Personality as a risk factor for workplace bullying must be considered carefully, taking into account not only victim characteristics but also how perpetrators perceive them and how organizational shortcomings in handling and preventing bullying allow such behaviour to emerge. Victim precipitation theory (Elias, 1986) suggests that certain traits or behaviours may inadvertently trigger mistreatment, though this idea is controversial as it can imply victim blame. Originally, it referred to situations where victims more clearly initiated conflict (Wolfgang, 1957), but later applications emphasize how perceived provocation or vulnerability can shape victimization.

From a social interactionist perspective (Tedeschi and Felson, 1994), personality may influence how individuals are perceived and treated in social contexts. Traits such as high conscientiousness or neuroticism may lead others to view a person as demanding or difficult, thereby increasing the likelihood of negative reactions (Aquino and Thau, 2009; Tepper et al., 2006). In this view, aggression is purposeful behaviour used to enforce social or performance norms. When such reactions are not managed, they can escalate into bullying. Thus, it is not personality itself but how certain traits are interpreted and responded to within the workplace that creates risk.

Building on school bullying research (Olweus, 1993), Zapf and Einarsen (2020) distinguish two personality-linked pathways: the vulnerable victim, characterised by insecurity and high neuroticism, and the provocative victim, marked by irritability, low agreeableness, or aggression. Vulnerable individuals may invite predatory bullying (Einarsen, 1999), whereas provocative victims may display behaviours that provoke negative reactions or mutual hostility, potentially setting in motion a cycle of escalating conflict (Baillien et al., 2016)—that is, dispute-related bullying (Einarsen, 1999). Taken together, personality contributes to the risk of bullying through its social meaning rather than inherent traits. When personal characteristics are misinterpreted or not managed properly in organizational contexts characterized by, for example, weak leadership and poor conflict management, they may become catalysts for persistent mistreatment.

Although personality is regarded as a relatively stable property of adults (Roberts, 2009), it may change through self-regulatory processes across the lifespan (Hennecke et al., 2014). Such changes may arise from experiences of discrepancies between one’s internal state and external standards, originating, for example, from social or environmental conditions (McCrae and Costa, 2008), including adverse or traumatic life events. Theoretically, exposure to traumatic events may shatter basic assumptions about the world as benevolent and meaningful (Janoff-Bulman, 1992), leading to enduring changes in how individuals relate to others. Regarding the Five-Factor Model, traumatic experiences have been particularly associated with increased neuroticism and decreased agreeableness and conscientiousness (Bucciol et al., 2024; Löckenhoff et al., 2009). Hence, it seems reasonable to assume that exposure to workplace bullying could lead to changes in personality traits, either temporary or more permanent.

While traumatic events represent one possible pathway to change, the TESSERA framework (Triggering situations, Expectancy, States/State expressions, and ReActions; Wrzus and Roberts, 2017) highlights another: personality development as the outcome of repeated sequences of everyday experiences. Building on state–trait research within a lifespan perspective (e.g., Fleeson, 2001; Roberts et al., 2008), the framework proposes that each sequence—beginning with a situational trigger and followed by expectations, state expressions, and reactions—contributes incrementally to long-term trait change when repeated over time. As workplace bullying represents a form of repeated and prolonged exposure, it can be seen as a context in which such negative sequences are frequently activated: recurrent exposure to mistreatment creates expectations of hostility, evokes states of anxiety or defensiveness, and leads to reactions of withdrawal or conflict. When accumulated over time, these repeated sequences may gradually shift an individual’s personality profile—for instance, increasing neuroticism or reducing extraversion (Farley et al., 2025).

