College Women’s Perceptions of A Friend-Based Intervention to Prevent Alcohol-Involved Sexual Assault

Abstract

Sexual assault (SA), particularly alcohol-involved SA, remains prevalent among college women. Because SA often begins in social contexts, bystander intervention has become a popular approach to prevention. Bystander interventions train individuals to intervene on behalf of others, including strangers, despite research indicating that intervention is more likely to occur when the bystander has a relationship with the target. Shifting the focus to friends as potential bystanders capitalizes on the qualities of relationship and responsibility that facilitate intervention. College women (N = 35) participated in focus groups (N = 8) during which they viewed a video prototype of a friend-based motivational interviewing (FMI) intervention session conducted with a friend dyad and provided feedback about the relevance and feasibility of using such an approach to reduce SA among friends who drink together in social settings. Content analysis of focus group transcripts yielded three themes: (a) Friends as Natural Bystanders, (b) The Role of Alcohol in Intervention, and (c) Receptivity to FMI intervention. Women indicated that they feel responsible for keeping their friends safe and that this sense of responsibility facilitates helping behaviors. Women also described ways through which alcohol intoxication can affect helping behavior. Women expressed enthusiasm for the FMI intervention approach and identified its emphasis on friendship and flexible approaches to personal safety as strengths. Findings highlight the promise of FMI intervention approaches that capitalize on the strengths of women’s friendship to create safety goals that align with participants’ values and overcome barriers to intervention, including alcohol intoxication.

In spite of increased recognition of the problem, rates of sexual assault (SA) remain consistently high on college campuses, with college women being disproportionately targeted for assault compared with college men (Anderson et al., 2018). Approximately 20–25% of college women experience SA, including forced sexual contact, rape, or attempted rape, over the course of their college experience (Muehlenhard et al., 2017), and alcohol consumption is a factor in the majority of these assaults (Abbey, 2002; Mohler-Kuo et al., 2004; Testa & Livingston, 2018). These rates are roughly the same as when the problem was first identified by Koss and colleagues in the 1980s (Koss et al., 1987; Koss & Oros, 1982), despite numerous prevention efforts focused on reducing perpetration (DeGue et al., 2014). This indicates the ongoing need for effective programming that aims to increase women’s safety through a variety of avenues that target both prevention and risk reduction.

College SA often begins in the context of parties or social events (Abbey, 2002) where others could potentially intervene. Because of this, bystander intervention has emerged as a promising approach to prevent college SA (Banyard et al., 2004, 2007). Training in bystander intervention is based on the bystander intervention model proposed by Latane and Darley (1970). According to this model, intervention requires that a bystander: (a) notice the event; (b) interpret the event as one that requires action; (c) make the decision/take responsibility to act; (d) know how to intervene; and (e) implement intervention. Bystander intervention training approaches for preventing college SA (e.g., Bringing in the Bystander, Banyard et al., 2004; Mentors in Violence Prevention, Katz, et al., 2011) are grounded in this framework and aim to increase protective efforts at each step of this model. For example, education about the prevalence of SA and its associated risk factors can enable college students to be more attuned to situations where a potential for SA exists and to recognize when assistance is needed (Steps 1 and 2). Bystander intervention training also strives to cultivate prosocial attitudes and to promote helping behavior to foster a sense of responsibility for intervening (Step 3). To facilitate response to a potential SA threat, participants are taught skills for safely intervening (Steps 4 and 5). In sum, the bystander intervention training is designed to promote bystanders’ abilities to recognize and respond to potential threats. Bystander intervention approaches have been effective in increasing participants’ knowledge of SA, reducing rape myth acceptance, and increasing attitudes and efficacy toward intervening and intentions to intervene (Banyard et al., 2007; Coker et al., 2011; Katz & Moore, 2013; Kleinsasser et al., 2015). Yet, research also shows that bystanders often hesitate to intervene due to a variety of concerns (Blayney et al., 2021; DeMaria et al., 2018). Overcoming barriers to intervention is critical to successful implementation.

Despite the promising effects of bystander intervention training on college students’ attitudes and intentions, the impact of these interventions on actual bystander behavior has been modest (Katz & Moore, 2013). Traditional bystander intervention programs train individuals to intervene on behalf of a stranger, yet ambiguities about sexual intent inherent in college drinking settings and perceptions of victim worthiness may pose significant barriers to intervention (Pugh et al., 2016). Additionally, college men and women differ in terms of their willingness to intervene in SA situations, with college women reporting more positive attitudes toward bystander intervention and greater intentions to intervene in a SA situation than college men (Hackman et al., 2022; Hoxmeier et al., 2020). Shifting the focus of bystander intervention training to specifically target female friends may offer a promising alternative to traditional approaches. Female friends are often present in social contexts where college drinking occurs, and individuals are more likely to intervene on behalf of a friend or someone they know as compared with a stranger (e.g., Bennett & Banyard, 2016; Burn, 2009; Katz et al., 2015). Barriers that may inhibit a man or a stranger from intervening are less likely to occur with female friends. For example, female friends are better positioned to notice when an assault may be unfolding and are less likely than strangers or men to hold negative judgments about the target of sexual aggression, especially if she is intoxicated or is dressed or has acted in a way that could be deemed as “provocative” (Burn, 2009; Pugh et al., 2016). In addition, social contexts involving alcohol are often imbued with expectations for casual sexual encounters (i.e., “hookups”; Lindgren et al., 2009; Paul et al., 2000), potentially making it challenging for an outside observer to determine whether a sexual interaction is consensual (Pugh et al., 2016), whereas a female friend may be better equipped to make this judgment (Blayney et al., 2021). Given the expectation for sex to occur in drinking settings, college men, especially those who are heavy drinkers, may be less likely to notice and interpret the situation as one requiring intervention or may fear backlash from other men in the situation (Carlson, 2008; Hoxmeier et al., 2020; Orchowski et al., 2016; Schipani-McLaughlin et al., 2020).

