Introduction

Gender-based violence (GBV) is a concept that encompasses a continuum of violence and abusive behaviors on the basis of sex and gender, intersecting with other dimensions such as age, ethnicity, disability, and sexuality (Hearn et al., 2022; Strid et al., 2021). These facets of violence are found in a variety of contexts, including the university environment, where unequal power relations and organizational culture are specific risk factors (O’Connor et al., 2021). In recent years, many societal conditions have revitalized interest in these issues, beginning in the United States with the Obama administration, the #Timeisup campaign, and the increase in Title IX investigations initiated by the Department of Education (Eriksen et al., 2022). With the subsequent global explosion of the #MeToo movement and the acceleration of activism against gender-based violence, sexual harassment (SH) in higher education has received particular attention, to the extent that many higher education institutions and universities worldwide, particularly in Europe, are becoming sensitive to GBV and developing best practices to address the issue (Jones et al., 2021; Strid et al., 2021; USV React, 2019; Zavos, 2018).

SH is described as any unwelcome sexual advance, any unwelcome request for a sexual favor, any verbal or physical conduct or gesture of a sexual nature, or any other conduct of a sexual nature that may be expected or perceived to be offensive, humiliating or intimidating to the individual (UN, 2008).

The systematic review carried out by Bondestam and Lundqvist (2018, 2020) on SH in higher education shows a prevalence that varies widely from 2 to 93% (Fnais et al., 2014), and depends on a variety of different factors. These authors analyzed publications up to 2017, therefore an update of the literature review is needed to understand whether and what development there has been in recent years.

In the last ten years, many campus-based SVSH (sexual violence and sexual harassment) interventions have been realized in university. The spread interest on SVSH reflects the need to intervene on these two phenomena as interrelated problems, as they often occur together, in the same contexts, like the academic one, characterized by close relationships complicated by power hierarchies (Bloom et al., 2021).

As for prevention, interventions generally cut across SV and SH, and bystander intervention training (e.g., The Green Dot evidence-based program, Coker et al., 2011, 2014; Cook-Craig et al., 2014) are the most widely employed initiatives which prompt statistically significant changes across behavioral, cognitive, attitudinal dimensions (Degue et al., 2014; Fenton et al., 2016); they also contributed to reduced victimization and perpetration at the community campus level (Coker et al., 2016). This kind of programs should be embedded in a broader set of multilevel interventions as recommended by the Center for Disease Control and Prevention, which incorporated the WHO’s guidance (Dahlberg & Krug, 2002) and the Socioecological Model (SEM; individual, interpersonal, community, and societal) as a framework for SVSH interventions (Bloom et al., 2021; Bronfenbrenner, 1977, 2005). Within this perspective, recent recommendations on SVSH prevention may be summarized as follows: 1. conceptualize violence as a community issue where all members have a role to play in prevention: the focus is not only on addressing individual behavior, but on making the larger environment a protective factor against violence and harassment, in which the survivor support is contemplated as a crucial factor (McMahon et al., 2021); 2. introduce evidence-based programs and training, such as bystander interventions (Banyard, 2013; Banyard & Potter, 2017; Orchowski et al., 2010, 2020); 3. build strong networks both inside and outside the university (e.g., Beres et al., 2019; Maxwell et al., 2010; Morse & Allensworth, 2015; Sisneros & Rivera, 2018); 4. evaluate the effects of interventions to prevent and contrast SH in the short and long term, using longitudinal cohorts and longer follow-up periods (Anderson & Whiston, 2005; Bondestam & Lundqvist, 2018, 2020; McDonald, 2012).

Furthermore, Bondestam and Lundqvist (2018, 2020) suggested that future research should aim to: 1. develop cross-cultural research projects that include non-English speaking parts of the world, comparative studies of different countries, and the development of cross-cultural research projects (Henning et al., 2017; McDonald, 2012; Willness et al., 2007); 2. promote the use of standardized definitions and research tools to identify and measure the different types of violence (Fedina et al., 2018); 3. adopt an intersectional perspective that incorporates the experiences of minority groups (Anderson & Whiston, 2005; Fedina et al., 2018; Guizzo & Alldred, 2024; USV React, 2019; Voth Schrag, 2017; Webermann & Murphy, 2022); 4. identify the underlying processes of SH to adopt a preventive perspective and develop functional interventions and strategies (Anderson & Whiston, 2005; Voth Schrag, 2017).