Summarizing cross-sectional studies by means of meta-analysis, Nielsen et al. (2017) found that exposure to workplace bullying was significantly associated with extraversion, neuroticism, agreeableness, and conscientiousness, but not openness to experience. Bullying was most strongly associated with neuroticism across the included studies. While there does not appear to be a clear and consistent victim profile (Glasø et al., 2007), there are indications that targets may differ from those not exposed, showing higher scores on neuroticism and lower agreeableness and conscientiousness (Glasø et al., 2009; Matthiesen and Einarsen, 2001). However, these studies do not clarify whether specific personality traits increase the likelihood of being singled out as a target, or whether the differences in trait levels among those bullied are the result of exposure to bullying—that is, whether they should be regarded as outcomes of the bullying process or as pre-existing characteristics constituting risk factors for bullying. This distinction is central to understanding whether personality functions primarily as a cause or a consequence of bullying, and shedding light on the direction of the association requires longitudinal evidence.

However, to date, longitudinal evidence remains scarce, and the findings are inconclusive. For example, when examining all Five-Factor Model traits, Nielsen and Knardahl (2015) found support for neuroticism as an antecedent of bullying, but when controlling for role conflict and role ambiguity—both well-established risk factors for bullying (den Van Brande et al., 2016)—the association disappeared, leaving only conscientiousness as a predictor of bullying across the two-year study period. In the same study, bullying predicted changes in agreeableness, conscientiousness, and openness over time. Persson et al. (2016) examined only neuroticism and extraversion and found that becoming bullied increased neuroticism, while no longer being bullied decreased neuroticism and increased extraversion. In a small study within a single private corporation, Podsiadly and Gamian-Wilk (2017) found that exposure to bullying behaviours predicted a decrease in agreeableness over time. In a large-scale four-year study, Farley et al. (2025) investigated personality change as an outcome of exposure to bullying and found that exposure was related to increased neuroticism and decreased extraversion, and that changes in bullying status were also associated with decreased conscientiousness. There are studies that have focused on specific aspects of personality traits in relation to the risk of bullying. Reknes et al. (2021) found that trait anxiety and trait anger affected the development of bullying over a six-month period. Zahlquist et al. (2022) investigated the same traits as potential links between conflict involvement and concurrent exposure to bullying behaviours over a 30-day period. They found that only trait anger moderated the association—that is, high conflict involvement combined with high trait anger was associated with a higher level of exposure to bullying behaviours.

As findings diverge across studies, the reciprocal effects between personality and bullying remain uncertain. Moreover, previous research has employed different methodological strategies and operationalised bullying in diverse ways. Hence, there is a need for further large-scale studies that employ longitudinal designs and consistent methods for measuring bullying (Nielsen and Knardahl, 2015). In response to this call, the present study addresses the following research questions using time-lagged data from a large national probability sample of Swedish employees.

In the second part of the study, we build on the longitudinal associations identified in Part 1 by examining potential mechanisms that may account for these relations.

The sample consisted of 58% women and 42% men, with a mean age of 49.44 years (SD = 10.50). On average, participants had worked at their current workplace for 10.11 years (SD = 9.62), and 96% held a permanent position. A clear majority (89%) were born in Sweden. In terms of educational attainment, 16% had completed compulsory education, 33% had graduated from upper secondary school, and 51% held a university degree.

Workplace bullying was assessed using the Negative Acts Questionnaire–Revised (NAQ–R; Einarsen et al., 2009), validated for use in a Swedish context (Rosander et al., 2024). The instrument includes 22 items describing negative and unwanted behaviours that employees may experience at work. When such acts occur repeatedly and systematically over time, they may constitute bullying. Responses were recorded on a five-point frequency scale ranging from never to daily. Internal consistency was high (Cronbach’s ⍺ = 0.89 and McDonald’s ω = 0.91 at both T1 and T2). At T1, those who reported being exposed to at least some bullying behaviours were asked how long the exposure had been ongoing, with a six-point frequency scale ranging from recently to more than 2 years.

Self-labelled bullying was also assessed. First, respondents were presented with the following definition: “Bullying occurs when a person, repeatedly and over time, is subjected to negative treatment by one or more individuals, in situations where the person has difficulty defending themselves. A conflict between two equally strong individuals is not considered bullying.” They were then asked: “How often have you been subjected to bullying at your workplace during the past 6 months?” Responses were given on the same five-point frequency scale as used for the NAQ–R.