College women are generally aware that they are more likely than men to be targeted for SA, and as such, they are invested in keeping one another safe from SA (Blayney et al., 2018). College women also often naturally engage in protective behaviors in drinking situations with female friends. These protective behaviors may involve intervening to prevent unwanted sexual interactions, but they are also more broadly aimed at reducing risk for SA in general. Qualitative research has been used to identify naturally occurring strategies that female friends may use to ensure one another’s safety, such as sending non-verbal signals to one another to interrupt unwanted situations, imposing agreed-upon rules about staying and leaving together at parties, and having one person drink less or remain sober to watch out for others’ well-being (Armstrong et al., 2014; Blayney et al., 2021). Although this research suggests that women are well-situated and naturally inclined to protect one another from SA, women in both Armstrong et al.’s and Blayney et al.’s studies also described circumstances that pose significant challenges to intervention and use of protective strategies, even among friends, suggesting more can be done to strengthen and enhance friend protective behaviors. Many of these challenges centered around issues related to alcohol consumption and social concerns.

Although female friends are predisposed to protect one another and naturally engage in some assault protective behaviors, alcohol intoxication poses one of the greatest impediments to keeping oneself and one’s friend safe from SA (Armstrong et al., 2014; Leone et al., 2018; Pugh et al., 2016). Cross-sectional work with young adult victims of recent SA has shown that when bystanders are present, bystanders themselves are likely to be using alcohol and are more likely to miss opportunities to intervene when they were highly intoxicated (Haikalis et al., 2018). According to alcohol myopia theory (Steele & Joseph, 1990), alcohol narrows a drinker’s attention, causing them to become preoccupied with their own situation and interests. Thus, alcohol consumption has the potential to distort or delay perceptions of and responses to risk (Leone et al., 2018). This is especially true in situations where indicators of risk are subtle or ambiguous, such as in social settings where there may be expectations for consensual sexual encounters (Blayney et al., 2021; Haikalis et al., 2018; Leone et al., 2018). Alcohol can also exacerbate the diffusion of responsibility that occurs in group settings, impair effective strategizing of plans, and may focus attention onto other factors (e.g., norms surrounding interfering with potential sexual encounters, consequences of intervening) that could reduce the likelihood of choosing to act (Blayney et al., 2021; Leone et al., 2018).

Social concerns pose another barrier to intervening on a friend’s behalf (Blayney et al., 2021). Sexual hookups are normative in college drinking settings (Lindgren et al., 2009, Paul et al., 2000); thus, bystanders are tasked with determining whether an observed sexual interaction is consensual. Female friends are better prepared to make this distinction than are casual acquaintances or strangers, but even friends are concerned about misinterpreting a situation and interfering with a consensual encounter (Blayney et al., 2021). Women fear embarrassment and incurring the anger of others. Women may also abandon safety measures when they interfere with their own social objectives, such as hooking up themselves (Armstrong et al., 2014).

In sum, the extant literature suggests that female friends are in a unique position to protect one another from college SA, in part because their relationship and intrinsic sense of responsibility for their friend leave them well-positioned to recognize and respond to potential threats. Although many women naturally engage in behaviors to protect themselves and their friends from SA, bystander intervention may not occur for a variety of reasons, such as alcohol impairment, concerns about personal safety, social consequences of intervening, or a lack of knowledge skills or confidence to respond to a potentially risky situation. Taken together, these findings suggest that college women may benefit from prevention programming to enhance their use of protective behaviors with their female friends in a reliable, consistent manner, especially in social contexts that involve alcohol.

Bystander intervention training alone, even when presented to friends, is unlikely to be sufficient for reducing SA risk. Intervention skills deficits, motivation, perceived barriers to intervening, and lack of awareness about the potential benefits of intervening all may pose a challenge to traditional bystander training focused on friends. Alcohol impairment also can compromise bystander action. A more focused and personalized approach is needed. Incorporating motivational interviewing (MI) into the friend-based bystander intervention training may help to address these challenges. MI is a clinical approach designed to increase motivation to change risky or harmful behaviors by encouraging exploration and consideration of change in a style that is described as the “spirit of MI” (Miller & Rollnick, 2012). The foundation of this MI spirit is its “collaborative conversation style for strengthening commitment to change” (Miller & Rollnick, 2012, p. 12). The techniques of MI are designed to support autonomy, agency, and self-efficacy for change, by approaching the discussion of risk behaviors and providing feedback and advice in a supportive and non-confrontational manner. Ultimately, MI is used to increase readiness to make behavior change and to decrease barriers to such change. MI’s focus on respect for and empowerment of participants makes it particularly well-suited to increase women’s awareness of personal risk and to help them to commit to protective strategies that balance their chosen (as opposed to prescribed) safety and social goals.

For years, MI was conducted only in a traditional one-on-one (interventionist to client) format. However, increasingly research has shown that MI lends itself well to a dyadic format, with two people who know one another (e.g., peers, co-workers, friends, romantic partners), as it is within this kind of relationship that people can most successfully work together toward behavior change (Bourke et al., 2016; D’Amico et al., 2017; Miller & Rose, 2010; Monti et al., 2014). Importantly, a peer-based MI can use the responsibility and relationship of friends as a framework to foster collaborative efforts to increase readiness and decrease barriers to helping behavior. Thus, for a variety of reasons, a peer-based MI intervention that capitalizes on and amplifies the natural sense of relationship and responsibility that friends have for one another would seem an excellent fit for college-based SA intervention. Yet, whether women would be receptive to such an intervention and find value in participating with a friend is unknown.

