Empirical Findings on Prevalence and Risk Groups of Bullying in the Workplace

Dieter Zapf Jordi Escartin, Miriam Scheppa-Lahyani, Stale Valvatne Einarsen, Helge Hoel and Maarit Vartia

Introduction 105

The Frequency of Bullying 107

The Duration of Bullying 111

Gender Differences in Bullying 112

Gender of the Victims 113

Gender of the Bullies 114

The Number of Bullies 115

The Organizational Status of Bullies and Victims 117

The Status of the Victim 117

The Status of the Bully 118

The Frequency of Bullying in Various Sectors 121

Categories of Bullying 123

Conclusion 127

Bibliography 143

Introduction

When this volume was published the first time (Einarsen et al., 2003a), the empirical foundation of bullying was rather limited. This has changed tremendously during the last years. There are now meta-analyses available on the effects of bullying on psychological and physical health (Bowling and Beehr,

2006; Nielsen and Einarsen, 2012; Nielsen etal., 2016; Verkuil etal., 2015), sickness absence (Nielsen et al., 2016), the relation between personality and bullying (Nielsen, Glasp et al., 2017), and the impact of methodological factors on prevalence rates of bullying (Nielsen et al., 2010), all a clear indication that the field of bullying research has matured. New studies have also been published with regard to many of the issues described in this chapter.

The phenomenon of bullying, which includes being exposed to persistent insults or offensive remarks, persistent criticism, personal or even physical abuse, has been labelled ‘mobbing at work’ in some Scandinavian and German countries (Leymann, 1996) and ‘bullying at work’ in many English-speaking countries (Liefooghe and Olafson, 1999). Typically, a victim is constantly teased, badgered and insulted, and perceives that he or she has little recourse to retaliate in kind. Bullying may comprise open verbal or physical attacks on the victim, but may also take the form of more subtle acts, such as excluding or isolating the victim from his or her peer group (Einarsen et al., 1994; Leymann, 1996; Zapf, Knorz et al., 1996). The following definition of bullying or mobbing seems to be widely agreed upon (Einarsen et al., this volume):

Bullying at work means harassing, offending, socially excluding someone or negatively affecting someone’s work tasks. In order for the label bullying (or mobbing) to be applied to a particular activity, interaction or process it has to occur repeatedly and regularly (e.g. weekly) and over a period of time (e.g. about six months). Bullying is an escalating process in the course of which the person confronted ends up in an inferior position and becomes the target of systematic negative social acts. A conflict cannot be called bullying if the incident is an isolated event or if two parties of approximately equal ‘strength’ are in conflict.

(Einarsen et al., 2003b, p. 15)

It should be noted that the increased attention bullying has received in research and practice during recent years has not led to an agreement on how to define and operationalize the phenomenon. Rather, there are authors/researchers who use more or less strict definitions with regard to the timeframe (e.g., within the last six months or at least six months) and the frequency of the bullying behaviour (e.g., at least once a week or less often than once a week) (cf. Einarsen et al., this volume; Hoel et al., 1999; Keashly, Tye-Williams et al., this volume; Nielsen etal., 2010; Zapf and Einarsen, 2005).

This chapter aims at summarising some descriptive empirical findings of bullying in the European workplace. We will start with the frequency and the duration of bullying. This is followed by an examination of the gender, number and status of bullies and victims, distribution of bullying across industries and occupations and the use of various categories of bullying. The empirical basis of this chapter is restricted to studies carried out in Europe (see Table 3.4 Appendix for an overview of the included studies). A worldwide comparison of studies of the years 2013-2018 can be found in Leon-Perez et al. (2019).

The Frequency of Bullying

For practical reasons, in particular it is important to know how frequently bullying actually occurs in organizations, because efforts to develop measures against it would depend on this information. However, it is not easy to provide reliable numbers. The problem is that the frequency of bullying depends very much on how it is measured (cf. Hoel et al., 1999; Nielsen et al., 2010; Nielsen et al., this volume). Furthermore, the measurement method employed is influenced by the general understanding of what constitutes bullying.

One of the major approaches in measuring bullying is using a questionnaire consisting of a list of bullying behaviours. Nielsen et al. (2010) called this the 'behavioural experience method’. Another approach is to use a precise definition, e.g., the definition presented above and then ask the respondents to label themselves as bullied or not, bearing this definition in mind. This method has frequently been referred to as the ‘selflabelling method’ (Nielsen etal., 2010). In the meta-analysis of Nielsen et al., the behavioural experience method led to a prevalence rate of 14.8% bullying, whilst the self-labelling method led to a prevalence rate of 11.3% when a definition of bullying was used, compared with 18.1%, if no definition was given (see also Nielsen et al., this volume). In the latter case, researchers have asked directly: ‘Have you been bullied during the last six months?’ (e.g., Rayner, 1997). This typically leads to a comparatively high amount of bullying, because people will also tend to say that they have been bullied when only occasional, minor negative acts have occurred.