Personality was measured using the 20-item version of the International Personality Item Pool (IPIP; Goldberg, 1999), known as the Mini-IPIP (Donnellan et al., 2006). This instrument is based on the Five-Factor Model of personality and includes four items for each of the five traits: agreeableness (α = 0.75/0.76; ω = 0.76/0.76), conscientiousness (α = 0.72/0.74; ω = 0.76/0.72), extraversion (α = 0.80/0.82; ω = 0.81/0.82), openness (α = 0.67/0.68; ω = 0.78/0.70), and neuroticism (α = 0.76/0.78; ω = 0.74/0.77). Participants were asked, “How well do the following statements apply to you?,” and responses were given on a seven-point Likert scale (1–7).

Sex, age, education, and foreign-born status, drawn directly from the Swedish National Population Register, were included as covariates in the main analyses, as all have been associated with the occurrence of workplace bullying (Einarsen et al., 2020a). Their inclusion is also important because the stability of personality traits has been shown to differ across demographic groups (Löckenhoff et al., 2008).

All statistical analyses were conducted using Stata version 19.5. Internal consistency was assessed using Cronbach’s alpha (α) and McDonald’s omega (ω). Alpha is reported to facilitate comparability with previous studies, whereas omega provides a complementary model-based estimate of reliability. Omega coefficients were computed using the omegacoef module (Shaw, 2020) in Stata. To investigate the reciprocal relationships between personality and workplace bullying over time, a two-wave cross-lagged panel model was estimated. The model included autoregressive paths for each construct from Time 1 to Time 2, as well as cross-lagged paths from each of the five personality traits at T1 to bullying at T2, and from bullying at T1 to each personality trait at T2. All variables were modelled as observed variables, and the analysis controlled for sex, age, education, and foreign-born status. Structural equation modelling (SEM) was applied using maximum likelihood estimation with missing values (MLMV). Model fit was assessed using the chi-square statistic (χ²), root mean square error of approximation (RMSEA), comparative fit index (CFI), and Tucker–Lewis index (TLI). RMSEA values below 0.05, and CFI and TLI values close to 0.95, were considered indicative of good fit (Hu and Bentler, 1998). Because workplace bullying is a highly skewed variable (as most individuals are not exposed), all coefficients were estimated using 5,000 bootstrap samples.

Previous studies have provided indications of reciprocal associations between workplace bullying and personality (e.g., Farley et al., 2025; Nielsen and Knardahl, 2015). However, the underlying psychological processes that might explain why personality contributes to bullying exposure, and why bullying in turn may shape personality, remain poorly understood. In the following sections, we present a theoretical rationale for these links, outlining why we expect (a) involvement in conflicts, (b) perceived stress, and (c) symptoms of mental health problems to be three key mechanisms. Specifically, based on the results from Part 1, we examine these factors as potential mechanisms through which neuroticism may act as a risk factor for bullying and through which bullying may influence subsequent change in neuroticism and conscientiousness.

Neuroticism reflects individual differences in emotional stability versus emotional reactivity (Cervone and Pervin, 2019). It indicates whether a person generally feels calm, secure, and relaxed, or tends to be anxious, tense, and easily upset. Individuals high on neuroticism are more likely to experience emotions such as anxiety, anger, irritability, guilt, and depression. Explicitly displaying such emotions could, according to victim precipitation theory (Elias, 1986), act as a trigger for harassment, as such reactions may be perceived by others as norm violations—both in terms of social conduct and performance expectations (Tedeschi and Felson, 1994). In the context of bullying, a provocative victim may elicit frustration in others and thereby provoke aggressive behaviour or engage in interactions that escalate conflict (Zapf and Einarsen, 2020). That is, as employees scoring high on neuroticism tend to be anxious and easily upset, their behaviour may be perceived by others as more provocative than that of their more emotionally stable colleagues. Hence, neurotic employees may face an increased risk of being bullied because they behave in ways that elicit aggression from others (Nielsen and Knardahl, 2015). According to this line of reasoning, high levels of neuroticism may activate an interpersonal conflict mechanism that mediates the link between personality and exposure to bullying.

