Experiences and Reporting of Unwanted Sexual Behaviors on Great Britain’s Rail Network: A Survey of Victims and Witnesses with an Embedded Randomized Vignette Experiment on Callback Effects

ABSTRACT

Unwanted-sexual-behaviors (USBs) on Great Britain’s rail network, though often seen as “lesser-harm” incidents, have profound psychological, emotional, and behavioral impacts on victims and witnesses. A national survey reveals these effects, highlighting the traumatizing effect of USBs, even years later. Victims share their dissatisfaction with reporting USBs and the need for empathetic communication in victim recovery and trust in authorities. The study also examines the impact of reassurance callbacks from police. While a single callback enhanced perceptions of safety and police effectiveness, multiple callbacks were favored, suggesting their potential to improve victim support and confidence in law enforcement responses to USBs.

KEYWORDS:

• Vignette experiment
• trauma
• unwanted sexual behaviour
• survey
• rail network

Introduction

Unwanted Sexual Behaviors (USBs) remain a pressing concern, with far-reaching effects, impacting people from all walks of life. Essential components of urban infrastructure, such as trains and train stations, are not exempt from this societal problem. The combination of crowds, close proximity between prospective victims and harassers, and the transitory nature of these environments create an ideal setting for USBs to occur. The consequences go beyond the immediate distress suffered by the victims, suggesting suffering throughout societies and challenging our notions of safe communal spaces.

Trains are crucial for connecting communities, driving economic activity, and providing opportunities for education and careers. The efficiency and accessibility of public transport are essential in promoting sustainable urban growth and establishing social equity. However, the prevalence of USBs in these settings poses a significant threat to the realization of these benefits, hindering individuals from fully engaging with and participating in the services offered to them. Women and girls are disproportionately affected, often altering their travel patterns, changing their schedules or avoiding public transportation altogether due to concerns about USBs. This limits the freedom of movements and has broader repercussions.

The term “USB” encompasses a wide range of behaviors deemed unacceptable, from inappropriate comments and gestures to outright physical violence. Even though these incidents are common, they often go unreported for various reasons, which include shame, fear of retaliation, and a lack of trust in the systems in place for seeking justice. The normalization of explicit comments in public transportation only exacerbates the issue, perpetuating a culture of silence and compliance.

Due to the gravity of the issue, both law enforcement and train operating companies have acknowledged that there is a pressing need for improved safety measures in Great Britain’s public transportation systems (https://media.raildeliverygroup.com/news/rail-industry-campaign-to-crack-down-on-sexual-harassment). However, these efforts are frequently hindered due to insufficient concrete evidence of the problem. While the abovementioned effects are globally recognized, USB is surprisingly understudied, particularly in mass transit systems, highlighting the critical need for thorough research and investigation.

This study aims to add to the body of evidence on USB by conducting a survey of individuals who have personally experienced or witnessed USB incidents at train stations throughout Great Britain. We asked detailed questions about the nature of the incident, its effects on those who experienced it, how they coped with the aftermath of the incident, and reporting patterns and attitudes toward the reporting procedures and outcomes. We then solicited responses about suggested ways to improve the services provided to those who report USBs to the authorities through a “bottom-up” approach.

We incorporated in the survey a randomized vignette-controlled trial to investigate the potential contribution of reassurance callbacks. Research has indicated that after initially reporting or presenting their cases to authorities, individuals who have been victims of crimes, as well as individuals in wholly separate sectors including healthcare and consumerism, place significant importance on receiving a follow-up contact (Clark et al., 2022; McKee et al., 2023). Callbacks to victims significantly contribute to the healing processes of those involved, especially when the criminal justice system cannot apprehend or prosecute the offender. Whether the same can “work” for USB victims is presently unknown, and we were keen to explore this strategy with our participants under controlled research settings.

Literature review

What is USB (as opposed to violence against women and girls or sexual harassment)?

As we examine the experiences of people who are subjected to USB, it is first crucial to identify the subtle yet significant differences between USB and “Violence Against Women and Girls” (VAWG), “Sexual Harassment,” and other related concepts.

USB encompasses various non-consensual actions – from verbal remarks to physical interactions – not limited to any specific context, level of harm, or gender. Thus, USB is a broad category that may include behaviors not explicitly covered by legally defined or severe terms, such as VAWG or sexual harassment (Bowman, 1993; C. B. Gardner, 1995). As such, VAWG and USB have different classifications and implications. VAWG, which includes sexual assault, rape and domestic violence, is often categorized as a high-harm crime due to its potential for significant physical, emotional and psychological trauma (Basile et al., 2014). These acts are recognized as violations due to their nature and impact (United Nations, 1993). On the other hand, the term USB can encompass less severe incidents, such as comments, remarks, or other forms of nonphysical or less aggressive sexual misconduct (see review by Ding et al., 2020). While these behaviors can also cause harm (though USB harm is presently undocumented), they may not always involve such significant levels of severity as VAWG (de Bruijn et al., 2006).

Furthermore, gender inclusivity is an integral part of USB, as it is not limited to individuals of a specific sex. Societal and cultural norms often associate USB with women and girls (Crenshaw, 2013; Morin, 2014). Other victim categories, including men and boys or members of the LGBTQIA2S+ communities, are known to be affected by USB as well, even if sexual harassment is disproportionately directed at women and girls (Fitzgerald et al., 1995; see also Nourani et al., 2020). From a legal perspective, terms such as sexual assault, rape, sexual coercion, sexual exploitation, domestic violence and stalking are not inherently gender-specific, and criminologists should be cautious and discuss these experiences more inclusively.

Finally, it is essential to underscore the varying degrees of harm inflicted. Though VAWG and sexual harassment are rightfully acknowledged for their immediate and severe effects, the umbrella term of USB allows for a more comprehensive examination of and response to a broader range of inappropriate behaviors. As recently summarized by Hoor-Ul-Ain (2020, p. 2), USB can be divided into three categories: severe, moderately severe and least severe – although this definition is exclusively about women but can be extended to include all victims:

Severe type of street or public sexual harassment includes rape, passing lewd or sexually explicit profane comments at a woman, physical acts such as following a woman, throwing things at her, groping, pinching or poking her. Instead, the moderately severe type encompasses sexual innuendoes, references to a woman’s gender or body that are not sexually explicit, such as catcalling, whereas the least severe category covers staring or ogling, whistling and all other offensive comments men make at women that are unwelcomed by them. (C. B. Gardner, 1995)

Thus, these experiences may not always align with legal definitions but can still contribute to creating hostile and unsafe environments. USB represents a broad spectrum of non-consensual actions, ranging from relatively minor offenses like sexual harassment, inappropriate comments or lewd remarks to groping, indecent exposure, and more overt and aggressive forms of misconduct (Basile et al., 2014). On the far end of the USB spectrum, one finds violence offenses, including widely acknowledged severe crimes such as rape and gender-based violence, often resulting in documented physical, emotional or psychological damage (United Nations, 1993).