A systematic update of the literature review on SH in academia is useful to see whether after 2017 research has evolved in line with these recommendations and those on prevention.

Objectives

The main objective of this review is to direct updated research findings towards an effective prevention perspective. Based on the considerations made in previous literature reviews (Bondestam & Lundqvist, 2018, 2020) and the recently cited suggestions to guide future work, this review looks at studies conducted since 2018 and has three specific goals: 1. to describe the phenomenon of SVSH in academia and the characteristics of recent research on the topic; 2. to situate research findings within a prevention-focused framework, with particular attention on risk factors; 3. to gather recent experiences and recommendations for building a complex system of multilevel effective preventive interventions.

Materials and Methods

Information Sources and Search Method

Following Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines (Liberati et al., 2009), a systematic review of the international harassment in academia literature was performed. To identify relevant articles, a rigorous search strategy was used on two databases, Web of Science and Scopus. After selecting some keywords for each main aspect of the search—focus, target, method -, the following search strings were used: “university” AND (“harassment” OR “misconduct” OR “violence” OR “abuse”, i.e. focus) AND (“faculty” OR “academics” OR “students” OR “perpetrators” OR “survivors” OR “bystanders” OR “staff” OR “employees”, i.e. target) AND (“prevention” OR “intervention” OR “prosecution” OR “policy” OR “guidelines” OR “best practices” OR “survey” OR “reporting”, i.e. method). Between January and March 2023, abstracts of the identified articles were screened for eligibility, and the remaining articles were assessed for eligibility based on a full-text reading.

Eligibility Criteria

The main inclusion criterion was related to the language: only articles written in English were considered. The only exclusion criterion was the year of publication, and the chosen reference point was a comprehensive review of the existing literature on the same topic (Bondestam & Lundqvist, 2018, 2020), which considered studies conducted up to 2017. Hence, all articles published before 2018 were excluded from this review.

For those studies which met the inclusion criteria full texts were retrieved and reviewers (O.B. and D.G.) independently assessed the eligibility of these texts and then grouped them for syntheses. When discrepancies or doubts emerged, these were resolved by a discussion with supervisory third-party reviewers (G.Z. and L.D.).

After a screening of the abstracts, these studies were grouped into two categories, according to their main topic: 1. SH studies and risk factors; 2. Prevention. Topic 1 includes a description of studies conducted since 2018 to date aimed at measuring and analyzing the phenomenon, with particular attention to associated victimization risk factors. Topic 2 includes the most recent research on SVSH prevention. The two issues are closely related because the development of effective preventive interventions cannot disregard knowledge of the phenomenon, as well as the precise identification of the risk factors to be countered.

Some subcategories were identified. With respect to Topic 1, following Bondestam and Lundqvist (2018, 2020), the most relevant subcategory was identified with Target (1. students, 2. faculty and staff, and 3. all targets), then, we considered Country (1. U.S. studies and 2. non-U.S. studies), Extent of the study context (1. single-college studies and 2. multi-college studies), Use of the same SH scale (1. adopted; 2. non adopted), Intersectionality (1. intersectional studies and 2. studies without an explicitly intersectional research focus). Furthermore, Individual, Relational and Community risk factors were collected.

With respect to Topic 2, articles were divided according to the previous category of U.S. and non-U.S. studies, and additionally divided into others related to the following main themes addressed in the studies analyzed: 1. multilevel and whole campus approach (studies on interventions which promote an integration of the individual, relational, community and social levels)Footnote 1; 2. bystander intervention programs (studies focused on bystanders); 3. Networks (studies on the importance of building networks, expanding interlocutors, and collaboration); 4. prevention and support processes (studies on prevention and support as a coordinated process). Special attention was given to the monitoring and evaluation reports that span the subcategories.

