect questioning techniques. The researcher also used a competitive game where same or mixed-gender pairs had the opportunity to switch partners to understand whether changes in opposite sex relationships were driven by women or men. The researcher combined measures from the survey and competition to create an overall opposite sex relationship index.
In addition to receiving ethical review and approvals from an institutional review board, researchers made efforts to address and account for ethical questions by conducting surveys privately, reading women an informed consent with the option to refrain from answering and never asking women to identify perpetrators. They also provided women with the Safecity and National Commission for Women (NCW) helpline, as well as contact information of a research personnel if they wanted their data deleted at any point.
Results and policy lessons
Looking at the outcomes for all women whether they were in the baseline survey or not, the researcher finds that sexual harassment awareness trainings reduced sexual harassment perpetrated by men, particularly extreme forms of harassment, but only when the training was provided to men. The trainings also reduced romantic relationships between men and women classmates.
Sexual harassment incidence: Trainings delivered to men (male intervention) reduced the incidence of sexual harassment perpetrated by men by 0.06 standard deviations, as reported by their female classmates. The training was most effective in reducing extreme forms of sexual harassment (such as nonconsensual groping, pinching and other acts) compared to mild or intermediate forms, decreasing extreme harassment by 1.05 percentage points (105 percent). This translated into 51 fewer women out of 1200 facing extreme forms of sexual assault in an academic year. Trainings delivered to women (female intervention) did not have a detectable effect on the incidence of sexual harassment. Overall sexual harassment remains lower by 0.08 standard deviations in male intervention classes approximately six months after the treatment albeit the results are weaker for long run because classes were shut down due to covid which could have reduced effectiveness of the training.
Opposite sex relationships: Training only men (male intervention) had no effect on friendships between men and women but reduced romantic relationships by 1.3 percentage points (64 percent). During the competition led by the researcher, women with male classmates that received the training were 14 percentage points (37 percent) more likely to prefer cooperating with other women as compared to women with male classmates that did not receive the training. Further, opposite sex relationships as measured by an index of women’s responses decreased by 0.13 standard deviations, while there was no effect on relationships based on men’s responses. These effects persist one to two years after the training showing that segregation over the long run increases although women are not more likely to be single. The female intervention reduced romantic relationships with men in the same class by 4 percentage points (126 percent).
Sexual harassment awareness: Men who received the training were 0.09 standard deviations more aware of sexual harassment behaviors than those that did not receive training. Awareness about legal mechanisms and ambiguous sexual harassment situations increased by 15 and 4.5 percentage points respectively (107 and 12 percent). Awareness increased largely for intermediate and mild forms of sexual harassment rather than extreme forms. This indicates that a greater ability to recognize what behaviors constitute harassment was not the driver of the reduction in extreme sexual harassment incidents. Women’s awareness did not improve, likely because all women received some information on sexual harassment before the program began. This is true for whether women were covered in baseline or not.
Perceptions and attitudes: The male intervention had no effect on men’s perception of legal costs but increased the perception of social costs by 0.05 standard deviations. The researcher did not detect any effect on men’s victim blaming attitudes or support for sexual harassment NGOs, indicating that men’s intrinsic attitudes toward sexual harassment did not change at least in the short run. This suggests that sexual harassment was deterred by raising the perceived level of social disapproval more strongly than changing intrinsic attitudes. The female intervention had no effect on men’s perception or attitude to sexual harassment.
Combined, these results indicate that sexual harassment awareness trainings can increase awareness and reduce sexual harassment. Delivering these trainings to men in a group setting may more effectively shift men’s perception of their peers’ attitudes toward sexual harassment and induce them to change their behavior. The long run results on sexual harassment and opposite sex relationships indicate that even with a small dosage of the training, repeated sessions would be helpful in reinforcing ideas of the training more deeply. The trainings present a scalable method to combating sexual harassment in universities and other settings like workplaces.
Encouraged by the results, colleges that collaborated in the study are working to introduce sexual harassment awareness training more regularly and increased awareness about their legal complaints committee. In practice they have struggled to do this because of the pandemic but are committed to doing so after the colleges have adapted their class delivery scheduled post pandemic.
Sharma, Karmini. Tackling Sexual Harassment: Short and Long Run Experimental Evidence from India. Working Paper, December 2022.