1Recent studies suggest that exposure to bullying in the workplace is a serious problem for many workers around Europe (see also Einarsen et al., 2003). As many as 5-10 % of European employees may suffer from exposure to bullying and harassment at work at any one time. It prevails in both private and public organizations, and finds its victims among men and women as well as among managers and workers alike. Studies also show that exposure to bullying in the workplace is a severe source of stress at work and may be a crippling and devastating problem for those exposed. Although single acts of aggression and harassment occur fairly often in everyday interaction at work, they seem to be associated with severe health problems in the target when occurring on a regular basis (Einarsen & Raknes, 1997 ; Einarsen, 2000).
2The aim of this paper is to present research on workplace bullying conducted during the last decade by the bullying research team at the University of Bergen (Einarsen, 1999 ; 2000 ; Einarsen et al., 2003). Since the late 1980s, our research has focused on the issues of “who is doing what to whom, where why and with what kinds of consequences” in relation to bullying at work. The group published its first peer-reviewed article in 1989 (Matthiesen et al., 1989). The empirical research has been based on large scale surveys with more than 10,000 Norwegian workers (Einarsen et al., 1994 ; Einarsen & Skogstad, 1996), case studies of targets (Matthiesen et al., 2003) and organizations (Einarsen & Raknes, 1997), as well as interview studies (Einarsen et al., 1994).
3Bullying is defined as a situation in which, over a period of time, one or more persons are persistently on the receiving end of negative actions from one or several others in a situation where the one at the receiving end has difficulties defending against these actions (Einarsen & Skogstad, 1996). Hence, the concept of bullying at work refers to all situations where one or more persons feel subjected to negative behavior from others in the workplace over a period of time and in a situation where they for different reasons are unable to defend themselves against these actions. The first core of the definition therefore relates to exposure to repeated and enduring negative acts in the workplace (Einarsen, 2000). The second core dimension is about an imbalance in formal or informal power between the involved parties. Typically, a victim is constantly teased, badgered and insulted and perceives that he or she has little recourse to retaliate in kind (Brodsky, 1976). We may distinguish between work-related bullying (such as being exposed to unreasonable deadlines, unmanageable workloads or other types of behaviors that make the work situation difficult for the victim) and bullying that is primarily person-related (such as insulting remarks, excessive teasing, gossip and rumors, social isolation and exclusion). These types of behaviors may be common and experienced by most people at work from time to time. As a single episode in a positive social climate, such actions may even be harmless. However, when behaviors that are perceived as unwanted by the recipient are systematically and continually aimed at a particular person, and especially in a situation where the victim feels defenseless against the actions or the persons performing them, they become acts of bullying. Among 137 Norwegian victims of bullying at work, social isolation and exclusion, devaluation of one’s work and efforts, and exposure to teasing, insulting remarks and ridicule, were the most common negative acts, as reported by a group of some 200 victims (Einarsen et al., 1994).
4Although there may be many unpleasant experiences requiring our attention and a need for a label that may initiate such attention, we must avoid bullying becoming a popular but misused concept. In Norway, bullying (or mobbing, which is our term) has evolved into a very popular term, and therefore has almost lost its meaning. In everyday language, it has even been used to describe good-hearted joking and horseplay, resulting in a situation where bullying was seen as something rather minor, to be easily accepted and tolerated. If someone resented bullying and reported being seriously hurt, he was seen as a neurotic and hypersensitive person, and was most likely to be blamed for his misfortune (Einarsen et al., 1994). Bullying at work is consequently long-term aggression, mostly of a subtle and psychological nature, directed towards a person who is not able to defend himself in the actual situation, leading to victimization of this person.
5Einarsen (1999) suggested the concepts of dispute-related and predatory bullying to explain the onset of two distinct types of bullying. While dispute-related bullying is preceded by a highly escalated interpersonal conflict and evolves out of an interpersonal dispute, ‘predatory bullying’ refers to cases where the victim has personally done nothing provocative that may reasonably justify the behavior of the ‘predator’. In the latter cases, someone is abusing his power or the target is a victim of scapegoating processes within the group. In the latter cases, a target may be attacked because he or she belongs to a certain out-group, for instance by being the first woman in a local police force, or the target may be bullied as an easy target of frustration and stress caused by other factors. Examples of predatory bullying are then exposure to a highly aggressive leadership style, being singled out as a scapegoat, and the acting-out of prejudice.