Results

Study Selection

Of the original 948 records found in the two databases, 440 records were screened according to the inclusion/exclusion criteria. All duplicates were removed. Of these, 136 articles were assessed for eligibility. After reviewing the full content of the articles, 68 records were excluded: 10 were written in languages other than English, despite the abstract being in English; 40 did not include SH, but only other facets of violence; 11 described the mere perception of the phenomenon by the target; 7 regarded different kinds of intervention that did not match the subcategories.

Figure 1 shows a schematic representation of the selection process.

Fig. 1
figure 1

Records selection process according to PRISMA guidelines

Full size image

Study Characteristics

Of the 68 articles included in the review, 31 were considered to refer to the topic of SH studies and risk factors and 38 to Prevention. Within Topic 1, 11 studies investigate students (sub-group 1.1. in Table 1), 8 faculty and staff (sub-group 1.2 in Table 1) and 8 all targets (sub-group 1.3 in Table 1); 6 are U.S. and 11 non-U.S. studies; 14 involve several universities and no one involve different countries; 7 studies have an intersectional research focus.

Table 1 Records within Topic 1- SH studies and risk factors
Full size table

As for Topic 2, 11 of the 38 articles are from non-U.S. contexts; 11 concern Multilevel and whole campus approach (sub-group 2.1. in Online Resource 1); 8 Bystander intervention programs (sub-group 2.2. in Online Resource 1); 11 Networks (sub-group 2.3. in Online Resource 1); 8 Prevention and support processes (sub-group 2.4. in Online Resource 1). One paper has been considered for both topics.

The SH section includes only empirical studies, while the prevention section includes commentary based on interventions realized in the academic setting, which produces recommendations and best practices.

SH Studies

Although born in the U.S. context, research on SVSH is gradually spreading to various parts of the world, with adaptation of scales and questionnaires or creation of ad hoc surveys aimed to capture cultural specificities. Considering the selected studies, research investigating the student population are now conducted also in other countries around the world (10 out of 15 studies are non-U.S.), while studies that focus on academics and staff or that consider the entire academic population are still in half or more of the cases dominated by U.S. research.

Of particular note are adaptations of standardized research tools in new contexts such as Japan (Takeuchi et al., 2018), Jordan (Spencer et al., 2022) or Canada (Dion et al., 2022). As for the need to use standardized definitions and scales to measure the different types of SVSH (Bondestam & Lundqvist, 2018, 2020; Fedina et al., 2018), the limitation of inhomogeneity among all included studies persists. There are many reasons for this: 1. adaptation of definitions and tools to the specific legislative context (Bondestam & Lundqvist, 2018, 2020; Pilgaard et al., 2022); 2. research objective of measuring forms of SVSH related to specific cultural or professional contexts which needs ad hoc tools (e.g., Gomez, 2022.) or 3. of detecting specific forms of violence for which there were no previous codified scales (e.g., Vargas et al., 2021).

The existing tools and definitions for detecting SH are numerous (Spencer et al., 2022). Among others, the SEQ (Fitzgerald et al., 1988, 1995) is a predominant scale within the SH literature, even in low- and middle-income countries (Ranganathan et al., 2021), defined by three dimensions as follows: sexual coercion,Footnote 2 gender harassment,Footnote 3 and unwanted sexual attentionFootnote 4 (Fitzgerald et al., 1988, 1995, 1997). These dimensions form the basis of many SH definitions, which are also used in some of the research included in this review: 8 quantitative researches use the SEQ, often in a version adapted to a specific non-U.S. context.

In terms of intersectionality, only 7 of the 31 studies reviewed focused specifically on the perspective of minority groups, suggesting that research in this area is not very well developed, although its importance is increasingly recognized both in academia and at the policy level (Musso et al., 2020). The lack of studies with an intersectional approach can also be seen in the few research studies that focus specifically on ethnicity. These studies, which are included in this review, mainly refer to the American context, while elsewhere attention to this specific factor still seems to be in its infancy.

Risk Factors in SHSV for Preventive Interventions

Many of the studies aim to identify several risk factors related to SVSH victimization that, if properly systematized, could offer useful indications for implementing effective preventive interventions (Dahlberg & Krug, 2002). Analysis of the collected research allows us to summarize what emerges in relation to the first three levels of the socio-ecological model (see Table 1).

