Bystander behavior in workplace bullying: a vignette study exploring how organizational space and situational strength influence intentions to intervene

The role of bystanders in workplace bullying has gained increasing attention over the last decade (Coyne et al., 2019; Hodgins et al., 2020; Jönsson and Muhonen, 2022; Rosander and Nielsen, 2023; Pouwelse et al., 2021). The bystander is an important actor with the potential to act in a bullying situation and can be described as the “third” actor, next to the perpetrator and the victim (Bowes-Sperry and O'Leary-Kelly, 2005). While not being directly involved, the bystander can intervene and thus, mitigate and potentially prevent harmful situations from occurring (Darley and Latané, 1968).

Research indicates that bystanders may respond and act in different ways as a response to workplace bullying. Bystander behaviors can be categorized according to how active or passive they are and if the behaviors are constructive or destructive (Ng et al., 2020; Paull et al., 2012; Thompson et al., 2020). This results in four broader categories: active constructive (e.g. trying to stop bullying); active destructive (e.g. supporting the perpetrator); passive constructive (showing empathy toward the victim but not trying to intervene); and passive destructive (e.g, doing nothing) (Coyne et al., 2019; Mulder et al., 2016; Ng et al., 2020; Paull et al., 2012; Thompson et al., 2020).

The way a bystander intervenes in the context of workplace bullying can be explained in different ways. In some prior studies, the research focus has been on the traits of the individual and their relationship with the target (Hellemans et al., 2017). In other cases, bystander behavior is understood in relation to the work environment and organizational setting (Hershcovis et al., 2017; Jönsson and Muhonen, 2022). For example, a previous study on bystander behavior in the healthcare sector showed that socialization processes, dysfunctional organizational culture, culture of silence, deficiencies in management, and fear of becoming the next target influence bystander actions (Jönsson and Muhonen, 2022).

Against this background, the current study explored whether and how spatial aspects influence bystanders’ intention to intervene, using the concept of “situational strength” to analyze different spatial contexts.

Workplace bullying was initially described and investigated as acts that occur in the physical work environment. Previous research indicates that a poor physical work environment is associated with higher reporting of both being a target and a witness of bullying (Salin, 2015). However, over the last 15–20 years, the concept of space has become relevant in a different way, especially vis-à-vis cyberbullying and online harassment (Farley et al., 2024; Forssell, 2020; Platts et al., 2023).

Space as a concept can be helpful in understanding workplace bullying and bystander intervention. For example, studies on school bullying have shown that space is highly relevant since bullying tends to occur in unstructured spaces (Craig et al., 2000; Leff et al., 2003), unsupervised spaces (Atlas and Pepler, 1998), and undefined spaces (Astor et al., 2001). Researchers have also identified hotspots or locations for bullying in schools, such as playgrounds, lunchrooms, hallways, stairwells, and bathrooms (Andrews and Chen, 2006; Astor et al., 2001; Cowie and Oztug, 2008; Craig et al., 2000; Fram and Dickmann, 2012; Leff et al., 2003; Migliaccio et al., 2017; Vaillancourt et al., 2010).

According to Kitchin and Dodge (2011), space can be considered an active element in the production of social relations that provides a defining context for social interaction, as spatial contexts are produced through social relations. This means that the workplace is not only a physical place but also a place that is given meaning by people’s practices, language, behavior, and memories, which in turn shape social relations (e.g. Kitchin and Dodge, 2011; Cowen Forssell, 2019). In the same way as work organizations are carriers of conventions of behavior, norms, and values that may influence social behaviors and interactions between organizational members, spaces can be viewed as embedded in context-specific norms for social interaction.

It is widely assumed that human behavior occurs due to combinations of individual differences (e.g., personality, intelligence, and experience) and situational characteristics surrounding each person (Cronbach, 1957). Situational strength is defined as “implicit or explicit cues provided by external entities regarding the desirability of potential behaviors. Situational strength is posited to result in psychological pressure on the individual to engage in and/or refrain from particular courses of action; this pressure in turn is posited to reduce relevant behavioral variance and attenuate subsequent trait–outcome relationships” (Meyer et al., 2010, p. 122).