Some researchers who administered questionnaires using the behavioural method have used a fixed cut-off point (e.g., Bjorkqvist et al., 1994; Notelaers and Einarsen, 2013). Respondents scoring higher than the cut-off point were considered to be victims of bullying. Usually, these studies report a prevalence rate as high as 10-17% bullying (cf. Table 3.4, Appendix). Other researchers using the behavioural experience method applied a strategy developed by Leymann (1996) which we will call the ‘Leymann criterion’: Here, the Leymann Inventory of Psychological Terrorization (LIPT; Leymann, 1990, 1996), or a similar questionnaire such as the Negative Acts Questionnaire (NAQ; Einarsen et al., 2009; Escartin et al., 2012; Notelaers et al., 2019) is administered. To be considered a bullying victim, the response to at least one item or to one general item on the frequency of bullying actions should be: ‘at least once a week’, and the duration of bullying should be ‘at least six months’. The weighted (for sample size) mean prevalence rate (see, e.g., Schmidt and Hunter, 2014) for studies using this strategy (see Table 3.4, Appendix, and the summary Table 3.1) was 9.6%. Those studies using the weekly criterion, but asking for bullying ‘within the last 6 months’, had a weighted prevalence rate of 11.2%. In the meta-analy-sis of Nielsen et al. (2010), the self-labelling-method led to a lower prevalence rate than the behavioural experience method. In the present data, studies were included that used some kind of definition, in most of the cases, similar to the one of Einarsen et al. (2003b) cited above. This led to a prevalence rate of 6.0%. When combining the self-labelling-method with the behavioural experience criterion of bullying ‘at least once a week’, we could not observe any differences between studies that asked for bullying ‘within the last six months’ or ‘more than six months’. To account for this finding is likely that most definitions included that the bullying w'ould go on for a longer time. We therefore considered both groups of studies. This led to a weighted prevalence rate of 3.0% bullying for the combined criterion in the studies included in the present review (see Table 3.1 and Table 3.4 in the Appendix). These data show that not all who are exposed to weekly negative behaviours feel victimized (9.6/11.2% vs. 3.0%) and not all who feel victimized are exposed to weekly negative behaviours (6.0% vs. 3.0%).

Compared to the previous versions of the chapter in earlier editions of this book (Zapf etal., 2003,2011), the overall sample sizes have more than doubled. The studies using the Leymann criterion and the ‘within last six months’ now provide similar prevalence rates suggesting that the frequency criterion is more important than the duration criterion. During the last 10 years, more studies involving large, partly representative, samples have been published. They report relatively similar prevalence rates for the definition or the combined criterion approach.

Table 3.1 Prevalence Rate of Workplace Bullying according to Different Criteria

Notes:

k = Number of studies included in the analysis. The sum of k is unequal 103. because one study could comprise more than one result.

Leymann criterion: Negative acts at least once a week for at least six months

Experience sampling 1/week: Negative acts at least once a week within the last six months

Experience sampling + Definition: Negative acts at least once a week plus self-labelled victim status

With regard to the experience sampling method some studies still report very high prevalence rates, but at the same time low rates for the self-labelling method (e.g., Eisermann and de Constanzo. 2011). In most of these cases, work-related items such as permanent high workload or being frequently interrupted at work appear to be the reasons for the high prevalence rates. However, in such cases, most employees do not consider themselves as bullying victims. Comparing studies published until 2000 and the following two decades does not show a clear trend (Table 3.1). Studies published until 2000 are limited in number. Comparing the other two decades after 2000, all prevalence rates decreased. An explanation for this could be that there is sufficient problem awareness since 20 years so that this factor did not further increase the prevalence rates of bullying, whereas successful intervention measures in recent years may have contributed to a decrease of the prevalence rates.

Moreover, we carried out analyses for different European regions' (see Table 3.1). One of the problems here is that most of the studies have been carried out in Northern Europe including Ireland and the UK. The variance among single studies is relatively high (see Table 3.4). A single large study (Evrin and Madziala, 2016) is, for example, responsible for the high prevalence rate according to the experience sampling method for Eastern Europe. Overall, high prevalence rates according to one criterion are counterbalanced by low prevalence rates in other criteria. It is, therefore, difficult to conclude that bullying is especially high or low in one of the European regions.