Individual-Level Factors

Numerous studies included in the review showed that being female strongly affects the likelihood of being a victim of SVSH, a characteristic that applies to both female students and academics and staff. In general, what exposes personnel most to SH and/or SV are previous vulnerabilities related to macro-dynamics of power, social inequalities and cultural marginalization. Among both students, academics and staff, lesbian, gay, bisexual, transgender, queer, intersex, or asexual (LGBTQ +) and black, multiracial or indigenous persons are more vulnerable to SH and/or SV. Greater vulnerability is also found among female students who are younger, from urban rather than rural backgrounds, with a disability, economic vulnerability and with propensity to use alcohol.

Relationship or Interpersonal Level Factors

Students who come from socially disadvantaged families are more likely to be victimized, as are students who have already experienced bullying or other forms of SV. In relation to sexual violence, one factor mentioned in connection with the peer group is the prevalence of alcohol or drug use. As far as staff and academics are concerned, the presence of other forms of discrimination tends to be associated with SH. According to most studies, hierarchical and asymmetrical relationships also provide fertile ground for harassment, especially when there is little support from colleagues.

Community-Level Factors

On a macro level, the academic context seems to foster a culture that could lead to the normalization of sexual harassment. It turns out that students often try to ignore or minimize the harassment even though they find it confusing, disturbing and embarrassing. In addition, specific academic social power structures, such as fraternities or the inherent dynamics of the medical training process are cited as contextual risk factors for SH and predatory behavior. Among the faculties considered, medical schools appear to be particularly at risk. Among students and academics, graduate students are more at risk of being harassed by staff than undergraduate students. This is related to the inherently unequal distribution of power that creates an environment in which some individuals are more at risk of gender-based violence. In the specific case of graduate students, dependence on faculty for financial support and other forms of access to networks, necessary for career development, can increase the risk of abuse. General risk factors for academics include lower levels of job insecurity, lower job satisfaction, low employee incentives, and an organizational culture that tolerates harmful behaviors or controlling actions.

Prevention

11 of the 43 articles selected were from non-U.S. contexts; this is a lower percentage than indicated in the previous section. This finding suggests that prevention efforts may have difficulty spreading beyond U.S. borders.

A Multilevel Whole Campus Approach

To prevent gender-based violence in the university, campuses require integrated approaches that engage all members of the community and span across individual, relational, community and societal level simultaneously (Beres et al., 2019; McMahon et al., 2021; Robinson et al., 2020).

First, it is suggested the creation of a group with a collaborative approach; it is the case of University of Otago (Beres et al., 2019), in New Zealand, where the group was composed of academic researchers and staff who worked on issues of sexual violence and the design of the initiative was preceded by consultation with over 100 staff and students for a full day.

Second, it is recommended that programs and training be selected based on demonstrated behavior change or, at least, attitude change (Beres et al., 2019).

The introduction of new, not previously validated training may be ineffective or have undesirable effects (Htun et al., 2022). It is also recommended to focus on more intensive interventions and bottom-up approaches for the target group that promote critical thinking, problem-solving techniques and opportunities for open discussion (Carey et al., 2022).

In the selected studies, in addition to the evidence-based programs, there are numerous initiatives that have not undergone rigorous evaluation; projects that have only been shown to be useful in improving participants’ knowledge and, in some cases, initiatives that have not produced any results. Given the need to integrate multiple interventions on a continuum of quality levels, projects that improve knowledge and realign cultures can reinforce the messages of evidence-based behavioral programs.

The projects, programs, and recommendations identified can focus on the individual, relationship, community, and societal levels or cross between multiple categories.

With regard to the individual level of preventive interventions, Webermann and Murphy (2022) emphasized the need to promote individual communication, emotion regulation and empathy skills that can be applied to a wide range of relationships. In addition, the authors emphasized the need to improve models for identifying students at high risk for misbehavior and use of force and to develop selective prevention strategies that target high-risk individuals, pointing to prevention strategies that aim to identify SVSH in the early stages. None of the identified articles reported specific experiences with projects aimed at these objectives.