Meyer et al. (2010) identified four dimensions of situational strength, as follows. Clarity refers to the extent to which cues regarding work requirements are clear and understandable. Consistency refers to the extent to which work-related cues are consistent over time and across sources. Constraints refer to the extent to which external forces limit individuals’ freedom of decision and action. Consequences refer to the extent to which the outcomes of actions have significant effects. Clear consequences encourage behaviors that lead to positive outcomes and discourage negative outcomes. Situational cues, provided by external entities, influence behavior. Strong situations, with clear and consistent cues, lead individuals to act similarly. In weak situations, where cues are inadequate or inconsistent, individual differences become more apparent (Dalal and Meyer, 2012). Strong situations lead to behavioral homogeneity, whereas weak situations highlight personality differences (Meyer et al., 2010).

In the case of workplace bullying and bystander intervention, one can argue that situational strength can either empower or inhibit bystanders in workplace bullying scenarios. In strong situations characterized by clarity, consistency, constraints, and clear consequences, bystanders are more likely to recognize bullying, feel responsible for addressing it, and have the support and motivation to act. Conversely, in weak situations, where these elements are lacking, bystanders may feel uncertain, unsupported, and less compelled to intervene, allowing bullying to persist.

With regard to preventing bullying in the workplace, the behavior of bystanders and the interventions of other people are of great importance (D'Cruz and Noronha, 2011; Pouwelse et al., 2021). For example, focusing on bystanders’ actions can be beneficial when designing methods or interventions to reduce the negative consequences of bullying. This can be achieved, for example, by actively promoting a positive social climate in the workplace or by encouraging social support and constructive behaviors in different situations (Einarsen et al., 2020).

The aim of this study was to explore whether and how spatial aspects influence bystanders’ intention to intervene. Specifically, the study analyzed related organizational and spatial contexts in the healthcare sector. Accordingly, this study presents a more nuanced picture of bystanders’ intention to intervene in the context of workplace bullying.

This study focused on the healthcare sector, as findings from earlier studies indicate that workplace bullying is more prevalent in the healthcare sector compared to other sectors (Zapf et al., 2020). According to a recent systematic review of bullying among healthcare employees, up to one in four healthcare professionals is exposed to regular bullying (Lever et al., 2019). As the healthcare sector has high turnover rates and shortage of staff (Jönsson et al., 2021), it is important to highlight aspects that could contribute to improving the work environment and staff retention.

This study employed vignette-based interviews. Vignettes have been used in many ways in prior studies, depending on the discipline, research approach (qualitative/quantitative), and topic of interest (e.g. Hughes, 1998; Barter and Renold, 2000; Jenkins et al., 2010; Jackson et al., 2015; Edgell et al., 2016). In qualitative studies, vignettes can be conceptualized as constructions that evoke and materialize the phenomenon under study, placing it in concrete contexts and allowing the researcher to explore participants’ views, feelings, experiences, interpretative resources, norms, and voices that are related to or arise from the situations (O’Dell et al., 2012).

The participants consisted of 24 staff members from different occupations in the healthcare sector (see Table 1). The participants were employed at one of the Regions, organizations responsible for providing public healthcare in Sweden. In a previous questionnaire-based study also about workplace bullying, the participants had expressed interest to participate in a follow-up interview study focusing on bystanders of workplace bullying. Most participants were women (75%), and their mean age was 47.42 years (standard deviation = 11.09; range: 28–70 years). The participants worked in different medical fields, including radiology, psychiatry, pediatrics, orthopedics, anesthesia, emergency care, palliative care, surgery, primary care, gynecology, and internal medicine. They had varied experiences with bullying. Some had experience as either a target or a bystander, while others had experience in both roles.