In terms of a vulnerable victim (Zapf and Einarsen, 2020), a disposition to express negative affect—encompassing emotions such as anxiety, worry, sadness, and depression—may also elicit mistreatment. In contrast to the provocative victim, the vulnerable victim may signal weakness, being less likely to retaliate if attacked and instead tending to withdraw. They may be perceived as demanding or interpersonally difficult, and thus seen as triggering aggression, but could also simply represent the most convenient outlet for frustration in cases of predatory bullying (Einarsen, 1999). This type of personality trait has shown the most consistent links to various indicators of victimization (Aquino and Thau, 2009). This suggests that depression, anxiety, and perceived stress may function as an emotional distress mechanism mediating the association between neuroticism and exposure to bullying.

However, the emotional distress mechanism could also be understood as a tendency to perceive behaviours as more negative, in line with the gloomy perception hypothesis (de Lange et al., 2005)—that is, the victim perceive others’ behaviours as more negative and more personally directed. In this sense, individuals high in neuroticism may interpret ambiguous social interactions as hostile or rejecting, thereby perceiving greater exposure to bullying behaviours even in the absence of actual increases in mistreatment.

Taken together, we suggest that the association between neuroticism and subsequent bullying can be understood through two distinct mechanisms: an interpersonal conflict mechanism (conflict involvement) and an emotional distress mechanism (perceived stress, and symptoms of depression and anxiety).

Exposure to workplace bullying represents a severe and prolonged social stressor (Zapf and Einarsen, 2005) that may have enduring consequences for personality development. Repeated experiences of devaluation, exclusion, and loss of control can elicit persistent stress and negative affect, which, over time, may become internalised as stable patterns of emotional reactivity. According to the reformulated learned helplessness model (Abramson et al., 1978), such experiences promote feelings of helplessness and hopelessness by fostering stable and global attributions of failure and lack of agency. These emotional states can develop into enduring dispositions of worry, tension, and vulnerability to stress—consistent with a more neurotic personality profile.

From a broader trauma perspective, exposure to bullying can also be understood in terms of shattered assumptions (Janoff-Bulman, 1992). Persistent humiliation and social exclusion challenge fundamental beliefs about the world as benevolent and meaningful and about the self as competent and valued. When these assumptions are violated, individuals may become chronically vigilant to threat and rejection, further reinforcing negative affectivity and emotional instability. Thus, stress, anxiety, and depressive reactions may not only represent temporary strain but also form the emotional distress mechanism through which bullying contributes to heightened neuroticism over time.

Taken together, these perspectives suggest that exposure to bullying may foster increases in neuroticism through the emotional distress mechanism, whereby stress, anxiety, and depressive reactions become internalised as more enduring patterns of emotional instability.

At the same time, bullying may undermine the self-regulatory resources required for persistence, organization, and control. Conservation of resources theory (Hobfoll, 1989) posits that chronic stress and loss of valued resources trigger spirals of depletion, in which reduced energy and motivation impair goal-directed behaviour. Through the same emotional distress mechanism, prolonged exposure to stress and mental health symptoms may erode conscientiousness—manifesting in disengagement, avoidance, and reduced self-discipline. Consequently, bullying-related stress reactions may operate as a common pathway linking exposure to decreased conscientiousness.

Accordingly, the emotional distress mechanism may also account for reductions in conscientiousness, as prolonged stress and mental health symptoms undermine self-regulation, persistence, and goal-directed behaviour.