Taking the combination of self-labelling and weekly bullying as indicators of severe bullying, it can be concluded that a figure of between 3% and 4% serious bullying has emerged as an average prevalence rate for European workplaces in the sense of the above-given definition. For somewhat less severe cases (including bullying experienced less often than weekly and of a duration of less than six months), the meta-analytical results of Nielsen et al. (2010) as well as our own results based on the studies in Table 3.4, suggest a figure of about 10% bullying. Moreover, the meta-analytical results of Nielsen et al. on self-labelled bullying without a definition and the present data suggest that in many organizations, up to 20% of the employees are occasionally exposed to negative social acts frequently associated with bullying, such as being yelled at, teased or humiliated. Although this does not fall within the stringent definition of bullying, it does imply that these employees are exposed to severe social stressors at work which may also lead to symptoms of psychological strain (for a discussion from a methodological perspective see Nielsen et al., this volume).

When we started analysing the prevalence rate of bullying in the first edition of this book (Zapf et al.. 2003), there were hardly any studies available from outside Europe. As shown by the summaries of Keashly (2018) and Leon-Perez et al. (2019), this is not so anymore. Studies from other continents suggest

According to the criteria of the United Nations: www.worldatlas.com/ articles/the-four-european-regions-as-defined-by-the-united-nations-geoscheme-for-europe.html that the bullying prevalence rates might be even higher there (see also Nielsen et al., 2010). However, comparisons are difficult to draw because of the different measures of bullying used (Escartin et al., 2019), the influence of culture (see Grimard and Lee, this volume; Salin et al., 2019) and other contextual factors such as working conditions or leadership. More systematic cross-cultural studies are necessary here to be able to draw firm conclusions. To account for the differences in prevalence rates of bullying between European countries the EU Foundation (Eurofound, 2015), for example, pointed to the following factors: Differences in awareness of the phenomenon, its causes and consequences; the extent of debate and initiatives about bullying and harassment by trade unions, employers and governments; and the level of tolerance for violence and harassment within society (for an overview, see Hoel and Vartia, 2018).

From a practitioner’s perspective, a figure of 3% of employees reporting serious bullying in a 1,000-employee-strong organization means that around 30 people are exposed to bullying at any one time. Given that not only the targets, but also many of the bullies and bystanders are, in one way or another, likely to be negatively affected by the bullying situation, we would consider this to be a sizeable figure indicating a very serious problem.

The Duration of Bullying

In daily working life in Europe, the terms ‘mobbing’ or ‘bullying’ are often used to account for even minor conflicts and arguments. Therefore, the duration of bullying is an important criterion to differentiate between bullying and everyday conflicts in organizations (Baillien et al., 2017). Studies reporting on the duration of bullying are summarized in Table 3.2. These studies show that bullying is a long-lasting conflict. Looking at some large representative samples in Sweden (Leymann, 1996), Norway (Einarsen and Skogstad, 1996) and Germany (Meschkutat et al., 2002), the average duration of bullying was 15, 18 and 16 months respectively. Among bullied Finnish prison officers, 66% of the women and 53% of the men had been bullied for more than two years (Vartia and Hyyti, 2002). In the study by Hoel and Cooper (2000), 39% of the victims had been bullied for more than two years. Among victims in a Finnish municipal institution 29% had been bullied for 2-5 years and as many as 30% for over five years (Vartia, 2001). In studies of victims only, the average duration was much higher, with a mean of more than three years (e.g., Leymann and Gustafsson,

Table 3.2 Average Duration of Workplace Bullying in Months

1996; Zapf, 1999a). This difference is probably due to method discrepancies: Thus, if one tries to identify and enlist bullying victims via help-lines or self-help groups, etc., one will end up with a self-selected sample of more severely bullied victims (see also Nielsen and Einarsen, 2008). The figures for duration given above underscore that bullying is not a short episode but a long-lasting process that ‘wears down’ the victims, in most cases lasting much longer than one year.

Gender Differences in Bullying

A frequently asked question among the public is whether there are gender differences in bullying. Although data exist on the gender of bullies and victims, there is limited theorizing or in-depth research on this issue (Escartin, Salin etal., 2011; Salin and Hoel, 2013; Vartia and Hyyti, 2002). An overview that puts gender and its complex relationship into context can be found in Salin (2018).