The data were collected individually by the two authors between March 2019 and September 2020. The duration of the interviews ranged between 40 and 95 min, and they were conducted when the participants were at their workplaces, via Zoom/Teams or phone (owing to the COVID-19 pandemic).

The first part of the interview focused on the participants’ personal experiences of workplace bullying (either as a target, bystander, or both) and how they acted during the bullying situation. This data has already been published, so we will not explore these aspects further (Jönsson and Muhonen, 2022). In the second part of the interview, we introduced the vignettes, reading them aloud to the participants. The reason for closing the interviews with the vignettes was to broaden the focus from personal experiences to more abstract and general incidents (Barter and Renold, 2000). This can be particularly useful when highly sensitive topics (e.g. bullying) are discussed (Rahman, 1996). After presenting each vignette, we asked participants how they thought they would react if they witnessed the situation described in the vignette. The recorded interviews were transcribed verbatim.

In total, we used five vignettes focusing on negative acts related to workplace bullying, with different workplace bullying behaviors and various spatial contexts where the bullying occurred (see Table 2). The first two vignettes (No. 1: “Gossiping” and No. 2: “Physical violence”) were from Longo and DeDonno’s (2018) study, who based the vignettes on reported workplace bullying situations in the nursing context. They consulted a panel of experts to test the content validity of the vignettes. Vignettes No. 3 (“A reprimanding email”) and No. 4 (“Embarrassing photo on Facebook) were developed by Madden and Loh (2020) based on existing literature and research. These two vignettes focus on cyberbullying and were not originally used in a nursing context but in a study involving various white-collar professions. The last vignette, No. 5: “Undermining persons’ reputation,” published by Katrinli et al. (2010), was developed by experts in organizational behavior based on Leymann’s classification of bullying behaviors applied in the nursing context. The vignette exemplifies effects on the victim’s possibilities to maintain their personal and professional reputation. The content was validated by five experts in industrial psychology.

We translated the vignettes described above and modified the names in the vignettes to suit the Swedish context. Furthermore, the wordings in vignettes No. 3 and No. 4 were modified to suit the healthcare context. While all the original vignettes focused on the reader/participant as the victim of bullying, in this study, the participants were asked to take the role of being a witness to the bullying situations. The vignettes were read in the same order for all participants, starting with the less severe situation (No. 1) and ending with a severe situation where bullying escalated and the target was looking for a new job (No. 5). The type of spatial context of the vignettes can be classified as occurring in the physical environment or online and is related to either a professional or a semi-professional space. The vignettes were categorized based on the four facets of situational strength: Clarity, Consistency, Constraints, and Consequences. This resulted in three types of situational strength: weak, semi-strong, and strong (see Table 2). The categorization was independently conducted by the two authors, and no direct disagreements were found.

The focus of the analysis was to identify how the participants described their intentions to act as bystanders in the different spatial contexts and the reasoning behind their behavior. The transcribed data were analyzed following qualitative thematic analysis (Braun and Clarke, 2006, 2023). Each author read through the verbatim transcriptions several times while identifying the intended bystander behaviors that emerged from the material. All bystander behaviors were coded using initial codes, after which the authors discussed and revised the codes when necessary. As a last step the codes were organized into categories based on Paulls et al.’s taxonomy (2012). This taxonomy describes that some roles have positive outcomes for the victim and may therefore help reduce bullying or its impacts in organizations, i.e. constructive (for example defending, sympathizing, intervening etc.), while others are negative, i.e. destructive (for example avoiding, abdicating, collaborating, manipulating etc.). Whether the role is active or passive contributes to the extent to which the role may lead to positive outcomes. The roles and actions or inactions of bystanders, whether active or passive, will contribute to the escalation or de-escalation of bullying. In this study, the codes were organized into three broad categories – active constructive, passive constructive, and passive destructive. No active destructive behavior (i.e. supporting or encouraging bullying) was identified in the material (see Table 3).