The RealConsent program was introduced at the interpersonal level. The program has been shown to increase knowledge of aspects of sexual consent and prosocial behaviors and improve knowledge and skills for safely intervening in potentially problematic situations; it also impacts aspects of the community level by challenging social norms and reducing victim blaming (Beres et al., 2019).

Banner et al. (2022) mentioned a mandatory 20-min online training module on the culture of respect aimed at staff and faculty. The final evaluation of the training confirmed the need to integrate multiple interventions and showed some skepticism about the ability of education and training programs alone to address the problem, which was also found in other research (Carey et al., 2022). Other less than encouraging results were collected in the evaluation of an integrative tool that illustrates 13 dynamics of interpersonal coercion identified by college students and utilizes the socioecological model (Munro-Kramer et al., 2022): no significant or sustained differences in knowledge, attitudes, or self-efficacy were found between those who interacted with the tool (intervention) and those who did not.

Like several previously presented training and projects, bystander programs can be transversal and situated between the individual, interpersonal, and contextual levels (Banyard, 2011); these programs focus on recognizing early signs of sexually harmful behavior and developing skills to intervene. Within this framework, bystander programs play an important role in raising community awareness and creating “guardians” who can prevent certain forms of SH and/or offer support to victims. The research conducted confirms that after participating in bystander intervention training, faculty staff also felt that they could play a crucial role in preventing violence by modeling prosocial behavior, seeking to be perceived as allies by students, and challenging cultural norms related to SV (Robinson et al., 2020).

When considering the community level, it is important to focus on the characteristics of the universities, as noted by Casey & Hampton (2022), who analyzed qualitative data from staff and students to reflect on how adaptable the socio-environmental prevention model is for commuter campuses and emphasized that any intervention must be tailored to the specific context.

Another strategy recommended to act at the municipal level is the introduction of SVSH prevention measures, their monitoring and evaluation (Banner et al., 2022). The policy is thus linked to the other levels of prevention and aims to promote a climate on campus that supports disclosure of SVSH victimization.

Finally, as far as the social dimension is concerned, these interventions must always be considered in the context of the macro level. Above all, this means creating alliances with other universities and other actors involved in the fight against gender-based violence. On a more conceptual level, some research emphasizes the need to always consider gender-based violence as a problem related to social inequalities and to highlight the close connection between multiple structural forms of oppression when designing interventions (Atkinson & Standing, 2019; Banner et al., 2022; Hurtado, 2021). In summary, it is important that all of these levels of intervention 1. include validated programs, such as those for bystanders, 2. are connected through valuable networks and multi-stakeholder involvement, and 3. can take into consideration complaints and promote victim consultation and inclusion.

Bystander Intervention Programs

Bystander programs are currently one of the most widely used strategies for reducing violence. The reluctance to get involved is often due to a number of barriers, both on an individual and socio-cultural level. Being women and having a high acceptance of rape myths (Elias-Lambert, et al., 2023; Lyons et al., 2022b) – and associated personality traits (e.g., the dark triad, see Lyons et al., 2022a) – are examples of personal characteristics that correlate with greater difficulty in intervening. On a more social level, Stojanov et al. (2021) found further potential barriers in gendered power dynamics and in the social and group atmosphere (e.g., male group behavior and the group’s support of the perpetrator).

In general, bystander intervention programs consist of workshops that include both a theoretical part – aimed at raising awareness of the problem of SV and helping to identify behaviors and characteristics of a culture that enables violence – and practical activities (Stojanov et al., 2021) aimed at acquiring skills on how to intervene when witnessing a violent risk situation. In the 8 studies analyzed, the duration of the workshops ranged from 2 to 12 h, divided into several modules, and lasted for different lengths of months; in addition, evidence-based programs are reported in 3 studies, including: Bring in The Bystander (BITB, in Stojanov et al., 2021), RealConsent (in Salazar et al., 2019) and Sex, Safety and Respect (SSR, in McCall et al., 2020). Of these, BITB focuses specifically on bystanders, while RealConsent and SSR have a broader focus and work on multiple levels, i.e. also on perpetrators and victims.