In addition to the instruments used in Part 1, Part 2 includes the following inventories:

Symptoms of anxiety and depression were measured using the Hospital Anxiety and Depression Scale (HADS; Zigmond and Snaith, 1983). The instrument includes 14 items—seven assessing anxiety symptoms and seven assessing depression symptoms—referring to experiences during the past week. Each item is rated on a four-point scale (0–3). An example anxiety item is “I feel tense or ‘wound up’,” with response options ranging from not at all to most of the time. A corresponding example for depression is “I have lost interest in my appearance.” Internal consistency was α = 0.87/0.87; ω = 0.85/0.86 for anxiety and α = 0.84/0.85; ω = 0.88/0.88 for depression.

Conflict involvement was measured using a single item:” How often have you experienced difficulties cooperating, tensions, or been in conflict with one or more people at your workplace during the past 6 months?” Responses were given on a five-point frequency scale from 0 = never to 4 = very often.

Perceived stress was also measured using a single item:” Have you felt stressed at work during the past month?” Responses were given on a five-point frequency scale from 0 = not at all to 4 = almost all the time.

Based on the significant associations established in Part 1, follow-up mediation models that included baseline measures for all mediators and outcomes were tested. Indirect effects were estimated using 5,000 bootstrap samples, which is the recommended approach for testing mediation because the product of the a and b paths is not normally distributed and bootstrap confidence intervals therefore provide a more accurate basis for inference (Hayes, 2022). Baron and Kenny’s (1986) stepwise procedure, which for a long time served as a common framework for testing mediation, has since been shown to be overly restrictive, and the bootstrap test is almost always more powerful than the Sobel test they recommended (Zhao et al., 2010). In contemporary mediation analysis, the presence of a significant indirect effect is sufficient to establish mediation (Hayes, 2022). All covariates were the same as in the cross-lagged model. Results are presented graphically in Figures 2–4; however, all results, including 95% bootstrapped confidence intervals (BootCI), are also presented in Supplementary Tables S4–S6.

The present results have several practical implications for how organizations can prevent and respond to workplace bullying. First, the finding of a reciprocal relationship between bullying and neuroticism points to a potential vicious circle in which exposure to bullying increases emotional distress, which in turn could alter dispositions in ways that heighten vulnerability to further mistreatment. This underscores the need for early intervention to break this cycle—by addressing emotional strain and conflict involvement before they consolidate into more enduring patterns of reactivity and relational difficulties.

Second, because emotional distress emerged as the central mechanism explaining all significant associations, attention should be directed towards the emotional consequences of bullying rather than assumed personality weaknesses. Providing timely access to psychological support and opportunities for recovery may prevent short-term stress and anxiety from developing into lasting dispositions of emotional instability or reduced self-regulation.

Third, the finding that conscientiousness decreased indirectly through emotional distress highlights that bullying not only undermines well-being but may also erode goal-directedness, organization, and self-discipline. Restoring a sense of control and agency among targets—through supportive leadership, clear work structures, and manageable demands—may therefore be essential for recovery and reintegration.

Finally, these findings suggest that preventive efforts should target the interaction between individual dispositions and the work environment rather than viewing personality as a fixed risk factor. Strengthening the conflict management climate, ensuring fair procedures, and reducing stressors such as role conflict and ambiguity can limit the activation of traits that otherwise make some individuals more vulnerable in adverse environments. By focusing on both the emotional and organizational mechanisms identified here, organizations can more effectively mitigate the personal and collective costs of workplace bullying.

A notable strength of the study is the large longitudinal probability sample of the Swedish workforce. One could argue that the time lag is too short to detect changes in personality resulting from exposure to bullying. This is not necessarily a limitation, but it requires discussion. Few, if any, participants started to experience bullying at the same time as the baseline survey was distributed—most had already been exposed to bullying behaviours for some time. The majority had been exposed for at least a year at T1, and about one third of those for more than 2 years. Thus, the timeframe of the study is in reality longer than merely the period between T1 and T2, which has important implications for the interpretation of the results. Because all outcomes and mediators at T2 were adjusted for their T1 levels, the cross-lagged and indirect paths reflect incremental change from T1 to T2. Given that most bullied participants had been exposed well before T1, any change in personality traits that occurred prior to baseline is not captured. This implies that the estimates presented here are conservative with respect to the cumulative impact of bullying. Nevertheless, different timeframes should be examined in future research to provide a more detailed understanding of how the bullying process interacts with personality traits.