The USB experience in public transportation

In public transportation, USB creates an atmosphere of fear and insecurity (Woodcock & Osmond, 2015). In their exploration of some of these behaviors, Ceccato and Loukaitou-Sideris (2020) revealed a startling reality: most women who rely on public transit have encountered some form of harassment. Similarly, a recent survey shows that almost half of women and a third of men in England and Wales have experienced USB on trains and train stations (2CV, 2020). We acknowledge that the precise degree to which such behavior occurs is unknown because we lack evidence from large-scale, systematic data (see Allen & Vanderschuren, 2016; N. Gardner et al., 2017; Gekoski et al., 2017). As highlighted by Ding et al. (2020, p. 270), “[t]he variation in victimisation rates, from 15% to 77%, likely stems from a mix of factors including differences in research methods[…], how harassment experience may vary according to the specific transit environment, differences in the built environment, and differences in the socio-demographics of the riders surveyed.” We can, however, ascertain that USB takes place across social and economic boundaries, impacting individuals from many backgrounds (although not all social classes utilize public transportation at the same rate) (Giuliano, 2005; Glaeser et al., 2008).

Daily encounters with USB on trains often involve unwanted staring, catcalling and verbal harassment, with almost a fifth of women reporting exposure to indecent acts within the last year, a number which increases to nearly 30% over their lifetime of using public transit (Fielding et al., 2021; Suth, 2003). These experiences are reinforced by interviews and surveys conducted with transport users, highlighting feelings of vulnerability and fear experienced while using public transportation. This is a prevalent issue in many countries worldwide (Anand & Nanda, 2022; Ceccato & Loukaitou-Sideris, 2022), though we are unaware of data that provide a breakdown of these experiences in the mass transit system for LGBTQIA2S+ victims.

Gaining insight into the various elements contributing to the incidence of USBs within public transport is paramount for formulating efficacious preventative and intervention measures. The existing body of literature indicates a variety of factors that contribute to the USB phenomenon, encompassing individual, situational and societal dimensions.¹ We focus here on the situational dimensions.

One factor of crime and USB is its spatiotemporal concentration: crime often occurs in predictable and spatially confined places (Weisburd, 2015). According to Sherman et al. (1989), crime concentrations occur in distinct environments within a city, setting them apart from other areas. These locations, known as crime hot spots, have the potential to either attract or create criminal activity (Brantingham & Brantingham, 1995). Crime generators entice individuals who may not have had prior criminal intentions to take advantage of tempting opportunities. On the other hand, motivated offenders are lured to crime hot spots by the distinct criminal opportunities in these areas, making the locations crime attractors (Frank et al., 2011, p. 1). Public places like train stations can be considered criminogenic due to their ability to provide a sense of anonymity, which is often necessary for specific criminal acts to occur (see recently in Peterson et al., 2023). For these reasons, USBs will likely be facilitated by the right conditions in these hot spots (see Ariel et al., 2023; Ceccato et al., 2020, pp. 5–7). Although no direct evidence of USB offenders’ choices exists, we can speculate that these generalized place based criminology research concepts also fit USBs.

Finally, the design and surroundings of public transportation locations play a significant role in facilitating – and reducing – opportunities for USB in specific locations. Factors such as crowded spaces, inadequate lighting and the transitory nature of public transportation can create an environment where harassment is more likely to occur, and potential targets are more vulnerable. Curiously, these design elements have always been conducive to victimization of passengers; evidence from the United States suggests that sexual harassment on trains was a common experience for women in 1880–1930 (Segrave, 2014), much like it is today. This highlights the importance of understanding situational dynamics, as these environmental factors provide valuable insights for improving safety in public settings through effective design and management strategies.

Mass transit USB in Great Britain

Just as the global evidence suggests a significant variation between different countries, studies from Great Britain are characterized by heterogeneity of prevalence – although the incidence does tend to be lower than in other countries. Different sampling techniques, definitions of USB and periods are some of the reasons for the variations (see Malacad & Hess, 2011). YouGov (2012; in Gekoski et al., 2017, p. 9) surveyed 523 women in London about their encounters with unwanted sexual interaction on public transportation and reported that 19% of all women, 31% of women aged 18–24 and 24% of those aged 25–34 reported unwanted sexual attention. A more recent survey corroborated these results, where “about a quarter of the population were very or fairly worried about sexual harassment on public transport” (Adams et al., 2020, p. 54). Twyford (2013) uncovered a TfL report that showed that 12–15% of women have reported experiencing unwanted sexual behavior. The age group most commonly targeted consists of women aged 16 to 24, with staring, inappropriate touching and verbal comments being the most frequent forms of victimization. These incidents often occur on buses, particularly during evening hours between 5 PM and 11 PM (which coincides with general temporal trends of crime in the mass transit system; see Ariel, 2011).

The most recent survey shows that almost half of women and a third of men in England and Wales have experienced USB on trains and train stations (2CV, 2020). The larger-than-usual rates seem to be linked to this study being the only survey that looked explicitly at USB rather than VAWG or sexual harassment, thus covering more unwanted behaviors.

Impact of USB: emotional and psychological effects, behavioural changes and coping mechanisms

Research on the effects of USB is scarce, but we can draw lessons from research on VAWG and sexual harassment. When individuals experience VAWG, their well-being is significantly affected, both psychologically and physically. Studies have shown a correlation between these incidents in public transportation and increased levels of anxiety, depression and post-traumatic stress disorder (PTSD) (Apell et al., 2019). Those who experienced harassment on public transport often felt vulnerable and powerless, with those feelings lingering long after the incident. Mushtaq et al. (2015). further emphasized the lasting effects of repeated exposure to harassment, citing chronic stress and subsequent psychological distress as consequences.

Fear and intimidation can significantly hinder the use of public transportation by a variety of groups, particularly women and marginalized individuals. This issue can exacerbate existing inequalities and restrict access to essential services such as education, employment, and social mobility opportunities. Ceccato (2012, 2013) has highlighted the impact of feeling unsafe in public places, even if crime is not prevalent. This can lead individuals to avoid the area completely or at certain times of the day. This issue is especially problematic in places where vulnerable populations rely heavily on public transportation and spend significant time in public places (Ceccato & Loukaitou-Sideris, 2020).

Reporting behaviors, bystander interventions and institutional responses to USB

Behaviors of reporting

The problem of underreporting poses a significant barrier to effectively addressing USB incidents. In Japan, for example, over “90% of those who had been exposed to groping on a train did not report it to the officials, and over the half of those being groped reported it to no one […] nearly 60% of those who had experienced groping did not consider groping on trains to be a significant problem that prevented them from using the railway often” (Shibata, 2020, p. 171). In the UK, Garius et al. (2020) highlight that “only 15% of victims of the most serious sexual offences reported the incident to the police, and we may deduce […] that this percentage may be lower for victims of less serious assault” (p. 373). Based on a relatively small sample of respondents, Solymosi and Newton (2020) showed that 59% of passengers had encountered some form of sexual harassment within the past three years while commuting – whether walking, waiting or using public transportation – but fewer than 5% had reported such incidents. Thus, an underreporting phenomenon exists, especially for “lesser harm” USBs.

The lack of reporting can be attributed to a variety of factors, but none of the extant studies look at USB; again, inferences can be drawn from research on the reporting of VAWG and sexual harassment. These include the psychological reaction of victims, who may feel guilt and fear, as well as the societal acceptance of harassment on public transportation as an unavoidable part of life (see Ayres et al., 2009; Woodcock & Osmond, 2015). Additionally, the nature of public transport, which is constantly moving and crowded, creates challenges in identifying offenders and leads to a lack of motivation to report incidents to authorities. This also contributes to a climate of impunity among offenders, which can perpetuate repeated offenses (Heinen, 2023).