6Dispute-related bullying seems to be of three kinds : aggressive behaviors used as tactics in an interpersonal conflict, malingering as a tactic, and resentment to perceived wrongdoing or unfair treatment by one’s opponent. Although interpersonal struggles and conflicts are a natural part of all human interactions and must not be considered as bullying, there may be a thin line between the fights between two parties in an interpersonal conflict and the aggressive behavior used in bullying. In some instances, the social climate at work turns sour and creates conflicts that may escalate into harsh personified conflicts and even office wars, where total destruction of the opponent is seen as the ultimate goal to be gained by the parties (Glasl, 1994).
7Denying the humanity of one’s opponent, thus clearing the way for manipulation, retaliation, elimination and destruction (van de Vliert, 1998), is considered a typical element of a highly escalated conflict. If one of the parties acquires a disadvantaged position in an intense social and interpersonal dispute, he or she may be turned into a victim of bullying (Bjorkqvist et al., 1994). If the parties in an interpersonal conflict perceive that their identity or self-image is denied or attacked, intense emotional reactions are typical (van de Vliert, 1984), including feelings of being insulted, of fear, suspicion, resentment, contempt, anger and so forth. People may then subject each other to bullying behavior or resent the behavior of their opponent to a degree where they feel harassed and victimized even though there are few observable signs of bullying behavior by the alleged offender. It may also be true that claiming to be a victim of bullying may be used as a strategy in interpersonal conflicts, in some cases even used by both parties. In highly intense interpersonal conflicts, aggressive outlets may come from both parties, making the situation rather complex (Einarsen et al., 1994). Hence, it may in some cases in practice actually be rather difficult to differentiate between what bullying is and what is to be regarded as a case of bullying, since the conflicting parties as well as non-involved third parties may perceive and label the situation quite differently. Typical of these cases is also the fact that while the target has a strong sense of being a victim of bullying and being on the receiving end of a host of highly unfair behaviors and sanctions, the alleged offenders as well as many colleagues refuse to acknowledge the perceptions of the target and instead explain the situation as being one where a highly difficult and neurotic person is misperceiving the situation or even just “getting what he deserves”. From a conceptual point of view, the difference between bullying and an interpersonal conflict is not necessarily found in what is done and how it is done (Einarsen et al., 2003), but rather in the frequency and duration of what is done and the ability of the parties to defend themselves and their reputations in the actual situations. As opposed to an interpersonal conflict, bullying is not a mutual and reciprocal process where both parties have the same opportunity to aggress and where the effects of the different parties on the opponent actions must be seen to be equal. Bullying is about having unequal power and about being exposed to negative acts over and over again without being able to defend oneself in the actual situation.
8The concepts of “subjective bullying” and “objective bullying” have also been proposed in order to solve this difficulty of deciding when something is and isn’t bullying, which in my experience can be difficult in many cases of dispute-related bullying (Einarsen et al., 2003). For instance, Niedl (1995) argues that the definitional core of bullying at work must “rest on the subjective perception made by the victim that these repeated acts are hostile, humiliating and intimidating and that they are directed at himself/herself” (p.49). Hence, Niedl argues for bullying to be defined as a subjective construct where the focus is on the perceptions of the targets. “Objective bullying” on the other hand, refers to situations where there is clearly observable evidence or statements from third parties that bullying is taking place. Einarsen and coll. (2003) argue that the stigmatization processes involved in bullying, the often subtle nature of the negative acts, and the fact that power differences are more visible from the point of view of those experiencing it, makes “objective bullying” the main concept of bullying.
9Based on both interview studies and survey data, we have concluded that bullying appears to be a gradually evolving process (Einarsen, 1999). During an escalating interpersonal conflict at work, a person may acquire a disadvantaged position, and may gradually be the subject of stigmatizing actions by colleagues or shop-floor management. These aggressive behaviors may be quite a number of different activities used with the aim or at least the effect of humiliating, intimidating, frightening or punishing the victim. The stigmatizing effects of these activities, and their escalating frequency and intensity, make the victims constantly less able to cope with their daily tasks and the cooperation requirements of the job, thus becoming continually more vulnerable and “deserving targets”.
10During the early phases, indirect and discrete types of behavior prevail. Later on, more direct aggressive acts appear. The victims are isolated and avoided, humiliated in public by being the laughingstocks of the department, and so on. In late phases of the process, both physical and psychological means of violence may be used. Victims of long-lasting bullying are also attacked more frequently than those with a shorter history as victims. In early phases, the victims are attacked only now and then. As the situation escalates, the frequency of the attacks becomes more frequent and harsher, and after some time the victims are attacked on a weekly or even a daily basis (Einarsen et al., 1994 ; Einarsen et al., 2003).