The Current Study

The role of friends in protection against college SA has been largely overlooked in the bystander intervention literature. To our knowledge, there have been no efforts to formally target friends in bystander intervention training. Because women are more receptive to bystander intervention training and report greater willingness to intervene compared to men, targeting female friends as part of bystander intervention training using an MI approach may help to overcome some of the barriers to helping behaviors and to increase the likelihood of intervention. Such an intervention would also need to address the role of alcohol in recognizing and responding to risk but should do so in a way that allows women to balance personal, social, and safety goals. To this end, we have developed a friend-based motivational interviewing (FMI) intervention designed to reduce college women’s risk of alcohol-involved SA (Read et al., 2022).

In the current study, we were interested in learning how such a friend-based intervention would resonate with college women. We also wanted to gain insight into college women’s perceptions of the extent to which their own or their female friend’s alcohol use may impact their ability to implement protective behaviors. Using data collected from focus groups conducted with college women, we sought to answer the following research questions to assess women’s receptivity to FMI intervention in general: (a) What are college women’s perceptions of the role of friendship in the prevention of SA? (b) How, if at all, does alcohol intoxication affect women’s abilities to keep themselves and their friends safe from SA? and (c) What do college women see as the strengths and weaknesses of an FMI approach? Given the exploratory and developmental nature of the study, there are no hypotheses.

Method

Author Positionality

All of the authors are cisgender White women residing in the Northeastern United States. The first author has a background in educational psychology and over 20 years of experience studying the role of alcohol in the SA of young women and adolescents. She has extensive experience using mixed methods and qualitative interviews to understand perceptions of SA risk and protective factors among young women, adolescents, and parents of young children. The second, third, and fourth authors are doctoral students in clinical psychology within 10 years of college graduation. Each have spent several years studying SA and aspects of alcohol use, as well as providing women with clinical services and education on the topic of SA. The fifth author is a clinical psychologist whose program of research spans over 20 years and focuses on the treatment of trauma and alcohol-involved SA. Her work has been largely quantitative with some experience with analysis of focus group data. Through their years of studying SA empirically and listening to women’s concerns in a clinical capacity, the authors have come to understand SA as a complex social problem with cultural roots in patriarchy and misogyny. Although perpetrators are solely responsible for their acts of sexual aggression, the authors believe that it is imperative to empower women with knowledge and skills to preserve their safety and well-being.

The FMI Intervention: Protecting Allies in Risky Situations

FMI Intervention: Protecting Allies in Risky Situations (PAIRS) is an FMI intervention that incorporates the principles of bystander intervention and was developed to be delivered to dyads of female college friends who regularly use alcohol and attend parties together at least twice a month. The PAIRS intervention is grounded in MI methodology, which utilizes a collaborative conversation style for strengthening commitment to change and to motivate and prepare women to work together to reduce SA risk while out together in drinking contexts. The intervention is delivered by an interventionist who has been trained in MI. The dyad of female friends meets with the interventionist for a single session about 1 ½ hr in length. Following typical MI format, the intervention includes a discussion of the friend dyad’s social context and alcohol use, with exploration and consideration of personal risk situations (e.g., “As you know, this is a study about how women might help to keep one another safe when going out. To help us start to think about this with you, I’d like you to describe a typical night when the two of you are out together. This way I can understand in more detail what happens”; “How do you know if a friend is at risk? What are the signs you look for?”). During the session, the MI-trained interventionist offers information about SA risk, the cognitive and behavioral effects of alcohol, and bystander intervention, as well as a menu of options regarding protective strategies (e.g., keeping track of each other’s whereabouts, leaving party together). Readiness to engage in shared efforts at protection is assessed throughout the intervention and used to formulate a protection plan:

We’ve talked a lot about risk for unwanted sexual experiences and ways women may protect themselves. Now I’d like to spend a bit more time hearing your thoughts about making any changes to what you currently do to intervene to help keep each other safe from sexual assault.

The protection plan involves identifying two protective behavior strategies that the dyad collaboratively decides that they want to implement together. Plans to enact the strategies, potential barriers to enacting, and suggestions for overcoming these barriers are also discussed.

Consistent with MI theory (Miller & Rollnick, 2012), the interventionist seeks to establish a positive rapport with participants, assess areas of risk, build motivation to engage in protective behaviors, facilitate change discussion, and ultimately help the participants create a sustainable plan that includes personalized protective behavioral strategies. The bystander model is introduced as part of the intervention, and the interventionist also elicits explicit discussion about the ways in which alcohol may impact helping efforts:

Looking at these steps [of the bystander model] and based on what you’ve just read, what role do you think that alcohol plays? In what ways do you think drinking might make any or all of these steps harder?

Consistent with the “spirit” of MI, interventionists provide information in a way that is intended to be collaborative. The PAIRS program is unique in its dyadic structure that capitalizes on the strength of friendships and targets ways that the friend dyad may support, encourage, and share responsibility with one another in protecting against SA.

We conducted focus group interviews to obtain feedback about the FMI intervention and to determine its feasibility and relevance to college women. We video recorded a mock PAIRS intervention session with two female undergraduate students who were friends and who regularly went out together. The mock participants were not given a script to follow and were encouraged to respond as though they were actually participating in the intervention. College women who met eligibility criteria proposed for the intervention were invited to participate in a focus group discussion during which they viewed the video depicting the mock intervention session and discussed their reactions to the intervention approach.