Before participating, all respondents were presented with a consent form that included information about the study. Participants provided active consent to participate and were informed that the study design was longitudinal. Ethical approval was obtained from the Swedish Regional Ethical Review Board (reference no. 2018/385).

In this section, we describe each vignette separately and briefly describe how the context is related to spatial aspects and situational strength. Next, an analysis of the empirical material is presented with a focus on bystanders’ intentions to act.

The situation described in Vignette No. 1 is taking place in the nurses’ lounge in a physical environment that can be described as a semi-professional space, where the boundaries between the private and professional are blurred. A lunchroom can function both as a room where staff detach from work for a moment and talk about more private things, and also an area where they may talk and reflect upon their jobs, talk to colleagues about a patient, or discuss staffing and scheduling. As the bullying behavior occurs in a semi-professional setting without clear codes of conduct and in a weak situation (i.e., without clear signals), it can be more difficult for bystanders to know how to act. In this setting, the participants described behaviors that could be labeled as active constructive, passive constructive, or passive destructive, as described below.

From the analyses, it became evident that participants who had a managerial position currently or previously were more prone to engage in active constructive behavior, by intervening directly and confronting bullies, for example, by pointing out that gossiping was not an appropriate behavior at the workplace. They considered it important to intervene in order to act proactively; otherwise, there was a risk that the situation could escalate and get out of their hands.

Others stated that they would intervene in a subtle way, which could be classified as active constructive behavior, but in a more indirect sense. This was expressed as, for example, talking to the bullies afterward, not confronting them in an aggressive way but rather highlighting their feelings as bystanders about the situation. One resourceful way of confronting the bullies was when the witness said something positive about the target or praised her in front of the bullies. Positive comments could defuse negative ones, and in this way, the gossiping could be stopped from escalating further.

Some participants stated that they would engage in passive constructive behavior by showing social support for the target, especially if they were close colleagues. Others said that they would turn to their managers, as it was difficult to intervene with colleagues. They also reflected that it is difficult for the witness to confront subtle behaviors, such as gossiping and rolling eyes.

It was quite common for participants to say that they would act in a passive, destructive way, that is, pretending not to have heard, walking away, and doing nothing, as they would not like to get involved in the situation. Some would greet the target when she came in, but they would not tell her that the bullies/perpetrators were gossiping about her.

A few respondents shared that they wished they had the courage to intervene, but thought that it might be too risky to do so. If the witness confronts the bullies, they risk “losing their faces,” and that in turn could have negative consequences for the witness, for example, turning them to a target instead.

Vignette No. 2 also presents an incident that occurs in a physical environment, but in a professional setting—a meeting room. Compared to the lounge area, the meeting room has a clearer code of conduct, usually having a person to chair the meeting, an agenda, a fixed timeframe, and expectations of proper respectful behavior. Thus, there are several cues indicating strong situational strength. In the context of workplace bullying, this implies that it is easier to intervene as a bystander when a person breaks the rules by not respecting others, turning to talk, or becoming violent. At the same time, bystanders may be passive if they feel it is the chairperson’s responsibility to act. In this context, the participants described both constructive and destructive actions.

All the participants pointed out that when a situation evolves and becomes physical, as described in the vignette, those who witness it must intervene. Their intentions to act can be described as active constructive behaviors, either acting directly themselves to stop the situation or by involving the manager.

Those with managerial experience stated that they would intervene directly, for example, send the bully home, contact the HR department, ask them both to come to the manager’s office, and explain the code of conduct. Physical threats and violence are serious assaults that can lead to the perpetrator being fired.

It was clear that this kind of behavior was totally unacceptable and contravening the professional code of conduct—“we are here for the patients.” Everybody is entitled to express their opinions and complete sharing their perspectives without being interrupted. Even non-managers thought that they would intervene by trying to separate the persons, asking them to calm down, initiating discussion/dialogue, and stopping the bully earlier by saying that they wanted to hear what the target was saying. It was also mentioned that it was safer to intervene in meetings, as there is a clear code of conduct.