The bystander model views SV as a community problem, so that all members of the community are addressed as potential bystanders and involved in the development of solutions (Beres et al., 2019). Of the studies analyzed, 2 involved only students, 3 involved faculty/staff and 3 involved students, faculty, and staff simultaneously.

An overarching aspect of the studies is their short-term effectiveness, particularly in terms of improving knowledge and awareness of issues related to violence (e.g., Martini & De Piccoli, 2021; Stojanov et al., 2021) and the acquisition of new skills, including practical tools that can be used in intervention (Elias-Lambert et al., 2022). Significant effects after a longer period of time (with a 6-month follow-up) are only reported in one study (Salazar et al., 2019), in which the RealConsent program contributed to both the prevention of SV perpetration and the increase of prosocial bystander behavior via several theoretically proposed mediators central to the program. The relevant aspect of the study is the change in bystander behavior because, as McCall et al. (2020) point out, “changes in understanding are one thing and changes in behavior are another”: the indicators of success will be significant reductions in rates of sexual violence on university campuses. The effectiveness of the studies examined in terms of a short-term increase in knowledge thus represents a solid starting point for the intended long-term changes in the behavior of involved individuals leading to a reduction in rates of SV.

Networks

Dealing with SVSH in academia requires a coordinated commitment from all members of the scientific community and a focus on several areas (Lichty et al., 2018; Rizzo et al., 2020). Starting with students, they should be more involved in outreach, event planning and policy development. One way they can begin to raise their voices to address SVSH in academia, for example, is through the creation of institutionally recognized clubs (Lichty et al., 2018) that allow campaigns and events to be designed and executed with limited institutional interference. Faculty and staff are important in providing role models, resources, and support to students, intervening in incidents of SV (Finley & Levenson, 2018), and helping to shape the academic environment in general (McMahon et al., 2021). Specifically, faculty may be involved in addressing SVSH as: 1. researchers who conduct ongoing evidence-based studies and write scholarly papers on the topic (e.g., sociologists who study gender and/or law and society, see Gronert, 2019); 2. lecturers who develop new materials and curricula; 3. advocates who support student survivors, staff and faculty who are actively working to end SVSH; 4. policy makers who work with faculty senate and college administration to revise existing policies (Graham et al., 2019; Sharoni & Klocke, 2019). Thus, resources to support faculty engagement in preventing and addressing SVSH in academia (e.g., the U.S.-based group Faculty Against Rape (FAR), see Sharoni & Klocke, 2019) are critical.

At a higher level, higher education leaders shape the ethos and climate of the school and set the direction for students, faculty and staff. When they understand the issues, they are better able to create and support the necessary services, procedures and policies (McMahon et al., 2021).

All these levels need to be held together by proposing initiatives for all key college members (see e.g., Hill & Crofts, 2021) to build a strong internal network working in the same direction (Moylan et al., 2022) and also looking for further allies outside the institution. The community response would be more effective if the institutions worked together in a multidisciplinary team to share information and promote community awareness. One way to accomplish this is to create networks between different academies (see, for example, McGann et al., 2020), which could enhance their ability to address SVSH through shared information on resources and events, webinars and training, and shared research, innovations, and best practices (McGann et al., 2020).

The final step is to also build a "communication bridge" to key people outside the academic community, such as local service providers (e.g., staff training in bars and clubs where SV is likely to occur, see McMahon et al., 2021), police and policy makers (Burman et al., 2020).

Addressing these issues requires activism inside and outside institutions, at multiple levels: advising students on strategic approaches to voice their opinions (e.g., Lichty et al., 2018); reviewing university policies and procedures; organizing programs and initiatives within and outside of universities (e.g., McMahon et al., 2021); lobbying industry bodies and policymakers to develop effective policies (Page et al., 2019).

Prevention and Support as Integrated Processes

Support services are considered a form of tertiary prevention: while the SV or SH has already occurred, support can help to mitigate the impact of the incident (Powell & Henry, 2014). This requires clear processes, reporting procedures, training and expertise of those responsible for implementing these processes, as well as those acting as points of contact. The overwhelming majority of harassed individuals did not formally report their experiences. Retaliation, fear of recrimination, lack of trust in the system, inaction, career damage, and lack of clarity about procedures often discourage reporting (Russell et al., 2021; Vargas et al., 2022).