Bystander intervention

Recognizing bystander engagement’s role in preventing and reducing USBs is widely acknowledged (see Fenton et al., 2016). The presence of proactive bystanders who are willing to intervene can have a deterrent effect on potential harassers and provide crucial support to victims. Despite this, multiple obstacles, such as fear of reprisals, uncertainty about the appropriate response and the assumption that someone else will step in, can hinder bystanders from taking action (Banyard et al., 2019). Kamau and Wright (2022) also highlight the need for all actors to play a role in preventing these incidents through a zero-tolerance approach.

However, we lack evidence on bystander interventions for victims of USB in mass transit environments. We know that the settings in which harassment takes place play a part in the likelihood of bystanders interjecting once they witness an incident. Lewis et al. (2021, pp. 292–293) observed that the way women react to sexual harassment varies greatly when encountering it in public transport and on the streets, as compared to in the workplace. This can be attributed to the distinct nature of these environments’ interactions and social norms. In a workplace setting, harassment often occurs between individuals with established power dynamics, involving someone the victim knows. In sharp contrast, instances of harassment in public spaces, such as streets and transportation, are mostly committed by strangers. In the public domain, harassment can take the form of overt displays of masculinity. In contrast, on public transport like the underground, it may be a discreet and solitary act (although we do not have published evidence on this matter).

Institutional response

The roles played by transit authorities and law enforcement organizations in tackling the issue of USBs in public transportation cannot be overstated. Taylor-Dunn et al. (2021) emphasize the significance of an organization’s dedication to guaranteeing safety, maintaining transparency in the management of reports and upholding responsibility. These factors play a crucial role in fostering confidence and establishing a secure environment for transportation. Their study advocated for a complete approach highlighting the importance of well-defined policies, rigorous staff training and efficient communication techniques to avoid and effectively handle USB incidents.

Despite widespread recognition of the pressing need to prevent USBs, particularly violence against women and girls, the criminal justice system often falls short in this regard (Martin et al., 2019). After a USB incident has been reported to the police, victims frequently express discontent with the reporting process, feeling neglected by a system that is meant to protect and serve them (Vijayasiri, 2008). The crucial impact of police interactions on public confidence in law enforcement has been highlighted by a number of researchers, including Avery et al. (2020) and Bradford et al. (2009). These interactions are necessary conditions for shaping the public’s trust in policing and emphasize the need for a compassionate and victim-centered approach (Lay et al., 2023). Therefore, USB victims must be taken seriously and have their complaints addressed appropriately, which is not a view held by many victims (Pemberton, 2014).

Regardless of the tangible outcomes of reporting USB to law enforcement, the level of support victims receive has been suggested to be invaluable for people who report offenses to the police. Just procedures, whether they lead to favorable court outcomes or not, are positively regarded by members of the public (Mazerolle et al., 2013). The provision of sufficient and transparent information on the progression of the case was also found to be valued by the complainants.

Indeed, it is common for criminal justice cases to end after the initial report of a crime. This can happen for various reasons, such as lack of sufficient evidence, victims refusing to testify, or prosecutors failing to secure a conviction even if the perpetrator has been identified and charged (as detailed in a forensic study by Buchnik et al., 2023). Only a tiny percentage of cases result in charges and convictions (ONS.gov.uk). While some victims may receive adequate support throughout the process, particularly those who have suffered physical harm, the high number of cases that do not achieve a court resolution can discourage individuals from reporting crimes. This, in turn, may prevent victims from accessing necessary treatment and legal compensation.

It is important to note that in cases where incidents are not reported to the police, limited action can be taken to address the crime issue. On the other hand, it is unclear whether victims desire a court outcome and whether these satisfy them. The literature on restorative justice shows that victims are not as revengeful as commonly assumed, and many prefer to repair the harm caused by their victimization, to have closure and to express their voice (see Sherman et al., 2015; Strang et al., 2013). We suspect the same applies to USB offenses, but research is needed.

Police reassurance callbacks to victims

Using reassurance callbacks is an innovative approach within victim support services that offers ongoing assistance and fosters a sense of safety after an incident. The method is founded on the principles of victim-centered policing and trauma-informed care. The main goal is to provide victims with a sense of security, validate their experiences and inform them about available support resources.

According to the principles of procedural justice – i.e., the concept of fairness in the procedures used to resolve conflicts and distribute resources, and how justice is administered are conducted, focusing on the fairness and transparency of these processes – establishing polite and supportive interactions with law enforcement can lead to improved cooperation and perceptions of justice among victims (Bottoms & Tankebe, 2012). Empirically, there is limited evidence of the effectiveness of this approach in law enforcement, and none at all in the area of USB. Hickey et al. (2021) provide evidence for using reassurance callbacks by law enforcement to convey empathy, provide updates on cases and address safety concerns. Implementing these practices may enhance victims’ overall experiences. Clark et al. (2022) tested the effectiveness of reassurance callbacks from local police officers to the victims of vehicle and cycle crimes. The policy was found to increase the satisfaction of bicycle theft victims by a margin of 5%-15%. In a follow-up to and an expansion of the Clark et al. (2022) study, McKee et al. (2023) examined the impact of reassurance telephone callbacks on the satisfaction of victims of vehicle crime in the entire London metropolitan area. The test suggested that reassurance callbacks lead to victims expressing more favorable attitudes toward the police. This effect was found to be more pronounced among minority victims, especially those who reported their crimes online.

The present study

As reviewed above, there is an informative body of evidence on USB on the mass transit system of Great Britain. There are, however, gaps in several areas. First, the available research focused on sexual harassment or VAWG but did not include other USBs or victim types. We aim to expand on this line of inquiry, with incidents not covered by the definition of VAWG, as USB includes “lesser” incidents regarding the level of harm to which victims are exposed.

Second, research on VAWG excludes non-female participants; even though female victims are more common, the experience of other involved parties remains an understudied area of scholastic and practical interest. In this sense, the experiences of witnesses have remained almost entirely understudied. People who experience USBs vicariously are not only indirectly victimized but may suffer emotional effects as well.

Third, we are interested in a more detailed account of the USB experience. What are the environmental conditions in which USBs occur? Where and when do they occur? What is the involvement of bystanders? These criminogenic descriptors are crucial for a more substantive understanding of USB events, as well as for finding clinical solutions for those who suffer from trauma.

Fourth, and directly linked to the clinical contribution of the findings, there is limited evidence on the effects that USBs have on victims and their coping mechanisms for the events. Are USB victims turning to therapy, their friends or counsellors? Do they change their routines following a USB assault? Do they have psychosocial maladjustments of which the criminal justice system should be aware?

Fifth, we are interested in the reporting patterns of USBs and the reporting experience of victims and witnesses. As we noted, there is a growing body of evidence in criminology on the importance of fair and supportive procedures for victims of crime. However, how victims or witnesses of USBs who reported the incident perceive the reporting process and the outcome is presently unclear.

Finally, we are keen to understand the potential effect of reassurance callbacks on USB victims, especially in crimes not prosecuted by the criminal justice system. As there are promising results from prior trials on the effect of reassurance callbacks regarding property crimes, we tested the effect of this approach with victims of USBs in train stations. We used a scenario-based experiment which presented real victims of USB vignettes with either reassurance callbacks or standard care (i.e., no reassurance callbacks). We then tested the effect of an additional reassurance callback relative to a single callback to investigate whether further calls are causally linked to increased satisfaction, expectations being met, and the safety of victims – taking into account the unique dynamics of public transit environments and the needs of those impacted by unwanted sexual behaviors.