Some participants believed that the managers should intervene, and if the manager was chairing the meeting, they should have intervened earlier, before the situation escalated.

This could be regarded as passive destructive behavior in the sense that responsibility was transferred solely to the manager, rather than taking action as a colleague.

In Vignette No. 3, the bullying occurs via cyberspace, involving the professional space. Here, the code of conduct, stipulating acceptable ways of communicating and behaving, becomes blurred. What is considered professional in cyberspace can differ among colleagues and managers if not explicitly discussed and expressed in policies. Because the content is related to work, the situation is defined as professional, while the situational strength can be labeled as semi-strong, with both low (clarity, consequences) and semi-strong (consistency, constraints) cues. In this context, we identified different types of constructive bystander behaviors.

This vignette describes a cyberbullying situation, and several participants stated that they would support the target in different ways. Many participants mentioned that they would try to support the target both emotionally by showing empathy and by giving concrete advice and information, for example, scheduling a meeting with the manager to have a dialogue and discuss the situation. Some even expressed that they would offer to accompany the target to meet the manager.

Some said they would encourage the target to contact a manager higher up in the hierarchy, as this is an unacceptable situation. Others mentioned they would try to discuss the situation at a group level, as everybody can make mistakes, and it was considered better to talk face-to-face than to send emails. For instance, according to a respondent, one could intervene in a subtle way by asking the manger, “Are you aware that you sent this e-mail to everyone?” (Participant 3).

As it might be difficult to intervene when the perpetrator is one’s manager, some participants would recommend the target to contact respective labor unions and refer to the organizational policies for guidance.

Several participants reflected on the manager’s behavior, which was considered unprofessional and contravening the code of conduct as a manager in many ways. The manager should have acted at an early stage if there were problems in the project, which should have been communicated in person. Likewise, negative feedback should be given privately, face-to-face, and not in a public space, such as a group email.

Vignette No. 4 presents an incident that occurs in an online setting, a semi-professional space. Although this happens in a collegial context, the image is posted on a private Facebook account, making it difficult for bystanders to change the situation. We categorized this vignette as a weak situation because there are no clear cues to guide actions. Several participants responded that they were not active on FB and, therefore, had and would not encounter this kind of situation. However, it was not unusual to have an FB account for work, but it was mainly used for planning joint activities, birthdays, and so on. The participants described different types of constructive behaviors.

Others would intervene, tell the bully off, point out that it was not acceptable to post these kinds of pictures of colleagues on a private FB page, ask them to remove the pictures, and apologize for posting such content. Some considered it illegal to post pictures like this, and they would ask the bully to delete the picture or report them.

Others reflected upon the somewhat blurred boundaries between private and professional life in relation to social media such as FB; for example, whether the situation described in the vignette could be considered as workplace bullying. It could be difficult to act as a manager when the pictures are captured outside work hours and posted on a personal FB account, even though the manager could ask the person to remove them.

A common behavior the participants mentioned was showing social support for the target; for example, talking with her, giving advice, and discussing what happened at the party. The target could also be advised to contact FB to remove the pictures. While one participant mentioned they would not act face to face but instead post a supportive comment on FB, another participant considered it difficult to comment on FB, as there was a risk that you could become a target yourself. In addition, it can be difficult to comment on FB, because this confirms that one has seen the picture and gotten involved, as the bully wants.

Vignette No. 5 describes yet another situation taking place in a physical environment but undermines the target’s reputation in both private and professional spaces. The situation is similar to the first vignette related to gossiping, but with a longer duration, presumably occurring in different spaces in the workplace. This situation is considered professional because gossiping occurs in the workplace. As the situation escalated over time, there was a lack of clear cues during the process, indicating weak situational strength. The participants mainly described constructive actions.