11Scandinavian cases of bullying seem mainly to be dispute-related, although both predatory bullying and mixed cases do occur. However, in countries with a more masculine culture and a culture where the power distance between different groups and between superiors and subordinates is larger than in the Scandinavian countries (see also Hofstede, 1980), predatory bullying may well be the most prevalent kind of bullying (Einarsen, 2000). Figure 1 illustrates the different stages that seem to be involved in the escalation process of dispute-related bullying.
Figure 1. Different stages involved in the escalation process of dispute-related bullying
12Einarsen & Skogstad (1996) report data on the frequency of bullying from 14 different Norwegian “Quality of working life” surveys (n= 7986) including a wide range of organizations and professions such as school teachers, university employees, hotel and restaurant workers, clerks, electricians, psychologists, health care workers and industrial workers. The sample, consisting of 43.9 % men and 55.6 % women, is mostly employed in public sector organizations (85 %). The rate of union membership is generally very high in Norway, and some of the studies are even conducted in collaboration with labor unions. Hence, most respondents are members of a labor union.
13The results of these studies showed on average that 8.6 % of the respondents had experienced bullying and harassment at work during the last six months (Einarsen & Skogstad, 1976). At least some 4.5 % of the respondents were victims of very serious bulling. Many of these victims had been victimized for a long period of time. Mean duration of all reported bullying episodes was reported to be 18 months. Hence, bullying as reported by these victims was not isolated episodes or short conflict intermezzos, but rather ongoing situations where the victims repeatedly experienced aggression from others at work.
14Organizations with many employees, male-dominated organizations and industrial organizations had the highest prevalence of victimization in the last six months. Older workers had a higher prevalence rate than younger workers. The prevalence rate among those between 51 and 60 years of age was 10.3 %. Even if men and women did not differ in the prevalence of bullying, significantly more men were reported as bullies. While 49 % were bullied by one or more men, 30 % were bullied by female perpetrators. Ninety per cent of all male victims reported men among the bullies, while women were bullied to a larger extent by both men and women. Victims reported superiors as bullies as often as they reported colleagues as tormentors. Twenty per cent of the victims were bullied by both superiors and colleagues (Einarsen & Skogstad, 1976).
15A closer look at a large (n = 500) almost all-male industrial organization showed that a large part (89 %) of the workforce had been subjected to some kind of harassment in the last six months (Einarsen & Raknes, 1997). On a weekly basis, 7 % of the men reported being subjected to at least one of the following behaviors from co-workers or supervisors : ridicule and insulting teasing, verbal abuse, rumors and gossip spread about oneself, offending remarks, recurring reminders of blunders, hostility or silence when entering a conversation, or the devaluing of one’s effort and work. As many as 22 % reported being subjected to one or more of these acts at least once a month. Bullying seemed to be highly embedded in this culture, making it a rather common phenomenon.
16However, bullying also takes place in female dominated organizations. In a national representative study of assistant nurses, we investigated whether being part of a minority group may be a risk factor for exposure to bullying (Eriksen & Einarsen, 2004). In this profession, males only comprise some 4 % of the total number of workers. The results showed that 10.2 % (n = 25) of the men and 4.3 % (n = 265) of the women reported that they had been exposed to bullying at work during the previous six months. This association between gender and exposure to bullying at work remained strong and significant also after adjustments for age, marital status, service sector, number of working hours per week, frequency of night shifts, personal commitment to the work unit, perceived mastery of the work, smoking, physical leisure-time activity, and health complaints. Hence, we may conclude that male assistant nurses, being a minority in their profession and in the organizations where they are found, are more often exposed to bullying at work than their female colleagues. Yet, it may not be gender in itself, but rather the fact that these men belong to a minority group in this particular setting that makes them vulnerable to bullying.