Methods

Study design

We combined a cross-sectional survey with a two-stage randomized vignette component and gave them to a representative sample of the UK population. This design aimed to facilitate a multifaceted exploration of the nature, impact and perceptions surrounding USB in public transport settings and the reporting experience of victims and witnesses. We used a mix of multiple-choice questions as well as multiple questions that elicited open-ended responses from the participants.

The randomized vignette component added a unique dimension to our study. Participants were eligible only if they experienced or witnessed USB in Great Britain’s trains or train stations. They were presented with scenarios depicting USB incidents, with or without reassurance callbacks scenarios systematically varied. This approach allowed an experimental examination of how different factors – such as the severity of the incident, the response of bystanders, and the demographics of the victim and perpetrator – influence public perceptions, attitudes and the propensity to report such incidents.

Participants and sampling

We recruited participants in partnership with YouGov. YouGov possesses a comprehensive and inclusive panel of individuals from various backgrounds in Great Britain. The study included a sample of participants who met the criteria of (a) being 18 or older and having used rail services at least once over the previous 12-month period;¹ (b) having experienced or witnessed USB in the train stations or on trains; and (c) consented to take part in this novel survey. We adopted a stratified sample approach to achieve adequate representation across several demographics, such as age, gender, socio-economic position and geographical location. Overall, 1,144 participants took part in the study.

Data collection procedures

YouGov uses active sampling for its commercial public opinion surveys, including nationally and regionally representative research online. Restrictions are put in place to ensure that only those contacted participate in active sampling. All YouGov survey respondents are selected from panels of registered users, and only those selected can contribute. A survey panel out of millions of potential participants is then recruited through advertising and strategic partnerships with many websites. When a panelist is hired, many socio-demographic details are recorded. Then, YouGov draws a sub-sample of the panel representing British adults in age, gender, social class, and education to complete a survey for nationally representative samples. In our case, the strict inclusion criteria mentioned above were implemented.

Survey instrument

The survey comprised dimensions that appeared in a collection of studies on VAWG or related sexual harassment studies in the mass train system and added new constructs. As alluded to earlier, it added an embedded randomized vignette experiment on callback effects. A copy of this instrument can be found in Supplementary Materials A.

The survey begins with a formal definition of USB. Studies have found that while people may experience USBs, many do not necessarily see themselves as victims, and the majority of individuals who encounter the behaviors described in the Sexual Experience Questionnaire do not identify themselves as victims of harassment (Fitzgerald et al., 1995; Lonsway et al., 2013; Magley et al., 1999). We therefore provide a series of USB examples to avoid any confusion. Only participants who confirm that they have experienced or witnessed USB in accordance with this definition were eligible and those who did not were excluded from the survey.

We first asked participants a series of questions designed to determine their sociodemographic status and their level of use of public transportation (we included only participants who use the train network in Great Britain) and to identify those who have experienced USB as victims, witnesses or both. We then asked a series of questions nested in four primary dimensions:

1. The nature of the USB incident they experienced or witnessed and the most recent encounter details.

2. The emotional and psychological effects the event had on the participant and the coping mechanisms used by the participant to manage the aftermath of the incident(s), including any behavioral adaptations they implemented (e.g., change of travel arrangements) or help services they may have used following the incident.

3. Reporting behavior, such as to whom the incident was reported and what the reporting experience was like, including satisfaction with the procedures, the outcome of the complaint, and the emotional support that victims and witnesses received from the authority to which they reported both during the reporting process as well as after the complaint was lodged.

4. We then presented randomly assigned vignettes about reassurance callbacks versus regular care following the complaint. We aimed to gauge from the participants (who had experienced actual USB incidents as part of the selection criteria) their views on the utility of this approach relative to regular care. We asked a series of questions about the level of support they perceived they were receiving through this approach, as well as the safety and effectiveness of the police in supporting victims of USB. We then randomly assigned two additional vignettes. Participants were either re-contacted by the authority to whom they lodged the complaint in a follow-up reassurance phone call or were not given a follow-up call. They were then asked questions about expectations, perceived level of reassurance, safety and support through Likert Scale questions.

The vignettes began with a preamble to explain the purpose of this section, and then participants were randomly assigned to one of the two scenarios:

Scenario A1:

Imagine you have just reported to the British Transport Police a sexual harassment incident that occurred on a train. Two weeks after making the report, you do not receive any further communication or follow-up from the police, except a letter informing you that your report has been closed due to insufficient evidence. They do not contact you for reassurance or updates on the investigation beyond this formal letter.

Scenario A2:

Imagine you have just reported to the British Transport Police a sexual harassment incident that occurred on a train. Two weeks after the report, you receive a letter informing you that your report has been closed due to insufficient evidence. After receiving this letter, you receive a follow-up phone call from the police. They express concern for your well-being, inform you they have taken your report seriously, and assure you that it helps prevent future victims by understanding where, when and how crime is committed – even though they closed the case.

The second set of scenarios gauged participants’ views of receiving a second follow-up call or not:

Scenario B1:

Now, imagine the same scenario where you receive the initial reassurance call after reporting the incident. However, in this scenario, you receive a second phone call from the authorities three months later. During this call, they reassure you again by checking in on your well-being, providing an update on the investigation if available, and offering their support and assurance that they are committed to addressing the incident.

Scenario B2:

Now, imagine the same scenario where you receive the initial reassurance call after reporting the incident. You have no further contact from the police after the first reassurance call. The police do not check in on your well-being, do not provide an update on the investigation if available, and do not offer their support and assurance that they are committed to addressing the incident.

Randomisation into the two arms (in each scenario) achieved baseline equilibrium, with no statistically significant differences between the participants in terms of their baseline characteristics (see Supplementary Materials B).

Data analysis plan

Formally, we presented a series of multiple-choice choices to participants for which they could choose whether any of the options in each question applied to their experiences. Given USB’s multifaceted and complex nature, we did not force one response for most questions. Likert scale questions about certain constructs – safety, support, expectations, etc. – were also used to measure the degree to which participants viewed some aspects of the USB experience. In every question, given the subject’s sensitivity, participants were allowed to choose “prefer not to say.” Finally, many sections used open-ended questions to allow participants a voice and provide room for further qualitative analysis. The data from the survey can be found in Supplementary Material B.

We focus on descriptive rather than inferential statistics in this paper and aim to echo the voices of victims and witnesses of USB on the mass transit system in Great Britain. We provide frequency distributions for each question and present the findings as rates of respondents to each survey item. Whenever relevant, we compared the responses of victims and witnesses to estimate whether their experience of USB is significantly different using odds ratios (OR) and indicated the 95% confidence intervals (CIs) of this result.

Using thematic content analysis, we examined the responses to the open-ended questions in our survey. We thoroughly examined and revisited the response data based on previous studies.

We employed thematic content analysis to explore the open-ended reported experiences. This analysis was facilitated by software integrated into Research.net©, allowing for the systematic coding of responses to open-ended survey questions. Our methodology was designed to dissect the qualitative data provided by participants, enabling us to identify, analyze, and report patterns (themes) within the data. The process began with a reading of responses to gain a deep understanding of the data’s breadth and depth. This initial reading was critical in developing a coding scheme that accurately reflected the responses’ content and nuances.

The coding scheme was developed iteratively. Initially, a set of codes was generated based on a preliminary analysis of a subset of the data. These codes were descriptive and directly related to the participants’ responses. As we progressed, these initial codes were refined, combined, or subdivided into more specific themes, allowing for a nuanced understanding of the experiences shared by participants. This coding process was not linear but involved constant comparison across the dataset, ensuring that the coding scheme remained relevant and comprehensive as the analysis proceeded.