In some cases, the fears behind the decision not to report appear to be confirmed: using the example of first-time reporters (10% of victims) in an academic medical center (Vargas et al., 2022), it was found that a significant proportion of them reported institutional minimization (almost 50%) and retaliation (about 30%).

A more effective implementation of the socio-ecological model could respond to these difficulties, as it aims to challenge the social and institutional culture surrounding SV, including the culture of silence. From this perspective, recent studies (Jones et al., 2021; Mitra et al., 2022) recommend intervening with in-depth, grounded, and contextualized educational programs for higher education personnel to better respond to disclosures of SV that are tailored to the needs of vulnerable and marginalized communities; according to these scholars, training that focuses on skills and competencies is not enough. An educational approach based on feminist pedagogy is required for all staff involved. Another study (Newins & White, 2018) confirms the importance of always focusing on the goal of changing entrenched cultural aspects and shows that students with stronger feminist beliefs indicated that they were more likely to disclose sexual assault on campus.

On a much more concrete level, it would make sense to maximize transparency about the consequences for harassers. Clear behavioral expectations may help increase the frequency of reporting (Russell et al., 2021), including with an anonymous online link for students to report an incident (Wiersma-Mosley et al., 2018), and with clear dedicated websites (Bogen et al., 2019).

Finally, campus sexual violence resource centers should be institutionally supported and closely linked to the numerous prevention education offerings and campus entries (Mitra et al., 2022; Eriksen et al., 2022). From this perspective, the importance of building true “communities of care” to address SVSH is reaffirmed by designing interventions in an integrated manner and with a shared cultural challenge in mind, tailoring them to the needs of those affected, and strengthening alliances with all activatable resources that already exist within and outside the university context (Eriksen et al., 2022).

Discussion and Conclusion

The results of this systematic review suggest that, due to power imbalances, academia is a breeding ground for SH, which is particularly pronounced among academics, as higher prevalence rates are recorded here. Large power imbalances, unclear sexual harassment policies and procedures, brittle employment contracts, and opaque promotion practices — all things that are prevalent in academia — may contribute to an increase in harassment and less reporting at the organizational level. In addition, it is possible that motivated blindness and ethical fading make it more difficult for witnesses, victims, and harassers themselves to be fully aware that sexual harassment is occurring (Mayer, 2014; Tenbrunsel et al., 2019).

Sensitivity to the phenomenon of SH, analyzed across the continuum of SV and gender-based violence, is also developing in contexts other than the U.S., with scales and questionnaires being adapted or ad hoc surveys being created that aim to capture cultural specificities or adapt to local legislation. So far, the phenomenon has been studied mainly in relation to students, although there are more and more surveys that look at the reality of academics and staff or, from a socio-ecological perspective, the entire academic community. The review confirmed that vulnerabilities related to the macrodynamics of power, social inequalities and cultural marginalization are risk factors for SH: women, LGBTQ + , people of different races or indigenous people are most at risk of SVSH. Previous experiences of discrimination or victimization are also risk factors, as are hierarchical and asymmetrical relationships. Graduate students are particularly at risk of being victimized by staff and academics due to the unequal distribution of power.

Given the studies included in the investigation, it is currently difficult to make a cross-national comparison of the prevalence of SH. As Bondestam and Lundqvist (2018, 2020) noted, despite the large number of studies collected, the concept of SH still seems to elude standardized definitions, and only tentative steps have been taken to homogenize research instruments, for example by disseminating the SEQ scale outside the U.S. context. Another limitation concerns the use of intersectional perspectives, which are still insufficiently captured despite the increasing recognition of the relevance of intersectionality in research and at the policy level.

Although almost exclusively limited to the U.S. context, the greatest strength of research and intervention on these issues lies in the proliferation of prevention approaches that adopt a socio-ecological perspective, according to which people’s development and behavior can only be understood by broadening the view to include their psychosocial reality and associated risk factors.