17A rather popular view is that these types of behaviors are deeply rooted within the personality structure of the office or shop-floor bully. However, not much empirical evidence exists for this notion. Yet, in one study we did find that self-reported bullies described themselves as being high on aggressiveness and low on self-esteem, the latter being particularly true for a group of offenders who also saw themselves as a target of bullying (Matthiesen & Einarsen, 2005). These perpetrators, labelled as provocative targets, were also found to be low on social competence and high on social anxiety (see also Zapf & Einarsen (2003) for a discussion). A controversial issue in this research is the role of the victim personality. Studies indicate that such factors may in fact play a role, at least in the victimization process resulting from exposure to bullying behaviors in the workplace, and at least in some of the cases. A study of personality and personality disorders among 85 Norwegian victims of bullying at work using a comprehensive measure of personality called the MMPI-2 revealed some interesting insights (Matthiesen & Einarsen, 2001). As a group, victims portrayed a personality profile indicating a tendency to emotional and psychological disturbance on a wide range of personality factors. However, the study showed that victims of bullying were not a homogeneous group. One group of victims portrayed a profile indicating an extreme range of severe psychological problems and personality disturbances. This group, called the “seriously affected”, reported a range of emotional and psychological problems although they reported a relatively low exposure to specific bullying behaviors, a finding indicating that personality is important in determining how bullying is experienced and how it is reacted to. These victims were depressive, anxious, suspicious, uncertain of themselves, and troubled by confused thoughts. A second group, called the “disappointed and depressed” portrayed a tendency towards becoming depressed and being suspicious of the outside world. The third group, called the “common group”, portrayed a quite normal personality, in spite of having experienced the largest number of specific bullying behaviors. Such results may indicate that a specific vulnerability/hardiness factor may exist among some but not all victims of bullying at work. Persons who are already suffering from psychological problems are probably more likely to suffer long-term psychological and physical problems in the wake of bullying and serious personal conflicts. Persons with psychological problems, low self-confidence and a high degree of anxiety in social situations may also be more likely than others to feel bullied and harassed, and they may find it more difficult to defend themselves if they are exposed to the aggression of other people.
18However, a caution must be put forward. Interviews with victims as well as case studies (Einarsen et al., 1994) reveal that bullying seems to exist only in organizational cultures that permit or reward such kinds of behavior (see also Einarsen et al., 2003). Bullying will only take place if the offender feels he or she has the blessing, support, or at least the implicit permission of his superiors to behave in this manner (Brodsky, 1976). In some organizations, bullying may even be institutionalized as a part of leadership and managerial practice. Authoritarian leadership styles are still highly valued in many companies (Hoel & Salin, 2003). Blaming it all on the “psychopaths at work” or even a “neurotic” victim is therefore in most cases a too simplistic solution to the problem of why bullying takes place.
19In addition to the values and norms prevailing in the organizational culture, the quality of the psychosocial work environment seems to be an important cause of bullying. A work situation characterized by role conflict and a lack of interesting and challenging work tasks, combined with a negative interpersonal climate in the work group, seems to be a high risk situation for bullying (Einarsen et al., 1996). A high degree of ambiguity or incompatible demands and expectations about roles, tasks and responsibilities may create a high degree of frustration and conflicts within the work group, especially in connection with rights, obligations, privileges and positions. This situation may then act as a precursor of conflict, poor inter-worker relationships, and a need for a suitable scapegoat, especially if the social climate is characterized by low trust and interpersonal tension. A typical characteristic of workplaces where bullying prevails is also low satisfaction among many employees regarding the leadership style of their managers and supervisors ; it is either too aggressive or too laissez-faire. In fact, as many as 50 % of bullying victims claim to be bullied by a superior, again linking bullying closely to leadership. These findings can be summarized by Leymann’s (1993) theoretical claim that four factors are prominent in eliciting bullying at work: (1) deficiencies in work design, (2) deficiencies in leadership behavior, (3) a socially exposed position of the victim, and (4) a low moral standard in the department.
20When working with victims of long-term bullying, what strikes one the most is the intense and pervasive health problems they display. Looking at both our own research and the research of others in this field, a clear conclusion can be drawn from all the research findings : Exposure to systematic bullying at work causes a host of negative health effects in the target (see also Einarsen & Mikkelsen (2003) for a review). Although single acts of aggression and harassment do occur fairly often in everyday interaction, they seem to be associated with severe health problems when occurring on a regular basis (Einarsen & Raknes, 1997). To be a victim of real or perceived intentional and systematic psychological harm by another person seems to produce severe emotional reactions such as fear, anxiety, helplessness, depression and shock. Such victimization seems to change the individual’s perceptions of his work environment and life in general to one of threat, danger, insecurity and self-questioning (Janoff-Bulman, 1992), which may result in pervasive emotional, psychosomatic and psychiatric problems, according to a host of recent studies (O’Moore et al., 1998 ; Einarsen & Raknes, 1997). In a study of male industrial workers, we found a significant negative association between exposure to bullying at work and measurements of psychological health and well-being (Einarsen & Raknes, 1997). In fact, exposure to bullying explained 23 % of the variance in psychological health and well-being. The strongest relationship existed between experienced personal derogation and psychological well-being. A study of a random sample of Norwegian assistant nurses (Einarsen et al., 1998) showed that nurses reporting exposure to bullying portrayed significantly higher levels of burnout, lower job-satisfaction and lower psychological well-being as compared to their non-bullied colleagues. In another survey conducted among 2215 members of 6 different workers’ unions, significant relationships were found between exposure to bullying and psychological, psychosomatic and musculoskeletal health complaints (Einarsen, et al., 1996). The strongest associations were found between bullying and psychological complaints where measurements of experienced bullying predicted 13 % of the variation. A total of 6 % of the variation in musculoskeletal problems could be statistically predicted by measurements of exposure to bullying.