The thematic content analysis, supported by the automated coding capabilities of our software, provided a structured approach to understanding the complex phenomena of USBs in public transport. By systematically coding and analyzing the data, we uncovered our participants’ multifaceted experiences, offering insights into the prevalence, nature, and impacts of USBs.

We then used word clouds to visualize common texts that appeared in the open-ended responses using high-frequency words or phrases used by participants. We used wordclouds.com to create visual representations; the size of each word indicates the frequency at which the words appear in the text.

Finally, we analyzed differences between the two experimental arms using independent sample t-tests and chi-square statistics. Then we analyzed the magnitude of the differences between the groups in terms of effect sizes (either Cohen’s d or phi, depending on the distribution of the data), as well as the associated 95% confidence intervals, to estimate the degree to which significant differences emerged as a result of the reassurance callback or the additional contact by the police.

Results

Baseline characteristics

As listed in Table 1, our participants looked similar to the UK population (see www.ons.gov.uk). The median age of the sample was 45.0; they were White (80.8%) and heterosexual (76.8%), and the majority of respondents held a university degree. About a quarter of respondents were residents of London, and 43% earned an annual salary of up to £49,000 (69% were employed, and 15.4% were retired). Under 60% were married or in civil partnerships, and a similar number owned a house. About 30% used the railways at least weekly.

Table 1. Sample characteristics (n = 1,144).*

Variable
Median Age	45
Education qualification (highest attained)
University Undergraduate or Graduate Degrees	60.0%
No formal qualification	1.6%
% Female	73.9%
Region Lived
London	23.4%
All other English regions	64.5%
Scotland	6.3%
Northern Ireland	1.3%
Wales	4.5%
Income – gross household
Up to £24,999 per year	19.0%
£25,000–£49,000 per year	24.0%
£50,000–£99,000 per year	27.6%
£100,000+ per year	11.9%
Marital status
Married/Civil Partnership	59.0%
Divorced/Widowed	8.1%
Not in a Relationship	29.9%
House Tenure
Own	61.9%
Rent	24.6%
Other	8.6%
Employment Status
Working	69.2%
Retired	15.4%
Student	2.3%
Unemployed	9.8%
Ethnicity
English/Welsh/Scottish/Northern Irish/British	80.8%
All other	17.6%
Sexuality
Heterosexual	76.8%
All other	17.1%
Use of Trains
Daily	8.8%
Weekly	20.7%
Monthly	18.7%
Less often than monthly	51.7%

^*Figures do not round up to 100% due to participants skipping questions or refusing to respond.

Nature of USB incidents on trains and in train stations

73.4% of participants of sample participants experienced first-hand USB on trains or in train stations in Great Britain; 29.7% witnessed USB in the same circumstances (Figure 1). About 1 out of 4 would report the incident to the authorities.

Figure 1. USB while on a train or at a train station: personal experience or witnessed experience (n = 1,144).

Table 2 provides an account of USB in terms of the type of sexual harassment participants experienced or witnessed, or when the event is more likely to take place and the role of bystanders in these situations. Verbal sexual harassment is the most common type of USB, and verbal comments or jokes of a sexual nature are experienced by 53.0% of the direct victims and 74.3% of witnesses. Touching or groping, as well as staring or leering, was experienced by more than half of the participants, while nearly one out of ten of the participants witnessed indecent exposure. Nearly 39% said they experienced frottage (i.e., pressing or rubbing against people for sexual gratification). 13 participants said they were raped.

Table 2. Nature of the sexual harassment incident on a train or at a train station.

Variable	Experienced Sexual Harassment	Witnessed Sexual Harassment^
Type of Sexual Harassment Incident (lifetime exposure)
Verbal comments or jokes of a sexual nature	53.0%	74.3%
Unwanted touching or groping	51.8%	62.2%
Sexual gestures or mimes	30.5%	25.0%
Indecent exposure	20.5%	9.2%
Sharing explicit images or messages without consent	3.6%	5.3%
Frottage (pressing or rubbing against people)	38.7%
Staring or leering in a sexually suggestive manner	51.4%
Using names like “slut” or “whore”	14.3%
Unwanted sexual communications, like e-mails, texts, DMs	4.6%
Stalking	9.4%
Upskirting	2.9%
Rape	1.1%
Pressure or offer of something in exchange for sex	4.3%
Encounter Details (for the most recent instance)
Recency
Within the last week	0.5%
Within the last month	3.0%
1 to 3 months ago	8.2%
4 to 6 months ago	8.8%
7 to 12 months ago	8.8%
13 months to 2 years ago	12.1%
More than 2 years ago	53.3%
Can’t remember	4.2%
Time of Occurrence
Weekday, between 7 AM – noon	10.3%
Weekday, between noon − 5 PM	10.6%
Weekday, between 5 PM − 9 PM	23.0%
Weekday, between 9 PM − 7 AM	10.1%
Weekend, between 7 AM – noon	2.1%
Weekend, between noon − 5 PM	5.8%
Weekend, between 5 PM − 9 PM	12.4%
Weekend, between 9 PM − 7 AM	12.1%
Can’t remember	13.0%
Witnesses to the most recent incident experienced
Yes, definitely	35.8%
Yes, but not sure	30.8%
No, there were not	23.1%
Can’t remember	9.4%
Reaction of Witnesses
They pretended not to see what was happening	59.7%
They watched at a distance what was happening	33.7%
They got involved – tried to stop it	17.7%
They got involved – asked victim for their wellbeing	15.7%
They talked to the offender	7.7%
Other	10.3%

^{^}Empty cells indicate that the response option is not available for participants.

Perhaps unsurprisingly, given the composition of the sample (median age 45), as well as what we know about the declining risk of USB as the age of victims increases,² more than half reported that their encounter took place more than two years prior to the survey. Nevertheless, 29.3% had experienced USB within 12 months of the survey. Afternoon rush hours (weekdays between 5 PM − 9 PM) were the riskiest time for USB, coinciding with general crime patterns in the mass transit system (Ariel, 2011).

Given the congested nature of this environment, USB is likely to be witnessed by others, with only 23.1% of the participants suggesting that no bystanders were present during their victimization. However, victims of USB shared that in 6 out of 10 incidents, witnesses either pretended not to see what was happening or watched from a distance (33.7%). Some witnesses approached the victim to ask after their well-being (15.7%), while some got involved and tried to stop the incident (17.7%) or talk to the offender (7.7%). A common theme in this sense for both victims and witnesses of USB was the lack of bystander intervention from the immediate environment, which was also expressed in quantitative analysis (Table 2).

As noted in the method section, we asked participants about their experience with USBs. Multiple themes arose from these responses (Table 3; see Supplementary Materials B). First, the modus operandi of offenders is broad, ranging from verbal harassment, including catcalling and aggressive language or rude comments, through acts of a sexual and wanton nature like lewd acts, masturbation in the victims’ presence and exhibitionism, to intimidation or direct threats. Witnesses shared similar encounters but noticed more invasions of personal space in incidents they viewed. A recurring theme was alcohol and drugs, with many harassers believed to be intoxicated during the time of the USB act. Most participants suggested that the primary victim category was females, particularly those whom they construed to be vulnerable young transport users, with misogyny appearing as a common theme in open-ended comments expressed by both victims (n = 692) and witnesses (n = 251).

Table 3. The USB experience in the train environment: thematic analysis of participants’ experiences (based on 692 comments by victims and 251 comments by witnesses).