Some would confront the bully by trying to point out that the accusations were not true and emphasize that colleagues should not interfere with each other’s private lives. Others suggested intervening at a group level without pointing out the perpetrator, so that the colleagues could discuss as a group that talking behind someone’s back was not acceptable behavior. Pointing out the bully or confronting her could be counterproductive, as this could worsen the target’s situation. There is also the risk that the bystander will become a target.

Several participants, including the managers, thought it was difficult to act when the situation escalated over time, as described in the vignette. The witnesses may only be partially aware of what is going on, as neither colleagues nor managers are around all the time. Thus, some would advise the target to leave and seek another job, which should be easy considering the general lack of nurses.

Similar to the responses concerning the vignettes discussed above, some participants highlighted the managers’ responsibility in handling the situation. It was considered important to confront the bully at an early stage; if that did not help, the next step was to relocate the bully. The manager needed to be made aware of the situation and that there was a high risk of a competent coworker leaving the job. It was also mentioned that a lack of knowledge on how to act as a bystander in these types of situations prevented them from acting. It was considered important to distinguish relationships as colleagues and friends, and that it is unacceptable to make fun of colleagues and their private lives.

This vignette-based study explored bystanders’ intentions to intervene during workplace bullying occurring across different spatial contexts. The vignettes presented scenarios involving various types of spaces, including physical and online environments, semi-professional and professional settings, and situations where the situational strength ranged from weak to semi-strong and strong. Although we did not explicitly ask if they considered the vignettes to be bullying situations, we perceived that the participants found the vignettes to be realistic and relatable.

The analysis of the interview data identified three main categories of bystander behavior: active constructive, passive constructive, and passive destructive, consistent with the taxonomy developed by Paull et al. (2012). Upon closer examination, it became evident that most of the behaviors described in response to the vignettes were constructive. The fact that none of the respondents described any active destructive behavior might be explained by social desirability (people’s tendency to give responses they believe will be viewed favorably by others).

In most cases, participants indicated that they would respond constructively, either by taking action themselves or by informing a manager. This is in line with previous research describing different types of constructive actions by bystanders (Pouwelse et al., 2021). Additionally, we observed instances where bystanders supported the victim afterward or confronted the perpetrator. In situations where destructive behavior was noted, some respondents reflected on their reluctance to take an active constructive role, fearing that doing so might make them targets of bullying. This is something that was foremost hightlighted by the participants who had previous experience of being a target of bullying. However, recent research by Nielsen et al. (2021) suggests otherwise—actively intervening bystanders did not have an increased risk of becoming victims of bullying, whereas inactive bystanders were found to have a three-fold risk of becoming victims at follow-up (Rosander and Nielsen, 2023). Nonetheless, there is a common perception that intervening as a bystander is risky, although this may not be true. Additionally, the consequences of being an active bystander in professions such as nursing or healthcare services may differ from those in the general workforce, as reported by Rosander and Nielsen (2023). Participants also noted that they were more likely to intervene if they had a personal connection with the victim or if they witnessed similar incidents multiple times.

When analyzing the empirical data through the lens of the situational strength theory (Meyer et al., 2010), no clear patterns emerged. According to this theory, situations with strong situational potency are expected to elicit uniform behavior among individuals. However, this was not observed in our study. Table 4 illustrates the different situations and types of intervention intentions identified in the empirical material.