21In view of the particular symptom constellation found in many studies, it has been argued that many victims of long-term bullying at work may in fact suffer from post-traumatic stress disorder (Björkqvist et al., 1994 ; Leymann, 1992). The PTSD diagnosis refers to a constellation of stress symptoms following a traumatic event, where the trauma first of all is relived through returning, insistent and painful memories of the event, recurring nightmares, or by intense psychological discomfort to reminders. Second, the patient avoids situations associated with the trauma, which may include memory problems with the actual event. Third, the patient may lack the ability to react properly emotionally, for instance by having reduced interest in activities that used to bring joy, by showing limited affect or by the feeling of having no future. Patients with PTSD are also hypersensitive, be it with sleeping problems, difficulties in concentration, by being highly tense and irritable and with bursts of fury, by having exaggerated reactions to unexpected stimuli, or by reacting with physical symptoms to reminders of the actual traumatic situations.
22An early Swedish study among 64 victims of bullying at work attending a rehabilitation program concluded that 65 % of the patients suffered from PTSD (Leymann & Gustaffson, 1996). In a study conducted among 102 victims of long-term bullying at work recruited among members of two Norwegian national associations against bullying, 75 % of the victims portrayed stress symptoms indicating a post-traumatic stress disorder (Einarsen et al., 1999). Even 5 years after the bullying had ceased, as many as 65 % reported symptoms indicating PTSD. On the Hopkins Symptoms CheckList, a total of 76.5 % scored above a level indicating psychiatric pathology as compared to 21.4 % for females and 12.4 % for males in a control group. The level of post-traumatic symptoms was highly related to the intensity of the reported aggressive behaviors, and was especially salient if the aggressive behavior was perceived as being of a personally degrading nature.
23Similar results were found in a group of 124 former Danish victims of workplace bullying (Mikkelsen & Einarsen, 2001). In addition, this study revealed that symptoms of post-traumatic stress were significantly associated with the shattering of a range of basic assumptions about oneself and other human beings generally held by healthy individuals. Also, victims as a group held significantly more negative assumptions about themselves and others as compared to a control group. According to Janoff-Bulman (1992), post-traumatic stress following victimization is largely due to the shattering of basic assumptions that victims hold about themselves and the world, in which the feeling of personal invulnerability comprises an important part. The sense of invulnerability is tied to the three core beliefs : a) the world as benevolent, b) the world as meaningful, and c) the self as worthy. The three core beliefs enable the individual to confront the physical and social environments as if they were stable, orderly, coherent, safe and friendly. A traumatic event such as exposure to persistent aggression by colleagues or managers and supervisors may present information incompatible with these existing mental models, as shown in this particular study. This may again result in a state of extreme anxiety and hyper-arousal, in the long run causing a breakdown in basic psychobiological systems. Exposure to bullying may also evoke our ancient and existential fear of being ostracized and excluded from the group to which we belong.
24Interview studies show that victims typically report being normal and healthy prior to their victimization and that as a result of exposure to bullying, they have subsequently developed severe health problems. In fact, many victims claim that their health has been ruined due to the bullying. Down is shown an example of how a victim describes the magnitude of her health problems. This narrative illustrates how a victim feels that a questionnaire designed to measure psychological and psychosomatic problems following exposure to traumatic events may not sufficiently capture just how much her mental and physical health have been damaged by exposure to bullying (Einarsen et Mikkelsen, 2003).
25Example of a female victim’s narrative of her health problems when responding to a self-report questionnaire measuring psychological and psychosomatic problems:
“This questionnaire did not give me the opportunity to show just how much my health has been ruined. How I went from being a healthy person to a patient with 6 hospitalizations. How my tolerance for stress has changed completely. The slightest stress (such as forgetting where I have put something) makes me stiff in the joints and I start to ache from my ankles right up to my fingers and my neck. My quality of life has been substantially reduced. Answering these questions stirs up many things that I thought were “buried” and forgotten. After three hours, my muscles ache and I have trouble focusing my eyes. Pain and negative feelings are aroused.” (Einarsen et al., 1999)
26However, given the retrospective design of our studies and the use of self-reports, we do not know whether the victims were particularly vulnerable prior to being subjected to bullying, perhaps due to an exposure to other distressing life events. Such exposure might account for the symptoms reported. Indeed, results of our study of 118 Danish victims of bullying (Mikkelsen & Einarsen, 2002) showed that many victims had experienced other distressing life events such as accidents, divorce and bereavements. However, 80.5 % of the victims in this study stated that none of these events affected them more negatively than the bullying they had suffered. Nonetheless, those victims who appeared to be most traumatized, as indicated by their scores on a self-report measure of PTSD, also reported feeling more negatively affected by an event other than bullying. Hence, in addition to the bullying, exposure to one or several other stressful life events might have contributed to their severe health problems.