Participants

Undergraduate college women ages 18 to 24 (M = 19.68, SD = 0.58) were recruited from Western New York to participate in a focus group discussion. Paper flyers and geographically-targeted advertisements on Facebook solicited college women to provide their opinions and feedback about an intervention “designed for friends to help protect one another against sexual assault.” A total of 331 women responded via phone or email expressing interest in the study, and 193 were screened over the phone for eligibility. The remaining 138 were unable to be reached for screening. Fifty-one women met the eligibility criteria, which included being enrolled in a 4-year college or university, having consumed four or more drinks of alcohol on a single occasion at least two times per month in the last three months, and attending parties or social gatherings where there is drinking at least twice per month. Of these, 35 women participated in one of eight focus groups. A majority of participants (69%) self-identified as non-Hispanic White, 17% identified as Asian, 6% as Black, and 6% as multiracial.

Procedure

The eight focus groups occurred during April and May of 2019. Each group was facilitated by two female moderators. After welcoming participants, the moderators reviewed the purpose of the study and participants signed consent documents. Focus groups (N = 8) ranged in size from three to six participants (Mode = 5). Each focus group lasted approximately 2.5 hr. Participants were compensated $50 in cash. All study procedures were approved by the University Institutional Review Board.

During the focus groups, the moderator played clips from the pre-recorded mock intervention session and paused the video recording at specific timepoints to inquire about participants’ thoughts and reactions to different aspects of the PAIRS intervention with respect to the role of friendship in protection against SA, perceptions of sexual risk, bystander intervention, and how alcohol use may influence bystander behaviors. For example, at one point during the video, the interventionist presented the mock participants with information about the effects of alcohol on behavior and cognition, and they discussed the relevance of this information for the mock participants’ lives. Following this depiction, the focus group moderator paused the video and asked, “Alcohol can be a factor that affects whether people intervene to help their friends. How well do you think this part of the intervention addressed this particular challenge in the discussion of protecting friends?” The moderators also queried participants about whether the conversation they were viewing appeared natural and realistic and whether they believed that college women would find the intervention helpful and be willing to participate (e.g., “In what ways do you think that this interview creates an environment where women may be open to making some of the changes that were discussed in this segment? Did these changes sound feasible? Realistic?”).

Focus groups were transcribed by a project research assistant (RA) and checked for accuracy by a second RA. Transcripts of focus group discussions were analyzed using directed content analysis (Hsieh & Shannon, 2005). This qualitative approach is used when the research questions are informed by prior research or theory (Hsieh & Shannon, 2005). In this case, we were seeking to understand participants’ perceptions of phenomena that had been identified as being influential to an individual’s likelihood of intervention (i.e., relationship and alcohol use). As is traditional in the analysis of focus group data (Krueger & Casey, 2015), the group and not the individual was the unit of analysis. Transcripts were read independently by four members of the research team to get a general sense of thematic content. After this, consistent with direct content analysis, transcripts were re-read by those four team members and coded for women’s perceptions (positive and negative) of the feasibility of recruiting friends in the protection against SA risk, the role of alcohol in SA risk and prevention, and overall receptivity to the FMI intervention approach. Coding and theme generation occurred through an iterative process through which the four coders read, compared, and refined the codes and created conceptual categories and hierarchies. Themes were organized around the original research questions. Disagreements were resolved through discussion. To establish trustworthiness of the final generated themes, the fifth author who was not involved in the initial coding reviewed and applied the themes to a sample of randomly selected focus group transcripts with high agreement.

Results

Based on study objectives and guided by our questioning, three major themes emerged from the qualitative analysis: (a) Friends as Natural Bystanders; (b) The Role of Alcohol in Intervention; and (c) Receptivity to FMI Intervention. These themes map on to the study objectives of: (a) identifying women’s perceptions of the role of friendship in the prevention of SA; (b) understanding how alcohol intoxication may affect women’s abilities to keep themselves and their friends safe from SA; and (c) assessing women’s reactions to the FMI intervention approach. We describe each of these themes below.

Friends as Natural Bystanders

In response to questions about the role of friends in SA prevention, participants in all eight focus groups offered descriptions of how friends assume responsibility for each other’s safety while in a social context. They described how they naturally feel a sense of responsibility toward one another, utilize protective behaviors together, and how having a close relationship facilitates the steps of the bystander model.

Relationship and Responsibility

Participants in all groups described how feeling responsible for one another is an inherent aspect of their friendship. Friends care about keeping each other safe in risky contexts, and this is at the forefront of their minds when they go out together. The willingness to intervene on behalf of a friend is illustrated in this quote by a woman who intervened to stop a man from harassing her friend: “I’d rather be called a bitch than get followed by a creep all night. I’m way okay with that. And I’m more than glad to be that bitch for my friends. My loyalty is strong.”

Moreover, this sense of relationship and responsibility was viewed by many in our sample as a unique component of female relationships, as participants noted that college women trust other women to be aware and feel a responsibility to intervene in unsafe situations:

It’s really engrained as girls that we need to look out for ourselves, and we need to look out for each other, and guys aren’t held to that because they’re the perpetrators.. . . I’ve been in creepy situations and all my girlfriends rush to me. Guys- [they have] no idea.