	Themes Shared by Victims	Themes Shared by Witnesses
USB categories	Unsolicited Physical Contact: A significant portion of the responses involved incidents where individuals experienced unwanted physical contact, ranging from groping and touching to more aggressive forms of sexual assault. These included men rubbing against women in crowded carriages, touching intimate areas, or more forceful actions like grabbing or attempted kissing. Exposure and Lewd Acts: Many respondents reported incidents where individuals (predominantly men) exposed themselves or performed lewd acts, such as masturbation, in their presence. This type of harassment was not only direct but also occurred through actions like flashing or exhibitionism. Verbal Harassment and Intimidation: Several responses mention instances of sexually explicit comments, catcalling, and aggressive, threatening language. This also includes instances where individuals were followed or stalked by harassers, adding to a sense of fear and vulnerability.	Stalking or Persistent Unwanted Attention: Some responses described situations in which individuals persistently followed or tried to engage with passengers who had expressed disinterest or discomfort Verbal Abuse: Many responses highlighted instances of verbal abuse, including offensive, derogatory or threatening language directed at passengers. Physical Assault: This included more direct physical aggression, such as groping, hitting and other forms of unwanted physical contact Invasion of Personal Space: This involved harassers invading a passenger’s personal space, often as a precursor to more direct forms of harassment.
Victim categories	The responses suggest that women, particularly young women, feel particularly vulnerable to these types of incidents. Additionally, there are instances involving men as victims, indicating that the issue spans across genders, although with a higher prevalence among women.	Incidents where harassment was specifically targeted at a person based on their gender, often involving misogynistic or sexually suggestive remarks.
Role of Alcohol or Drugs	In several instances, the harassers were described as being under the influence of alcohol or drugs, which might have contributed to their behavior.	A significant number of incidents involved harassers who were under the influence of alcohol or drugs, leading to aggressive or inappropriate behavior.
Response to USBs (authorities and bystanders)	A recurring sentiment in many responses is the lack of intervention from others or support from authorities. Victims often felt isolated in their experiences, with bystanders not stepping in to help or authorities not being present or responsive.	Some responses included instances where bystanders intervened to help the harassed individual. Several responses pointed to a lack of intervention or inadequate response from transport staff or authorities when harassment occurred.
Impact on Mental and Emotional Well-being	While not explicitly stated in every response, the underlying impact on the victim’s mental and emotional well-being is evident. The experiences described are likely to have long-lasting effects, contributing to feelings of fear, anxiety and mistrust in public spaces.	Witnesses often faced notable psychological and emotional distress. Incidents can leave lasting effects, fostering apprehension, unease, skepticism and a perception of risk in communal environments.

Many respondents suggested that there were no formal, capable guardians near the incident (e.g., police officers or security guards) or described inadequate responses from transport staff or authorities when harassment occurred.

The word maps (Figures 2 and 3) further represent these themes, with the mental and psychological effects on the participants’ well-being emerging as common terms expressed in the analyzed text, as we discuss in the next section.

Figure 2. Please describe the most recent incident in which you were victimized (n = 692).

The larger the word, the higher the frequency of its appearance in open-ended responses.

Figure 3. Please describe the most recent incident in which you witnessed victimization (n = 251).

The larger the word, the higher the frequency of its appearance in open-ended responses.

Emotional, psychological and behavioural effects of USB and coping mechanisms

Immediately after the event, victims felt violated, uncomfortable, angry and shocked. Many felt embarrassed and anxious as well. Some of these effects lingered for an extended period of time; in fact, the majority of victims of USB incidents felt anxious or stressed up to two years after it took place – a view shared by 61.2% of the participants. USB left victims angry (47.4%), with a quarter feeling fear or paranoia and low self-esteem. Nearly 1 in 10 felt withdrawn, isolated, depressed or hopeless and had difficulties concentrating or focusing. These and additional adverse effects are listed in Table 4.

Table 4. Impact and coping mechanisms.

	Per cent^^
Impact of the incident up to two years after it took place
No noticeable impact on well-being or mental health	15.4%
Anxiety or stress	61.2%
Anger	47.4%
Fear or paranoia	25.6%
Low self-esteem	22.5%
Difficulty concentrating or focusing	10.5%
Isolation or feeling withdrawn	8.8%
Depression or hopelessness	8.1%
Trouble sleeping or nightmares	7.5%
Other	5.5%
Attempts to get help to deal with the aftermath of the incident
Talking to friends or family	52.6%
Seeking support from a therapist or counselor	6.7%
Engaging in activities such as exercise, meditation, or hobbies	10.5%
Writing in a journal	5.7%
Joining a support group	3.3%
Avoiding reminders of the incident	17.5%
Taking time off work or school	4.0%
Using substances (e.g., alcohol, drugs)	4.5%
Talking to someone at the station/on the train immediately after	11.4%
None of these	25.7%
Feelings immediately after the incident
Uncomfortable	70.0%
Fearful	39.4%
Angry	61.3%
Helpless	32.6%
Shocked	43.3%
Violated	48.5%
Anxious	38.2%
Confused	15.4%
Embarrassed	38.5%
Sad	13.2%
Travel Modifications following the incident
Changed daily habits in general	25.1%
Changed daily commute or regular transportation habits	25.4%
Avoided getting trains at certain times of the day	36.6%
Avoided certain train stations or train routes	23.3%
Used trains less often	32.6%
Avoided traveling by train completely	11.2%
Avoided traveling by train or being at train stations alone	33.4%
Moving to different areas of a train or station to avoid certain people	61.8%

^{^}More than one response was allowed.

Some victims attempted to get help to deal with their victimization by talking to friends or family members (52.6%), therapists or counselors (6.7%), or joining support groups (3.3%). 10.5% engaged in positive behaviors like exercise, meditation or their hobbies, while 17.5% avoided any reminders of the incident and 4.5% self-medicated with drugs or alcohol. A quarter of the sample did not take action to deal with the incident’s aftermath.

The USB incident caused victims to change their daily habits: a third avoided getting on trains at certain times of the day, avoided certain train stations or routes, or used trains less often. The most common change was moving to different areas of the train or station to avoid certain people (61.8%).

Reporting USBs

Most participants did not report the USB incident to anyone, although the odds of reporting were greater for witnesses than for victims (OR = 1.683, 95% CI 1.414, 2.003). The most common authority figure to whom the event was reported was the rail or security staff member where the event took place (11.9% of victims and 17.1% of witnesses), followed by the British Transport Police – in person or on the phone or using the text helpline (approximately 8%). If the event was to be reported, 71.8% did it at the time or immediately thereafter. These patterns are displayed in Table 5.

Table 5. Reporting behaviour and satisfaction with reporting process and its outcomes.