27Building on the above line of research, a conceptual framework (Einarsen, 2000; Einarsen et al., 2003) has been developed that identifies the main classes of variables to be included in both future research and future organizational action programs (see Figure 2). Four things are important in the model. First of all, this model distinguishes between the nature and causes of bullying behaviors as exhibited by the alleged offender from the nature and causes of the perceptions of the target of these behaviors. Furthermore, it distinguishes between the perceived exposure to bullying behaviors from the reactions to these kinds of behaviors. Third, it focuses on the impact of the organization on both the behavior of the alleged bullies and the perceptions and reactions of the targets. Fourth, the target’s personality is likely to affect how the perpetrator’s behaviors are perceived and even more so how they are reacted to. Last, but not least, the conceptual model pinpoints that the target’s reactions to the bullying may alter the target’s personal characteristics (such as personal styles of coping or even personality), as well as the very organization itself and how it reacts to the particular target.
Figure 2. A conceptual framework for the study and management of bullying at work
From Einarsen et al., 2003
28Looking first at the perpetrator’s behavior, the model proposes that bullying arises out of a combination of an organizational culture that permits or even rewards this kind of misbehavior, and situational, contextual as well as personal factors that may cause a manager or an employee to act aggressively towards subordinates or colleagues (such as stress, conflicts or a highly aggressive personality). Hence, bullying behavior may be a result of the combination of a propensity to bully, due to either personal or situational factors, and the lack of organizational inhibitors of bullying behavior. Brodsky (1976) claimed that although both victims and bullies may suffer from personality disorders, for bullying to be established it must occur within a culture that permits or even rewards this kind of misbehavior. On the basis of survey data on the experiences and attitudes of British union members, Rayner (1998) concluded that bullying prevails due to an organizational tolerance of bullying at work. Ninety-five per cent of the respondents in her UNISON study claimed that bullying was caused by the fact that “bullies can get away with it” and “victims are too scared to report it”. Hence, bullying behavior may be a result of the combination of a propensity to bully and the lack of organizational inhibitors of bullying behavior.
29However, situations where one person offends, provokes or otherwise angers another person often involve substantial discrepancies between the subjective perceptions and interpretations of the conflicting participants. It is therefore vital to distinguish between the observable behavior of perpetrators and the perception of these behaviors by a target, again going back to the difference between subjective and objective bullying. Although the behavior of the perpetrators in most cases is the main cause of the perceptions of targets, perceptions may also be influenced by other factors. Studies of sexual harassment have shown that an incident that is considered mildly offensive by one individual might be seen as serious enough to warrant a formal complaint by others (Tersptra & Baker, 1991). Although personal factors of the victim may not be relevant as a cause of the bully’s behavior, they may be highly relevant when looking at the vulnerability of the victim when facing such persistent aggressive behavior. In a study of perceptions of sexual harassment at work using hypothetical scenarios, we found that personality factors did not affect whether or not female students perceived a particular situation as constituting a case of sexual harassment. However, their reactions to different types of unwanted sexual behavior from men were highly correlated with personality factors (Einarsen et al., 1997). Studies on bullying at work have also shown that variables pertaining to individual differences might affect the degree of reported stress symptoms following exposure to bullying at work. Einarsen et al. (1996) showed that self-esteem and social anxiety moderated the relationships between bullying and self-report measures of psychological, psychosomatic and musculoskeletal health complaints. Victims with high social anxiety reported more psychosomatic symptoms than did victims with low social anxiety. Also, Einarsen et al. (1996) found that victims with high self-esteem reported more psychological and musculoskeletal complaints than victims with low self-esteem. Furthermore, a study among Danish factory employees showed that state-negative affectivity partially mediated the relationships between exposure to bullying behaviors and psychological and psychosomatic health complaints, while generalized self-efficacy acted as a weak moderator of the relationship between exposure to bullying behaviors and psychological health complaints (Mikkelsen & Einarsen, 2002b). In a theoretical overview, Mikkelsen (2001) further proposes that individual variables such as perceived locus of control, attributional style and coping strategies are likely to influence the extent to which victims of bullying develop severe health problems following exposure to bullying at work.