Importantly, focus group participants reported that this sense of relationship and responsibility to their friends led them naturalistically to engage in a number of protective behaviors with one another. For example, women noted that part of their responsibility when going out with friends is to ensure everyone stays together and leaves the party with their friend(s), so that no one is left alone with a potential perpetrator after drinking. As one participant described:

I’m always like scanning the room making sure I can see my friends, knowing where they are, and if I can’t find them, the first thing I’ll do is like check my phone, see where they are, and if we’re still in the same house I’ll go find them.

However, they acknowledged that this natural instinct alone may not be sufficient.

I think a lot of girls do at least some of this process naturally type of thing; they just need more of the honing of skills and information because people already do it without this intervention, just not at the best optimum level.

Friendship and the Bystander Model

The steps of the bystander model can be summed into two main phases: Recognizing (Steps 1 and 2) and responding (Steps 3–5) to risk. The focus group discussion suggests that friends are well-suited for the first phase. As noted earlier, women often actively work to stay together at a party, so they are more likely to notice an event where assistance is needed. Importantly, friends know each other well, and that knowledge facilitates the interpretation of ambiguous social situations as potentially threatening. Reports from multiple women in the focus groups supported this. “I mean since I know them very well sober and also when they’re out, their behaviors, like noticing. . .little things like having trouble walking, or just acting off for them.” All of this information helps to increase a friend’s confidence to intervene, whereas a stranger may be uncertain whether there is a problem that requires assistance:

It’s definitely easier to decide if that’s your friend. I know if my friend is over there talking to some guy and she had some kind of body language. . .since you are their friend you know their body language and you know how they act, so if they’re acting suspiciously then you would have more of an inclination to go in and help rather than with a stranger.

The second phase of the bystander model (Steps 3–5) focuses on taking responsibility and acting to intervene. This phase includes deciding to intervene, determining what action is needed, and then implementing the intervention effort. The women in our focus groups noted that they would feel equipped to swiftly decide on a course of action in situations involving their friend, because there is less concern that their friend will be upset by their intervention efforts. Furthermore, the sense of personal responsibility toward their friend also would compel them quickly to action:

If it were you and a stranger that you don’t know, you don’t know how your actions would affect them because you kind of stop and go, “Am I actually helping them or am I not? Is this what they want?” So, I think with a stranger you’re actually interpreting it and trying to decide how you want to approach the situation.

Helping women to embrace and build upon the responsibility they already feel for their friend’s safety offers an important opportunity for intervention efforts. A collaborative approach that builds on these natural strengths of friendship, but that is perhaps more purposeful, planful, and informed by current research, can help to enhance the actions these women are already engaging in to keep one another safe.

The Role of Alcohol in Friend-Based Intervention

A second theme to emerge from the focus group discussions had to do with the role that alcohol plays in women’s efforts to protect their friends from SA in social situations. Participants provided their experiences and insight on the ways in which alcohol impacts their and their friends’ nights out. These narratives fell into a few subthemes: (a) the need for guardianship in alcohol contexts, (b) the ways in which alcohol interferes with the ability of friends to provide guardianship, and (c) the ways in which alcohol may promote intervention (i.e., liquid courage).

Need for Guardianship in Alcohol Contexts

Participants in all groups described how being intoxicated creates a need for guardianship due to cognitive and physical impairment that can render women vulnerable to assault. The potential for danger was especially apparent in cases of extreme intoxication. For example, “that guy came to her house, she wasn’t aware that he was all over her;” “I was at a party. . .and this guy had his arm around me, and I was so drunk that I didn’t even realize that he did that.” Participants talked about providing guardianship to their friends as well as describing situations in which they needed guardianship themselves. They indicated that friends routinely do this for each other, but they do not always plan or communicate about it in advance; thus, the potential exists to miss situations where assistance may be needed:

I went out with my roommate a few weeks ago. . .and then all of a sudden, she was just trashed and. . .she got barred at a club and she was like passing in and out in the Uber and I was like, thank God I wasn’t wasted because this would have been a mess of a scene if one of us wasn’t paying attention.

Alcohol Interferes with Intervention

Participants in seven out of eight groups also reported that they viewed alcohol intoxication as a significant barrier to successful intervention on behalf of a friend. For example, “if you were really drunk, you wouldn’t have been able to intervene as effectively. You know, it would have been more of a problem.. . . anything could happen when you’re really intoxicated.” Data yielded from focus groups revealed a myriad of ways that alcohol might interfere with friend-based interventions, from making it difficult to notice or interpret a threatening situation to interfering with one’s ability to assume responsibility for acting or to act as described throughout the section below:

One girl in my sorority was making out with a guy in the corner. . .but she was really intoxicated, and I’m like “are you okay” and I’m pulling her aside, but I didn’t even trust myself enough because I was drunk, and she was telling me that she’s fine. But I was like I don’t think you’re fine. I think if I was sober, I’d know you weren’t fine, so yeah, alcohol just makes things weird when you’re checking up on people.

Participants frequently described having a friend who is sober or less drunk as a way of guarding against danger due to alcohol intoxication. (e.g., “I mean having someone who’s ‘with it,’ can’t go wrong with that.”) However, one participant pointed out how sober companions may not be a given anymore. “You used to. . .have to have a D.D. [designated driver] when you went out before; we had like Ubers and other ridesharing. So there always would be guaranteed someone sober. But now that there isn’t.”

Participants also mentioned how some of the tactics they use to keep each other safe are less effective when they are intoxicated. For example, being too drunk to remember to check their phones, to keep in contact, or physically staying with their friend throughout the night would make it impossible to notice and determine whether assistance is needed:

We have designated times (when we’re out drinking) where we will meet up like every hour or so just because we know, like we won’t look at our phones, like we can’t hear; we are too drunk to look at them.