	Experienced Sexual Harassment	Witnessed Sexual Harassment
Reporting the Incident
No	77.3%	73.5%
Yes, to the British Transport Police in person	7.6%	8.1%
Yes, to the British Transport Police on the phone or using the text helpline	6.5%	7.6%
Yes, to the police other than the British Transport Police	5.1%	1.0%
Yes, to the rail or security staff where the event took place	11.9%	17.1%
Yes, to someone else working at the station (e.g., a newsagent worker)	2.3%	2.3%
Yes, to health services	1.1%	0.7%
Yes, to social services	0.7%	0.7%
Yes, at school	0.8%	0.7%
Reason for not reporting the incident
I wanted to but wasn’t able to find anyone to do so	11.1%	19.6%
I didn’t know what number to call/text to report it	13.6%	20.1%
I didn’t know you could report such incidents	23.7%	17.5%
I didn’t feel the incident warranted reporting	32.2%	14.3%
I felt ashamed to	15.8%	–^
I didn’t feel it would be taken seriously	50.7%	37.0%
I didn’t want to relive the experience	14.7%	–^
I was too frightened to	6.4%	11.1%
Timing of reporting
At the time of/immediately	71.8%
At a later date	22.1%
Satisfaction with the response received after reporting the incident	3.64 (1.31)^Ω
I had no response at all	7.8%
Satisfaction with the outcome of the report	3.20 (1.33) ^Ω
Informed about the procedures following the report
Yes, I was	46.8%
No, I was not	49.6%
Perceived fairness and just process?
Yes	51.4%
No	22.9%
Don’t know	24.3%
Authorities offered emotional support during the reporting process
Yes	33.6%
No	53.6%
Don’t know	11.1%
Authorities offered emotional support after the report
Yes	22.1%
No	66.1%
Don’t know	9.6%
Receive feedback on the actions taken against the harasser(s) if identified
Yes, full feedback	17.6%
Yes, partial feedback	15.8%
No	41.4%
N/A – harasser(s) were not identified	22.7%

^{^}not asked; ^Ωon a scale of 1 (very dissatisfied) to 5 (very satisfied), mean scores and (standard deviations).

Participants gave many reasons for not reporting the incident, with similar views expressed by both victims and witnesses. Over half the victims felt that the incident would not be taken seriously, a view that 37% of the witnesses shared. Interestingly, many participants shared that they did not know how to report the incident (13.6% of victims and 20.1% of witnesses) or did not know that such incidents could be reported (a third of victims and about 14.3% of the witnesses). 15.8% of victims did not report because they felt shame, and 14.7% did not want to relive the event.

Further depicted in Table 5 is the satisfaction from reporting the incident and the outcome of the complaint. Overall, complainants mainly were satisfied with the response they received (3.64 on a scale of 1–5). Half viewed the process as fair and just. However, 49.6% said they were not informed about the procedures following the report, and 41.4% did not receive any feedback on the actions taken against the offender (excluding cases where they were not identified). 53.3% and 66.1% did not feel that the authorities provided emotional support during or after the complaint process, respectively.

Reassurance callbacks

Test 1

The results of the randomized controlled trial portion of the study are presented in Table 6. We detected statistically significant differences across measures, indicating that the reassurance callback stimulus was preferred by the participants exposed to it relative to those who received no reassurance callbacks. Under the reassurance scenario, victims felt 68% more support and 19% safer, and the participants perceived the police as more effective in helping victims by 54% relative to the business-as-usual arm. All differences are statistically significant at the p < .001 level.

Table 6. Reassurance vs. no reassurance call backs; and 1 vs. 2 reassurance call backs – t-tests, chi-square statistics, and effect sizes (Cohen’s d and Phi).

		N	Mean (SD)	t/χ^2	d (95% CI)/Phi
Reassurance vs. No Reassurance Call Backs
Felt Supported^a	Usual Care	547	1.71 (1.60)	14.566 ***	.874 (.751, .997)
	Reassurance Calls	561	2.68 (1.16)
Felt Safer^b	Usual Care	553	2.05 (1.12)	7.302 ***	.438 (.319, .557)
	Reassurance Calls	559	2.54 (1.12)
Police effective in helping victims^c	Usual Care	435	11.30%	26.833 ***	0.178
	Reassurance Calls	409	24.90%
One vs. Two Reassurance Callbacks
Expectation of another reassurance call^d	One Callback	288	2.98 (1.52)	−5.574 ***	−.470 (−.637, −.302)
	Two Callbacks	273	2.30 (1.39)
Felt Safer^b	One Callback	288	2.15 (1.05)	9.400 ***	.798 (.625, .970)
	Two Callbacks	268	3.00 (1.08)
Felt Supported^a	One Callback	289	1.94 (1.04)	13.555***	1.149 (.970, 1.327)
	Two Callbacks	273	3.23 (1.20)

p ≤ .05; **p ≤ 01; ***p ≤ .001.

^aOn a scale of 1 (not supported at all) to 5 (very supported).

^bOn a scale of 1 (not safe at all) to 5 (very safe).

^cper cent agreeing.

^dOn a scale of 1 (would not expect another call at all) to 5 (expect very much).

Test 2

We show that assigning an additional call following the first reassurance call is advantageous within the subgroup of participants randomly assigned to the reassurance arm of test 1 in the study (n = 561). Participants who did not receive the second call expected further callbacks, and they felt less supported and less safe than those who received two calls. Again, all differences are statistically significant at the p < .001 level.

Discussion

This paper contributes to our comprehension of public safety by highlighting the lived experiences of those who regularly navigate the mass transit environment. Below, we summarize the results and discuss the survey’s theoretical, practical and future research implications.

Summary of findings

While we cannot ascertain, based on the results of this study, the prevalence of USBs on trains or in train stations, the findings illustrate the nature of these incidents and their situational factors. As in previous studies, we show that verbal harassment was the most common form, followed by touching, groping and more severe forms, including frottage and rape. Incidents often occurred during rush hours and were frequently unwitnessed or ignored by bystanders. This analysis reflects a concerning pattern of sexual harassment and assault in public transport environments. It underscores the need for more effective measures to ensure safety and support for victims, as well as a broader cultural shift toward addressing and preventing such behavior.

USB incidents were reported to the authorities at a low rate by both victims and witnesses. The primary reporting was to rail or security staff and the British Transport Police. Reasons for not reporting included skepticism about being taken seriously, shame, lack of awareness of reporting procedures, and a desire to avoid reliving the event.

Participants informed us of their immediate emotional responses to USB, including feelings of violation, discomfort, anger and shock. Long-term effects like anxiety, stress, anger, fear, low self-esteem, depression and social withdrawal were noted by the victims. Coping strategies varied from seeking social support to avoidance and substance use. Behavioural changes included altering daily routines and travel habits to avoid potential USB situations.

The experimental component of our study suggests that reassurance callbacks from authorities significantly improve victims’ feelings of support and safety. The effectiveness of the police in helping victims was perceived to be higher when reassurance callbacks were made. Further, the benefit increased with an additional follow-up call.

Theoretical implications

Routine activity theory

This study’s findings can be interpreted through the lens of routine activity theory, which suggests that the convergence of motivated offenders, suitable targets and the absence of capable guardians facilitates the occurrence of USB (Clarke & Felson, 1993). Thus, these incidents seem not random but are influenced by specific conditions. The high incidence of USB during rush hours can be seen as a convergence of these elements – overcrowded trains (providing anonymity to offenders and a more significant number of potential targets), the presence of potential victims, and a lack of guardianship (e.g., limited security or police presence), as found in Sao Paulo’s metro system as well (see Figure 2(c) in Ceccato & Paz, 2017).

Furthermore, the findings regarding the demographic characteristics of victims, particularly the vulnerability of young female transport users, align with the theory’s suitable targets. Victims of USB may be perceived as vulnerable or less capable of defending themselves, making them attractive targets to offenders. On the other hand, the characteristics of motivated offenders, often involving substance abuse or showing misogynistic tendencies, highlight the individual factors that drive the propensity to commit USB as well.