30According to the model, it is however not only the target’s personality that may influence how bullying is perceived and reacted to. Organizational factors, including an effective victim support system, are key factors that may moderate the victim’s perceptions and reactions in a situation where bullying may exist. These factors are therefore important both by inhibiting aggressive behavior in the first place, by potentially reducing the anxiety such behavior may create in a defenseless target, and of course in their own right as means for managing complaints and intervening in specific cases.
31The latter part of the model has clearly an individual, subjective, and most of all a reactive focus. Although bullying at work may to some degree be a subjectively experienced situation in which the meaning assigned to an incident will differ, depending on both the persons and the circumstances involved, this part of the model highlights the necessity in any strategy against bullying to take the victim’s perceptions and reactions seriously and as a real description of how they experience their work environment. This part of the model argues for the inclusion of a rehabilitation program in an effective organizational strategy against bullying.
32This conceptual framework also gives some credit to the dynamic process involved in the interaction between perpetrator, victim and organization. The stress reaction of a target and the consequential effects such perceptions may have may backfire and justify the treatment of the victim (Einarsen et al., 2003). The victim’s stigmatization process discussed earlier may even alter how third parties view the victim,
“which again may change how an organization tolerates, reacts to and manages a particular case of bullying” (Einarsen et al., 2003, p. 25).
33Following from this, it may happen that the alleged bully, the target, and the responses of important third parties and organization representatives may change in the course of the process. The escalation and the dynamics of interaction involved in the victimization process are therefore essential to the understanding of the phenomenon of bullying at work (Einarsen et al., 2003).
34Last but not least, socioeconomic factors and cultural factors will probably affect most of the proposed variables in the model. For instance, power inequalities do vary between national cultures, providing a more or less fertile soil for bullying to take place (Einarsen, 2000). Different national cultures may also have traditions for more or less autocratic and harsh leadership styles, may differ in how conflicts at work are handled, and may have a more or less permissive attitude towards aggressive behavior at work. Differences in national legal systems will relate to how much effort organizations may put into preventive measures and how much protection a target may expect from either the organization or society (Yamada, 2003). In Norway, the work environment act includes a general ban on harassment at work, yet many targets have found it difficult to receive the support of the court in cases of alleged exposure to bullying. However, in 2004, the prime minister of Norway announced a national strategy to prevent bullying at work. Such political statements and efforts, combined with the general awareness that they create, may of course affect how organizations and third parties react to cases of bullying, as well as the behavior of potential perpetrators and the reactions and behaviors of targets. In many countries, bullying is still taboo or even a phenomenon with no name, making it difficult for targets to raise their voices and complain about their treatment. In a highly unionized country such as Norway, where bullying has been on the agenda for more than 20 years, many targets will experience strong support from their unions, especially if the alleged bully is part of management (Sjøtveit, 1992). Socioeconomic factors such as the labor market will also affect how easily a target may change employment in order to escape bullying, while a country’s economy and competitive situation may influence how workers are treated and how much attention organizations are willing to pay to their “human side”.
35Some of our studies may illustrate such cultural differences. A study on cultural differences between the US and Norway in exposure to sexual harassment at work and the emotional reactions of targets to such experiences, conducted on a small sample of females working in two male-dominated manufacturing organizations, showed that Norwegian women reported less unwanted sociosexual behavior than the American respondents. While 60 % of the American women felt repeatedly exposed to sexually harassing behavior, this was true for only 27 % of the Norwegians (Einarsen & Sørum, 1996). However, in the American sample, 36 % saw themselves as victims of sexual harassment, while this was true for only 4.8 % of the Norwegians. Both samples reacted with anger, humiliation and resentment to such conduct. However, the Americans reported significantly more anxiety and resentment than did the Norwegians, who on their side reacted with more grumpiness. Thus, Norwegian women experienced sexually harassing behavior to a lesser degree than American women, and they felt to an even lesser degree victimized by such conduct. The results were explained by the small power differences, the feminine values and the small differences in gender roles that characterize Scandinavian culture.