Alcohol 100% inhibits our ability to help each other because I’ll be like, ‘dude I was so drunk I didn’t know where you were!’ I was really just trying to find them, like it definitely inhibits it when everyone is at like 10 drinks, like how I am I supposed to keep track of you if I’m trying not to vomit?

Alcohol also impacts women’s ability to be kept safe by their friends. This can include something as simple as keeping a friend in eyesight, as one participant noted: “I have friends who will get too drunk and like wander off . . .and like that’s not good.” In some cases, being drunk makes friends less likely to stick to their safety plans, such as going home together:

I’ve personally had experiences where like a group of girls all went out together, and then like toward the end of the night like some girls wanted to leave and some girls wanted to stay. . . and talk to guys but they were way too drunk to be there. So that’s kind of like a tricky situation where like, do I leave my friends here to talk to these random guys or do I like get them mad at me and force them to come home with me?

Liquid Courage

There also were participants who believed that the effects of intoxication might offer some benefit with regards to keeping others safe. This was discussed in five of the eight groups. Some women believed that this protective effect may occur due to increased sociability, stating, “Especially if you’re drunk, I feel like you’re very open to making friends and helping each other out.” As discussed earlier, women are more likely to intervene with friends because they feel responsibility toward them, and it is plausible that this increased camaraderie with acquaintances and strangers makes intoxicated women more likely to intervene with women who they may not know as well. Other participants believed they are less timid or afraid of offending men when they are drunk:

When I get drunk, I get a little bit more confrontational. . . I’m not at all shy even when it’s strangers and I think that they’re in a bad situation; I’m gonna go up to them and be like “hey, are you okay?”

If a guy is grinding on my friend or something like that, and she looks uncomfortable, if I’m drunk or something. . . I go over there, and I’ll start dancing on her trying to pull her away. I’m also being drunk but I’m embarrassing myself, so it gets the attention away.

In sum, there was a consensus that alcohol intoxication made these social situations less safe for women and their friends, requiring some level of guardianship and women noted how being intoxicated interferes with their ability to watch out for their friends or to accept a friend’s assistance. Alternatively, some women believed that there are situations where the disinhibiting effects of alcohol increases the likelihood of taking protective action.

Receptivity to FMI Intervention

We were interested in hearing women’s reactions to the FMI intervention, including both general reactions to an intervention of this type and format, as well as specific thoughts about aspects of the PAIRS intervention prototype presented in the video. Women identified many advantages to the FMI approach and commented on its relevance for their lives and feasibility for implementation. They also offered some critiques and highlighted some concerns.

Advantages of an FMI Intervention Approach

The focus group participants identified four features of the FMI program that they particularly appreciated: (a) the focus on friendship, (b) the customizability and empowering nature of the intervention, (c) the opportunity to enhance protective behaviors, and (d) increasing communication and shared commitment to safety.

Friendship

Women appreciated how the program capitalized on and enhanced the relationship that exists between friends: “They are building on what they already have as a friendship.” Several of the women commented on the significance of this aspect of the intervention:

I think it was so important that you brought it back to their friendship because that’s been like a common theme throughout the interview. And saying like “oh, this would strengthen your friendship,” like how will this change a relationship.

Women in all of the groups also reported that completing the program with a friend would allow participants to feel at ease: “Yeah, the fact that they’re friends and doing it together makes it more conversational and comfortable. I would be more comfortable if I was with my friend compared to being alone.”

Customizable/Empowering

Another positive aspect of the program according to participants was the way it incorporates women’s personal experiences and goals in the creation of a safety plan alongside the more general information provided. This aspect of the program was discussed in all eight groups. Focus group participants appreciated the flexibility of the plans: “I also feel like you presented a lot of good options that weren’t just like, ‘Oh, stop drinking so much.’ You gave a lot of other things that are helpful in those situations.” Women voiced that being able to decide which steps they were comfortable taking, rather than being prescribed solutions, was empowering and a particular strength of the MI approach:

I also think that basically every point, especially towards the end, was things that they came up with on their own so it wasn’t them being told “this is what you should do.” There is like the list but kind of letting them choose what is important to them and in their life, so I think it’s more relevant to them like things that they would actually do as opposed to things they’re just being told to do.

Reinforces and Enhances Protective Behaviors

Women in seven of the eight focus groups noted that the program helped to reinforce and enhance things that they already do, both alone and with friends, to reduce their risk for negative outcomes. Many of the women noted that these protective behavior tactics were not always explicitly discussed in their group of friends: “At first I thought this was stuff that I’m never going to do but when I read it [the list of protective behaviors], I was like I probably do 80% of all this but I don’t actively do it; it’s kind of more passive.” In addition to increasing consistency of previously utilized techniques, the intervention provided ideas for new ways to decrease risk. Participants were optimistic that the changes they would make would not be much more difficult than what they were currently doing and could improve their level of safety:

It’s already making me evaluate it, just hearing it, and I want to get my friends together and say like, “Come on guys let’s go talk to someone” because sometimes you have these unspoken agreements and you kind of have this checklist filled out mentally, but it’s good to talk about it. It can make you realize like, “Hey, we could be doing this better, we could change this. Maybe we could make ourselves safer.”

Communication/Accountability

Women in all of the focus groups appreciated how the program required friends to work together, support, and communicate with each other. Women indicated that they thought the intervention would encourage ongoing communication between friends:

I think the fact that they already talked about what they did like that was kind of the premise of what to change or make better. Makes it more involved with each other. They’ll probably talk about this conversation that they had with you or the interview when they go out the next time.