Environmental criminology

Environmental criminological theories, such as Crime Prevention Through Environmental Design (CPTED), could be relevant for understanding how the physical design of public transport systems can influence the occurrence of USB (Newman, 1973). Thus, understanding the structure of the train stations and carriages is needed to explain further what contributes to mitigating the risk of USB. Overcrowded trains, for instance, are not just a matter of commuter discomfort but can also be seen as facilitating environments for USB. The proximity of passengers, coupled with the anonymity provided by the crowd, creates an environment where offenders might feel emboldened to commit sexual offenses, assuming that there exists a lower risk of identification and apprehension.

Furthermore, the lack of surveillance or effective monitoring in train stations and on trains could contribute to the high incidence of USB. CPTED suggests increasing visibility through better lighting, CCTV surveillance, and placing staff or security personnel in visible locations can deter potential offenders. More research is also needed here, as the optimal designs that would maximize visibility and foster a sense of ownership and vigilance among legitimate train system users are warranted but remain to be discovered.

Psychological impact and trauma theory

The long-term psychological effects of USB on victims, such as anxiety, stress and depression, warrant further exploration. The literature on victims often overlooks crimes that cause “less” harm than physical assaults, such as violence or rape. Nevertheless, our participants, who were victims of USB that the criminal justice system undervalues relative to crimes that cause more tangible manifestations of harm, report on the psychological impact of USB in public transport, which can be considered as trauma.

Trauma Theory (Herman, 1992) can provide a framework for understanding USB’s complex and often prolonged psychological effects. Traumatic events like experiencing or witnessing a USB can lead to lasting psychological effects that profoundly affect an individual’s sense of self, their perception of the world, and their ability to feel safe and secure. The reported effects of USB incidents, including anxiety, stress, anger, fear, low self-esteem, depression and social withdrawal, align with the symptoms typically associated with trauma. These responses are not just immediate reactions to a distressing event but can persist long after the incident, impacting the victim’s mental health and overall well-being for an extended period. Unsurprisingly, many change their daily routines as a result of the USB: they want to avoid a reoccurrence of the trauma.

These trauma responses highlight the need to conceive more substantive trauma-informed approaches in both policy and support services. These include recognizing the signs and symptoms of trauma in victims of USB, providing them with access to appropriate mental health support, and ensuring that the response by authorities and the public is sensitive to their trauma. The fact that so many victims say that they did not feel they were taken seriously and their apparent desire for reassurance callbacks – even if their case is to be discontinued from a criminal justice perspective – calls for a new approach to helping victims recover and preventing further psychological harm.

Policy implication

Reassurance callbacks

Introducing reassurance callbacks as a policy response to USB in public transport could have far-reaching implications, extending beyond immediate victim support to broader aspects of law enforcement and public safety. These callbacks, wherein authorities proactively contact victims to offer support and updates, embody a critical shift toward a victim-centric approach (Lay et al., 2023), perhaps especially beneficial in cases discontinued in criminal justice system outcomes.

Firstly, implementing reassurance callbacks aligns with the principles of trauma-informed care. It acknowledges the psychological impact of USB and addresses it through empathetic, proactive communication. This approach necessitates that police officers or staff are trained in effective communication and trauma awareness, ensuring they can provide the necessary reassurance. Integrating such callbacks into standard response protocols can significantly enhance victim support and recovery, offering reassurance and a sense of security in the aftermath of an incident – although we acknowledge that field trials rather than a survey-based experiment are needed before we can make a sound policy recommendation. For example, there is a risk that callbacks can backfire if a substantive, long-term solution to the core risk of USB is not identified. Future studies should address this issue.

If the results are generalizable, then reassurance callbacks can substantially improve the perception of the effectiveness and responsiveness of law enforcement and transport authorities. Regular (rather than one-off) empathetic communication with victims can build and reinforce public trust in these institutions. This is crucial for encouraging the reporting of USB incidents and fostering investigation cooperation (Clark et al., 2022). Knowing that authorities provide ongoing support and updates might encourage more victims and witnesses to report such incidents. Increased reporting provides a more accurate picture of the prevalence of USBs, aiding in identifying patterns and hot spots for targeted interventions. Policies prioritizing victim engagement and follow-up can send a strong message about the seriousness with which USB is treated, thereby strengthening community relations (McKee et al., 2023).

USB hot spot policing

From a routine activities theory perspective, strategies to prevent USB in public transport should focus on disrupting the convergence of the three elements. This can include increasing the presence and visibility of security personnel or staff trained to identify and intervene in USB incidents and improving surveillance systems.

The absence of capable guardians is critical. The participants underscored the lack of bystander intervention and inadequate responses from transport staff or authorities when the harassment occurred, reflecting this gap. It may be that increasing the presence of guardians, such as police officers, security staff, or even vigilant commuters in “USB hot spots” could deter potential offenders. Note that ordinary street crimes and USB hot spots may not necessarily overlap (Moreira & Ceccato, 2021), and a specialized focus will likely be required to prevent crime at USB hot spots. We also note that witnesses often do not intervene in such incidents (Tripathi et al., 2017). However, further research is needed to substantiate this approach, as we are unaware of published evidence on the effect of USB hot spot policing initiatives (see proposal for this approach in Ariel, 2023).

Conclusions

Based on the results of this survey, unwanted sexual behaviors are a perversive and troubling phenomenon that causes adverse psychological and long-lasting effects on victims as well as witnesses using the rail network. While most victims and witnesses do not report incidents to the authorities, mainly because they do not feel they would be taken seriously, those who do file a report are generally satisfied with the response they receive. Nevertheless, we now know, as a result of this study, that there are issues in terms of the feedback and information victims receive and the emotional support they receive during and after the reporting. They expect the police to call them back after they have complained, preferably multiple times, even though their cases do not progress further in terms of further investigation and prosecution. A trauma-focused approach is needed to assist victims, while USB hot spot policing can potentially be used to prevent USB victimization.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15564886.2024.2330094.

Notes

1. This intentional omission of people who have not used rail services at least once over the previous 12 months of the survey carries a limitation. Our approach ensures that the population of interest is at least occasional passengers of rail services. However, we thus excluded people who were victimised or experienced USB beyond the immediate past 12 months and never used the rail services after the incident. Based on our data, we can only speculate what this group of victims experienced. It is likely that at least some of the events or symptoms reported by our study participants would exist at similar or even elevated levels for victims who never returned to ride trains in Great Britain. Evidence is warranted on this sub-group of USB victims.

1. At the individual level, numerous studies have highlighted the correlation between frequent engagement in USBs and distinct personal traits and behaviors. There is likely a strong link between USB behaviors and characteristics such as harboring negative views toward women, accepting interpersonal aggression, and displaying tendencies toward dominance and control. In addition, scholarly investigations have identified societal and cultural norms as major factors contributing to the perpetuation of gender inequality and the tolerance of violence against women. These norms can significantly influence an individual’s propensity to engage in acts of harassment (Brown & Messman-Moore, 2010).

At the societal level, the continued prevalence of USBs on public transportation can be attributed to many broader factors, including entrenched patriarchal systems, harmful gender stereotypes and the trivialization of these incidents. As cited by Hershcovis et al. (2021), cultural narratives that downplay the seriousness of harassment or lay blame on the victims can dissuade people from speaking up and perpetuate a culture of silence.

2. We note that the reasons for the inverse relationship between the risk of USB in the public domain and age are complex and linked to a wide range of factors, including the amount of time potential victims spend outdoors, in nighttime economy facilities, and offenders’ choices. See e.g., Uggen and Shinohara (2009), Ceccato and Loukaitou-Sideris (2021),