For example, in response to Vignette 1, which was categorized as a weak situation, a wide variety of behaviors, both passive and constructive, were described. This indicates that in less defined environments, individuals may respond in diverse ways, reflecting the lack of strong situational cues. Specifically, Vignette 1, set in a physical environment and semi-professional context, elicited active constructive (AC), passive constructive (PC), and passive destructive (PD) behaviors. Conversely, in Vignette 4, which was also categorized as weak, only constructive behaviors were reported. This vignette, set in an online environment, resulted in active constructive (AC) and passive constructive (PC) behaviors. Despite being a weak situation, the context seemed to inherently promote constructive responses, suggesting that other factors within the spatial context may influence behavior. In Vignette 2, categorized as a strong situation within a physical and professional context, behaviors were more focused on constructive actions, though some passive destructive behaviors were also noted. This vignette elicited active constructive (AC) and passive destructive (PD) behaviors, indicating that while there are some guiding cues, they are not strong enough to completely eliminate destructive responses. Vignette 3, categorized as a semi-strong situation in an online and professional context, predominantly resulted in constructive behaviors, both active and passive. This aligns with the theory that strong situational cues lead to more uniform and positive behaviors, as evidenced by the active constructive (AC) and passive constructive (PC) responses. Finally, Vignette 5, set in a physical and professional context and categorized as weak, also resulted in constructive behaviors. This vignette elicited active constructive (AC) and passive constructive (PC) behaviors, further suggesting that the nature of the spatial context can significantly influence the type of bystander intentions, regardless of the situational strength.

The analysis of Table 4 reveals that the situational strength theory did not consistently predict uniform behavior across different contexts. The situational strength theory suggests that strong situations, characterized by clear norms and expectations, typically lead to uniform behavior (Meyer et al., 2010). However, if the vignettes were not perceived uniformly by all participants, variability in their responses could be expected. The variability in responses across different vignettes can be attributed to differences in how participants perceived the situational cues. Ambiguity or uncertainty within the vignettes weakens situational strength, allowing individual differences to play a larger role in determining behavior. In addition, even in situations with strong situational cues, personal factors such as values, beliefs, and past experiences may influence how individuals interpret and react to these cues. Thus, the interaction between personality traits and situational strength may also play a role. While strong situations are expected to limit the influence of personality, in certain contexts, personality traits such as resilience, assertiveness, and contrarian tendencies can still moderate behavior.

Even though the participants in this study were asked to take the role of witnesses to the vignettes/bullying situations, we know that they had varied experiences with bullying. Some had experience as either a target or a bystander, while others had experience in both roles. These experiences might influence how individuals perceive and respond to different situational cues.

For example, someone who has been a target of bullying might be more sensitive to weak situational cues and may exhibit a wider range of behaviors, including passive destructive (PD) behaviors, due to heightened emotional responses. Conversely, those with bystander experiences might be more inclined to exhibit passive constructive (PC) behaviors, reflecting their previous role of observing and possibly intervening in bullying situations.

The absence of active destructive behaviors in participants’ descriptions can largely be explained by social desirability bias, where individuals aim to present themselves in a positive light, avoid negative judgments, and align themselves with the societal norms of morality and ethics (Nederhof, 1985). This bias may lead to an overrepresentation of constructive behaviors and omission or downplaying of destructive actions, reflecting participants’ desire to maintain a positive self-image and avoid moral discomfort.

In this study, the sample size was quite small (N = 24). This is a limitation in several ways. First, it reduces the generalizability of the findings. With fewer participants, the results may not adequately represent the broader population or encompass a diversity of perspectives, experiences, and behaviors (Carminati, 2018). Second, the limited sample makes it more challenging to identify consistent patterns and increases the risk of drawing conclusions influenced by the specific individuals interviewed. Furthermore, small sample sizes can hinder the achievement of theoretical saturation, where no new themes or insights emerge, potentially leaving significant aspects of the phenomenon underexplored.

In this study, we used vignettes to explore intentions to intervene as a bystander of workplace bullying. Vignettes are useful for examining intention and type of action but in future studies it could be important to assess how participants view the vignette (in relation to concepts, contexts etc.). In addition, in this study, only one of the vignettes was strong (in relation to situational strengths while three were week and one was semi-strong, this is something that can be more balanced in the future providing a more rigorous test of the situational strength.