Participants noted that creating a plan together increases accountability to one another: “And because you’re doing it with a friend you both are on the same page on the same strategies, and I feel like you guys can hold each other to it a little more.” Women indicated that planning with their friends before going out would make them more likely to follow through with their plans and prepare for barriers that would further increase the likelihood of success. Finally, women acknowledged that creating a plan after discussing the different components of the intervention (e.g., risks, protective behaviors, the role of alcohol) would allow participants to select and adapt techniques that incorporated what they had learned from the intervention itself:

It’s a very open conversation and you educated them [mock participants] on a lot of different things and sometimes it’s easier to make a plan when you have all of the information you need right there and you’re having them make a plan right there and that way . . . if they don’t talk about it ever again after this, they still have what they remember.

Criticisms of FMI Intervention Approach

Although the majority of the feedback about the intervention was positive, participants did identify some concerns about the FMI intervention approach. These concerns fell into three categories: (a) change is not needed; (b) change is unrealistic or overwhelming; and (c) expectation of change is unfair.

Change is Not Needed

In three of the focus groups, there were women who felt that it was unnecessary to change any of their behaviors. A few of these women believed that they were able to accurately judge the safety of their social context. As one woman noted, “I don’t think that way [about protective behaviors] because I just assume that in the environment that we are at all the time we are going to be safe.” Another woman stated:

I think obviously it depends on where you are or who you’re with; I think that has a big impact on everything. I feel like I’m very safe with my location, like I don’t really go places where I’ll feel uncomfortable.

Additionally, a few women indicated that they would not be interested in participating in an intervention like this because they were already familiar with the information being presented:

I don’t know if I would sit through one of these just because all of these things I’ve kind of already heard of. Even though I should be more aware of it and be more prepared, but I kind of already know everything.

Change is Unrealistic/Overwhelming

Women in six groups expressed concerns that making changes would be unlikely for participants. Some felt that individuals would not follow through on their plans: “A lot of people be like ‘I’ll be fine’ and they won’t really do it. I’m not saying they won’t, but I can see where it could be something like, it’s kind of forgotten.” A few felt that the MI format of the intervention would not fit with their friendships and lifestyle: “I feel like definitely this is kind of an unrealistic conversation, like me and my friends definitely wouldn’t talk about the pros and cons [of changing], we just do things to make sure everyone’s good and that’s about it.” When discussing the bystander model, some women felt that it was oversimplified on paper and that intentions to intervene are not enough:

It’s easy to say you’ll do something, and when you’re in the situation most of the time you don’t do it. So, looking at the (bystander) model, I could see how that could help someone but like when you’re looking at the situation you don’t know what to do.

A couple of participants felt that women would be overwhelmed by the prospect of making a change plan during the intervention and would prefer to make a plan on their own time:

I feel like I usually make a lot of plans and I’m not able to implement it. So, I feel like in the moment . . . I would be really nervous to actually write down and set my mind to it. I would rather wait and process what I learned and just make those decisions on my own time and gradually make additional changes instead of implementing a bunch of changes.

Expectation of Change is Unfair

Women in three groups voiced frustration that the focus of the FMI intervention, like other SA interventions, puts the onus on women to change their behavior. Importantly, they felt that the expectation that they needed to remain vigilant in order to stand guard over their friend’s safety was unfair:

It’s kind of frustrating, like I shouldn’t have to make sure my friends are okay; I should just trust that people aren’t going to suck. I’ll ask what they do but I want to go out and have a good night. I don’t want to have to worry, but I will.

Another woman remarked, “There was a topic brought up [in the video] about how it’s the female’s job to change stuff that she’s doing to protect herself against someone else. At what point is that fair?”

Although drinking reduction is not a goal of the program, the information provided during the session, along with the participants’ experiences often point to the benefits of reducing drinking. A few women felt that it was unfair to suggest that women decrease their drinking as a means of reducing SA risk: “Yeah, like I’m comfortable with myself, like I feel like I can handle things properly with my level of drinking, so why change . . . for another person’s actions like they [sexually aggressive men] should be the ones changing.”

Discussion

Overall, women in this study were largely positive toward the idea of an FMI intervention for addressing the problem of college alcohol-involved SA. There was a general consensus that such an intervention makes sense, given that women already are motivated to help keep their friends safe, and that they naturally watch out for one another and rely on their friends to help them avoid unwanted sexual situations. This finding replicates prior research (Bennet & Banyard, 2016; Blayney et al., 2018; Burn, 2009; Katz et al., 2015) suggesting that friends are a fruitful intervention target. Consistent with models of bystander intervention, participants described how their friendships facilitated their ability to recognize and accept responsibility to respond to potential risk. Nonetheless, several women noted that they do not always use protective behaviors consistently or may not make explicit safety plans with their friends, suggesting that they could benefit from intervention. These women viewed the FMI intervention as providing an opportunity to reflect on and strengthen their relationship through enhanced communication, planning, and shared commitment to one another’s safety. They also viewed the FMI intervention as having the potential to reinforce and promote consistent use of existing protective behaviors as well as to expand their repertoire of protective behaviors. Importantly, the FMI approach made college women feel empowered to work together to develop a tailored plan that would enable them to balance personal safety and social goals.

Although capitalizing on the strength of the relationship could enhance the likelihood of intervention, women also spoke candidly about how alcohol may interfere with effective intervention among friends. Consistent with alcohol myopia theory (Steele & Josephs, 1990), women talked about how, when intoxicated, they get pre-occupied with their own interests and may lose track of their friends or forget to follow safety plans (e.g., checking in with one another). This may result in failure to notice and/or interpret whether or not their friend may need assistance (i.e., the first two steps needed for intervention; Latane & Darley, 1970). They also indicated that even if they did notice that their friend needed assistance, they might be too intoxicated themselves to take protective action. In addition to interfering with one’s ability to recognize and respond to a friend’s risk, alcohol intoxication also can interfere with one’s willingness to accept help from a friend.