A further strength of the study is that it was the first to investigate the associations between personality traits and workplace bullying using two separate methods: self-labelling and behavioural experience. This provided a more nuanced picture of how bullying and personality traits are connected. Moreover, in all analyses, we adjusted for T1 levels of all mediators and outcomes, thereby modelling change rather than static levels. Controlling for prior levels typically absorbs a substantial portion of variance through stability paths, which often reduces the strength of other effects. This also contributed to conservative estimates and should be regarded as a notable strength. Furthermore, because most individuals are not exposed to workplace bullying, the phenomenon is highly skewed; therefore, all models were estimated using 5,000 bootstrapped samples—a method that reduces the impact of deviations from normality.

The data are based on self-reports, which could have affected the results, as there is a possibility of subjective interpretations, social desirability, and common method variance (Spector, 2006). However, the main variables of the study are of such a nature that anything other than subjective interpretation would probably introduce more error, as the participants’ subjective experience plays a crucial role in understanding the outcomes. For this reason, alternative and potentially more objective measures might not be appropriate (Nielsen et al., 2020). If social desirability influenced the responses, this would most likely contribute to underestimations of the effects, as exposure to bullying—particularly self-labelled bullying—is less likely to be reported from a social desirability standpoint. Finally, the longitudinal design and the presentation of the central variables in different sections of the questionnaire are factors that may have reduced the likelihood of common method variance (Podsakoff et al., 2003).

Although the Mini-IPIP is a well-validated instrument (Donnellan et al., 2006), it provides a more abbreviated assessment of the Big Five traits than other personality measures that capture these dimensions more comprehensively. This means that some nuances within each trait may not have been detected. The analyses were also based on observed scores, which may introduce measurement error. However, given the longitudinal survey design and the focus on change over time, this approach represents a reasonable and pragmatic choice. Future research could examine whether similar patterns emerge when using more extensive personality measures and latent-variable modelling.

Part of how we have discussed the findings is based on how bullies might react to individuals displaying different personality traits. We do not have data on the reasoning or perceptions of the bullies—we only have data from potential targets. Whether bullies actually perceive certain targets as provoking or otherwise triggering their negative behaviour therefore remains unknown. This is something that future studies should explore further.

Taken together, these strengths and limitations suggest that the results provide a conservative yet robust indication of how personality and bullying interact over time, while pointing to areas where future longitudinal designs could further refine this understanding.

The present study provides new insights into the complex interplay between personality and workplace bullying. The findings indicate reciprocal associations between bullying and neuroticism, showing that personality may influence exposure to bullying while also being shaped by it. As a novel contribution, we introduced two mechanisms to explain these associations—the interpersonal conflict mechanism and the emotional distress mechanism. Together, the two mechanisms fully mediated the effect of neuroticism on subsequent bullying, suggesting that neuroticism may contribute to behaviours or interactions that escalate tension and conflict at work, and that heightened emotional reactivity further increases vulnerability to bullying.

Changes in personality traits as outcomes of exposure to bullying behaviours were, in addition to neuroticism, also found for conscientiousness, and for both traits the emotional distress mechanism fully mediated the association. For conscientiousness, this implies that prolonged exposure to bullying can undermine goal-directedness and self-regulatory capacity.

Overall, the findings highlight that personality should not be regarded as a fixed risk factor but as a dynamic system that interacts with the social environment. Bullying can gradually erode emotional stability and conscientiousness, but these effects occur through emotional and interpersonal processes rather than through the traits themselves. Preventing and addressing bullying is therefore not only a matter of protecting well-being but also of maintaining the psychological resources that sustain adaptive personality functioning over time.