For future studies, it could also be valuable to focus on how bystanders act in real-life bullying situations that take place in different context. The physical and social context of a place can affect whether bystanders feel empowered to intervene or remain passive. Here, it would be interesting to focus on the involvement of the bystander in relation to time. This aspect has been highlighted by Bowes-Sperry and O'Leary-Kelly (2005) in relation to sexual harassment. They developed a model where the bystander intervention is categorized along two dimensions: (1) immediacy of intervention and (2) level of involvement.

In addition, future research should explore several intriguing areas based on the findings of the present study. One potential direction is to investigate the conditions under which bystanders might engage in active destructive behaviors. This could involve designing scenarios that minimize the social desirability bias to obtain a more accurate understanding of such behaviors. Additionally, conducting longitudinal studies could help track the long-term effects of bystander interventions on both bystanders and victims, shedding light on the sustained impact of different types of interventions.

Regarding the spatial context, the unique challenges and strategies for bystander intervention in online environments also warrant further study. Research could examine the role of anonymity, speed of information dissemination, and effectiveness of different types of online interventions.

Finally, investigating how organizational support and policies influence bystander behavior could be beneficial. Studying the impact of clear anti-bullying policies, reporting mechanisms, and support systems on the likelihood of bystander intervention could inform the development of more effective organizational strategies to combat workplace bullying.

To effectively implement proactive measures against workplace bullying, it is crucial to raise awareness of how different workplace environments/spaces can influence both bullying and bystander behavior. Although no clear patterns were observed, the four facets of situational strength offer valuable insights into this process.

Firstly, clarity is essential. Well-defined policies and procedures on acceptable behavior and bullying can make it more likely that bystanders will recognize bullying and know how to respond (Nielsen et al., 2024). For instance, creating a detailed anti-bullying policy that includes definitions, examples of bullying behaviors, and the steps employees should take if they witness or experience bullying can be very effective. Additionally, conducting regular training sessions to educate employees about the policy and procedures, such as through role-playing scenarios, can help bystanders practice how to respond to bullying (Kuntz and Searle, 2023).

Consistency in enforcing anti-bullying rules across all levels of the organization is another critical factor (Tuckey et al., 2022). Ensuring that these rules are uniformly implemented can boost bystanders’ confidence in intervening, knowing that their actions will be taken seriously. For example, if a manager witnesses bullying and takes immediate action, it sets a precedent that such behavior will not be tolerated, thereby encouraging bystanders to report incidents. Regular audits to check for compliance with anti-bullying policies, including anonymous surveys to gather feedback from employees about their experiences and observations, can further reinforce this consistency (Einarsen et al., 2020).

Constraints refer to external pressures such as strict rules and supervision that limit inappropriate behaviors like bullying. In such environments, bystanders may feel more compelled to act, thereby reducing their passive bystanding. Implementing strict rules and close supervision, such as a zero-tolerance policy for bullying and ensuring that managers are trained to monitor and address any signs of bullying promptly, can create a more secure environment. Providing clear and accessible reporting channels, like a dedicated email address for reporting incidents, ensures that employees feel safe and supported in coming forward.

Consequences include the outcomes of addressing or ignoring bullying. Bystanders who understand the potentially positive or negative consequences of their actions are more likely to intervene. For instance, highlighting the positive outcomes of intervening, such as creating a safer and more supportive work environment, and the negative consequences of ignoring bullying, like a toxic workplace culture and potential legal ramifications, can motivate bystanders to take action (Pouwelse et al., 2021).

Managers should consider these factors when working with staff to establish a workplace code of conduct that covers all relevant spaces, including offices, break rooms, and virtual environments. Promoting an inclusive and respectful workplace culture through team-building activities, diversity training, and open communication channels can further support these efforts. Additionally, offering resources such as counseling services, employee assistance programs, and support groups for those affected by bullying ensures that employees know how to access help when needed.

This work was supported by the Swedish Research Council for Health, Working Life and Welfare (FORTE) under Grant 2018-00228.