Though most of the discussions focused on how alcohol interferes with bystander intervention, a few women reported the perception that they were more—rather than less—likely to intervene when they had been drinking. These women asserted that the disinhibiting effects of alcohol gave them courage to face a potential aggressor and described situations in which they directly confronted a potential perpetrator. Though it has been posited that alcohol may facilitate intervention in specific social contexts (Leone et al., 2018), there is also concern that by acting in a confrontational way, women may be placing themselves and their friends as risk for harm if the potential perpetrator or others in the situation respond aggressively (Van Bommel et al., 2016).

The qualitative findings related to alcohol consumption replicate themes identified in other studies regarding how alcohol consumption may complicate bystander intervention (Armstrong et al., 2014; Blayney et al., 2021; Leone et al, 2018), by either interfering with the processes of recognizing and responding to risk or through eliciting “liquid courage” reactions, potentially increasing likelihood of aggression (Leone et al., 2018). Heavy alcohol consumption can pose a significant impediment to effective bystander intervention that needs to be explicitly addressed in intervention efforts. Although the FMI intervention described here does not explicitly target drinking reduction, using MI to explore the effects of heavy alcohol use may lead some participants to consider drinking reduction as part of their safety plan. Other possibilities may include taking steps to make protection cues more salient in an attempt to overcome alcohol myopia, such as setting a phone reminder to check on one another at regular intervals.

Despite the largely positive reaction to the FMI intervention prototype, a few important concerns were raised, consideration of which can be helpful in refining the current FMI and other friend-based interventions. First, a few women felt that they were already using protective behaviors and did not perceive a need to change anything that they were currently doing. As part of the FMI prototype, participants are presented with suggestions for enhancing safety as well as being given the option to develop their own strategies. Consistent with the spirit of MI (Miller & Rollnick, 2012), participants are free to adopt or reject strategies as they choose, so there is no pressure to set change goals that are not perceived as relevant or desirable. FMI interventions may be able to address this by guiding participants to think through how their current protective behaviors are most likely to be used in risky contexts. Such an exercise may reveal discrepancies between what women think they are doing and what they are actually doing.

A few participants also expressed frustration that they perceived the intervention as putting the onus on women to engage in protective efforts, rather than targeting men’s assaultive behavior. We appreciate this concern and validated the participants who raised it within the focus group discussions. At the outset of the intervention, the interventionist explicitly states that the blame for sexual aggression lies solely with the perpetrator, and women are never to blame for their victimization. However, in a world where there are motivated offenders willing to exploit opportunities, we also believe that empowering women to manage their social goals in ways that also maximize their safety is critical to women’s health and well-being. The framework of MI, which is focused on allowing personal choice and flexibility in determining whether and how behavior change should occur, is helpful in this regard. Data from our focus groups support that an FMI intervention can be empowering for women in this way.

Limitations and Future Directions

There are several limitations to acknowledge. First, it is important to note that women were responding to a video prototype of the PAIRS FMI intervention depicting two other women going through the intervention protocol and not to having gone through the intervention themselves. Second, though participants responded positively to the FMI intervention, this is not evidence of intervention efficacy. A trial of this intervention is ongoing. Finally, the interventionist, mock participants, and most of the focus group participants were White; women from other racial and ethnic groups may have different experiences or views that impact their perceptions of the FMI intervention and its relevance to their lives (Krebs et al., 2011). However, a strength of using an FMI format is that it would be easily customizable to meet the needs of diverse groups of women.

The next steps in the development of the FMI intervention include refining the intervention protocol based on participant feedback and testing its effectiveness using a randomized controlled trial with a large, diverse sample of college women. Once efficacy is established, efforts can be made to identify potential mechanisms of change. Furthermore, the current format of one interventionist working with a dyad of friends for an in-person intervention is labor and time-intensive. Future efforts should explore options to increase scalability, such as administration through a web-based format. Given the flexibility of the FMI approach, the intervention could also be modified for use with other populations such as with the military or non-college samples.

The subject of involving male friends as bystanders came up in a few of the groups. Although this may be a reasonable approach, it poses some significant challenges. First, SA is often perpetrated by men known to the victim, including male friends (Abbey, 2002), raising the concern that women may be victimized by the men they are trusting to protect them. Also, men tend to be less responsive to bystander intervention than women (Hackman et al., 2022; Hoxmeier et al., 2020), especially if they know the perpetrator. Gender-specific tailored approaches to bystander intervention that target men can be effective at reducing perpetration and increasing bystander behavior (Salazar et al., 2014); however, future research should examine whether and how FMI intervention can be adapted for use with mixed-gender dyads.

Conclusion

Findings from this qualitative study indicate that a FMI intervention was well-received by college women, who saw the intervention as a natural fit for the risk contexts that they encounter and for the way in which they naturalistically rely on friends in social settings. These findings, though preliminary, suggest promise for this novel intervention approach that capitalizes on the natural resource of women’s friendships to decrease risk for SA—a pervasive public health problem affecting a substantial portion of young college women in the US If the efficacy of this intervention is supported in future work, this FMI approach offers a new direction for bystander-based intervention that may help expand the focus of SA prevention efforts to those who are naturally most likely to help but who could benefit from learning to implement specific tools and protective behavioral strategies.