Currently, the most popular evidence-based programs focus on bystanders, which means that all members of the community are involved in creating solutions to address the issue and promote a culture against violence. Some of these programs, such as Green Dot or RealConsent, have already demonstrated long-term effectiveness and have proven to be effective in terms of both knowledge and behavior change. Others, that have undergone short-term evaluation, have shown good results in improving knowledge and awareness of the problem of violence and in promoting new skills, including practical tools that can be used in intervention.

In the logic of the socio-ecological approach, prevention is composed of evidence-based programs and parallel initiatives, events and communication campaigns carried out at different levels within multi-stakeholder networks; the aim is to produce redundant and frequent messages that reinforce each other. The studies included in the review show that evidence-based programs already exist but are underutilized, and that there is a tendency to develop new initiatives, courses and workshops whose evaluations are not always satisfactory.

At present, prevention initiatives still appear to focus on students, but there is a move towards a "whole campus approach" and increasing efforts to involve academics and staff. To this end, the number of actors involved in combating the phenomenon needs to be increased by forging strong alliances both inside and outside the college. It would be desirable to build better alliances also between different universities, possibly in different countries. The main recommendation is that this solid system of interventions, developed in a scientific and therefore reproducible framework, should be exported outside the U.S. context, as has been done with the research to identify and describe the phenomenon, which can be understood as a kind of “needs assessment” to which it is now necessary to respond.

For offices with limited resources, staffed by professionals tasked with both responding to and preventing assault, the call for “comprehensive approaches” may sound far too vague and unattainable (Sisneros & Rivera, 2018). To ensure that comprehensive efforts truly encompass all areas of the college campus, involve multiple offices and staff, and lead to shared responsibility and the “anchoring” of prevention work rather than merely “overlaying” it, ways must be found to engage more stakeholders and community members in the process (Hong, 2017; McMahon et al., 2021). In building these alliances, it is desirable for the university to find successful ways to work with partners from different backgrounds and languages and to train the ability to learn from those with specific practical or technical skills to be supported by improved strategies and interventions. An example of an operational model that can concretize the social-ecological approach is the model piloted at the University of Otago (Beres et al., 2019). From this perspective, intersectionality represents a possible extension of the socio-ecological model, as both aim to overcome the focus on individual aspects by emphasizing social contexts and power relations (Merz et al., 2023). However, the intersectional approach represents an added value because it asks critical questions about the actual inclusivity of the definition and methods used when it comes to including the “community” (Leung et al., 2004), it analyzes in depth the complex dynamics of marginalization and pushes for effective social change that goes beyond simple symbolic steps of inclusion (Hankivsky et al., 2010).

Finally, one of the points that emerged from the research is the need to act at a cultural level, by promoting interventions against SH that emphasize its connection with other forms of sexual violence and structural inequality, and at a practical level, by training coping skills that are implemented in everyday micro-interactions. Only by addressing both levels and taking into account the risk and protective factors distributed in the socio-ecological model it is possible to conceive the college as a “caring community” rich in opportunities for empowerment and support, that should also be offered to female doctoral students and young researchers who are particularly burdened by asymmetrical relationships and inequalities.

Limitations

As this is a general review of the topic of SH in academia, a very extensive and rich body of material has been gathered that would merit further investigation through targeted research using more precisely defined keywords. For example, policies against SH, which have been very present in recent years, were not sufficiently considered in the study. The same applies to the dimension of support, which is also not well represented, but which may reflect the many barriers to disclosure of specific cases of SH that is still largely tolerated in academia, in addition to the difficulty of addressing from a scientific perspective such a sensitive issue that is not yet addressed with clear and shared procedures in most contexts. Finally, the methodology used for the review, which focused on risk factors, led to a preference for studies using quantitative approaches. This decision affected the selection of studies included in the review, with some qualitative research on race and intersectionality being excluded. The reason for exclusion was that these studies addressed the topic in such a specific and detailed way that it did not fit the focus of our objective. A better integration between the socio-ecological model and intersectionality should overcome this kind of limitation.

Furthermore, the review found no targeted research on protective factors, i.e., evidence on what resources to promote in order to support the development of contexts and cultures that are free from SV. The perpetrator perspective is also missing: future research could focus on in-depth studies of the cultures that support SH or the neutralization techniques that normalize and legitimize its presence, which could provide an even better basis for improving prevention programs.