36In another study, we investigated whether empirical findings from the US on factors affecting the perception of sexual harassment were transferable to Norway and the UK. More specifically, we investigated to what extent the nature of a behavior, the respondent’s gender, as well as the initiator’s position vis-à-vis the respondent, influence the perception of sexual harassment among Norwegian vs. British citizens. American studies had shown these variables to be highly influential with regard to how social-sexual behavior is perceived and labeled within the workplace (Einarsen & Hoel, 2000). An experimental design using hypothetical scenarios was used to investigate how different types of social-sexual behavior were perceived and labeled in Norway and the UK. The two samples consisted of 116 female and 50 male Norwegian students and 133 female and 48 male part-time British students. The mean age was 30 years, ranging from 25 to 45 years. The results showed that while the British sample reacted as predicted by American theories on harassment, the Norwegians did not. Norwegians perceived seductive behavior to be less serious than did the British, while sexual coercion was perceived to be the most negative and the type of behavior most likely to be labeled as sexual harassment. Sexual coercion means that a woman is subjected to threats of punishment or actual sanctions if she refuses to be sexually co-operative. The Norwegians saw this as more serious than actual physical behavior and assaults. Again the results may be explained by the equality, the feminine values and the low power distance characterizing the Norwegian culture. Furthermore, the hierarchical position of the initiator as opposed to the recipient did not influence the judgement of the different kinds of social-sexual behavior in either of the two samples. In the US, which has a greater power distance than Norway and the UK, research shows that gender discrimination and seductive behavior are experienced as more negative and to a larger degree as sexual harassment if a formal hierarchical difference between initiator and recipient exists. This was however not the case either in the Norwegian or in the British culture. Although the latter studies have been conducted in relation to sexual harassment, such cultural differences are also likely to exist in relation to non-sexual harassment and bullying. For instance, Cowie and colleagues (2000) report on a study on exposure to bullying in the United Kingdom and in Portugal that the proportion of targets in the latter country was more than double that in the UK sample. One risk factor for bullying in the UK was length of employment, while belonging to a minority group was a clear-cut risk factor among the Portuguese. While the Portuguese were mainly bullied by their co-workers and to lesser extent by their immediate superiors, the British reported their immediate superior as the main perpetrator of bullying.
37The aim of this article has been to present an overview of theoretical and empirical contributions to the field of research on bullying in the workplace done at the University of Bergen during the last decade. Our main objective when conducting this research on bullying in the workplace has been to contribute to the prevention and constructive management of such problems, and to the healing of individual and organizational wounds resulting from such episodes. To accomplish this, it has been our conviction that different types of information have to be provided (Einarsen, 1996). First we must provide descriptive information on the phenomenon itself, from both the conceptual and empirical points of view. A review of our findings in this respect has been presented in this paper. Second, both theoretical and empirical information on the causes and consequences of the problem is needed. Descriptive data and personal experiences gathered by both victims and professionals may be helpful in this respect, but are by no means sufficient for the implementation of effective interventions, which may only be accomplished through the development of theoretical and empirically sound models of the causes and effects involved in bullying at work. Although both personality and psychosocial factors in the workplace seem to play a role in bullying at work, the causal links and the relative importance of these factors still require further research. A broad outlook taking into account individual, situational, contextual and social factors is definitely needed, as shown by the presented conceptual framework. It is our opinion that bullying must be seen as a complex social phenomenon characterized by multi-causality, involving a range of factors found at many explanatory levels, from individual to cultural and societal, depending on whether we focus on the actor’s behavior or on the target’s perceptions, reactions and responses (Einarsen, 2000).
38The third type of information needed concerns the actions that may be taken to resolve or prevent the problem. Research conducted by Dan Olweus (1993, 1994) on bullying among children showed that personality traits among victims and bullies are highly important causes of victimization in schools. However, the intervention program developed by Olweus (1991) has a strong focus on the school and the classroom as a social system, and involves all children, teachers and parents of the particular school (Olweus, 1991). These findings show that not all possible causes of bullying and harassment in the workplace may be taken into account when intervention strategies are developed. Research must therefore also be conducted on possible interventions and actions steps, including the cost-benefit of various interventions and strategies. It is therefore not sufficient to concentrate on developing and testing causal theories of bullying. Theories and models of action and interventions must therefore be developed, and empirical data on the effectiveness of the various actions must be gathered. Although prevention programs and intervention strategies have been proposed on the issue of bullying at work (Einarsen et al., 1994), these have not yet been based on systematic research and evaluations, and have thus not been presented in this paper. Our future research efforts will therefore concentrate on further research on the causes and consequences of bullying in the workplace in order to establish procedures and guidelines for the prevention and constructive management of bullying in organizations, and for the treatment and rehabilitation of those individuals exposed to the problem. Through further empirical investigations and theoretical, conceptual, and practical developments in terms of prevention, intervention and treatment, our vision is to contribute to the development of ethical organizations with working environments that foster dignity and diversity in the workplace as well as fair treatment and opportunities for all employees. Therefore, the issue of bullying at work is, in our opinion, an issue for all members of the working community and a basic issue of democracy and human rights in modern society.