Gender of the

Victims

One can argue that there exists some relation between female socialization and the victim role because women are said to be brought up to be less self-assertive and less aggressive, and tend to be more obliging than men (Bjorkqvist, 1994). Consequently, women would be even less able than men to defend themselves when bullying emerges. Moreover, for various reasons, women hold less powerful positions in organizations (Salin, 2018; Salin and Hoel, 2013). For example, they are less often occupying managerial or supervisor positions (Davidson and Cooper, 1992). To explore this issue, we carried out an analysis based on 80 samples of bullying victims, most of them listed in Table 3.4, which reported gender distribution among victims. Weighted percentages with regard to sample size (total sample size N = 14,119 victims; k = 79 samples) showed that 65.8% of victims were women and 34.2% men.

An analysis of those studies where the gender distribution of victims and the gender distribution of the total samples were available led to the following results: Of the more than 10,000 victim (N = 10,974 from k = 55 samples), a total of 66.4% were w'omen and 33.6% were men. These victims emerged from a total sample of nearly one hundred thousand employees (N = 99,431) with a gender distribution of 63.4% women and 36.6% men. This contrasts with the gender distribution of the workforce within the European Union (EU), where women make up 46.2% (Catalyst Eurostat Database, 2O19^[1]). These figures show that the men/women ratio of victims in our data base corresponds closely to the respective ratio in the overall sample, with women only marginally overrepresented among victims (a difference of 3%) whereas the gender distribution in our data base deviates substantially from the gender distribution in the EU (a difference of 17.2%). This suggests that the over-representation of women among victims is by and large due to the over-representation of women in the respective populations. Of course, one could argue that bullying in some sectors and occupations is higher because of their overrepresentation of women. For example, women make up around 70% of the healthservice sector worldwide (Boniol et al., 2019, p. 1) and there is evidence that the bullying prevalence rate is high in this sector

(e.g., Di Martino et al., 2003; Leon-Perez et al., 2019; Zapf, 1999a). However, if women’s attitudes and behaviours played a role, there should still have been an effect if the baseline (and thus the industry/occupation) is controlled for. All in all, there seems to be little evidence that women are more at risk because of any gendered socialization.

Nevertheless, in some samples, there exists a higher risk for women to be victimized. In the case of Nuutinen et al. (1999)’s police sample, the explanation of women’s higher risk of victimization may lie in their visibility in a male-dominated organization (see also Archer, 1999). Minority groups who differ from the main groups in salient characteristics carry a higher risk of being socially excluded from the group (Schuster, 1996; see also Zapf and Einarsen, this volume and Lewis et al., this volume). It follows that women may be seen as intruders in the male-dominated cultures of researchers, business professionals or the police force (Archer, 1999; Hoel et al., 2001). Yet, in a study among a large representative sample of assistant nurses where men only represent a small minority of less than 3%, male nurses were nearly three times more likely being a target of bullying compared to female assistant nurses (Eriksen and Einarsen, 2004).

Gender of

the Bullies

Less information is available on the gender of bullies. In the studies by Zapf (1999a), (N = 209) altogether 26% of victims reported being bullied by men only, 11% were bullied solely by women, whilst in 63% of all cases both men and women were identified as bullies. Einarsen and Skogstad (1996) reported that 49% of the victims were bullied by men, 30% by women and that in 21% of all cases the bullies were both men and women. The respective numbers in the study by Mackensen von Astfeld (2000) were: 32% men, 27% women and 37% bullied by both men and women. All in all. men seem to be clearly over-represented among the bullies in most studies (see also Meschkutat et al., 2002; Rayner, 1997; an exception is UNISON, 1997). This result corresponds to similar findings in research on bullying in schools (Olweus, 1994). Bullying, at least in part, includes forms of direct aggression, such as shouting or humiliating someone. There is substantial empirical evidence that this kind of aggression is much more typical for men than for women, who for their part tend to make more use of indirect aggression such as social exclusion or spreading rumours (Bjorkqvist, 1994). Moreover, as managers and supervisors appear to play a dominant role in bullying scenarios (see below), and the fact that men are over-represented in such positions, this may explain why men are more often identified as bullies than women.

Finally, Leymann (1993a, 1993b) reported that women are more often bullied by other women, whilst men are more frequently bullied by other men, which he explained in terms of labour market segregation. Similar results were reported by Einarsen and Skogstad (1996), Hoel et al. (2001), Mackensen von Astfeld (2000), Meschkutat et al. (2002), Niedl (1995), Rayner (1997), and Zapf (1999a). Whereas women are sometimes exclusively bullied by men, it appears to be rare that men are exclusively bullied by other women. This finding may be explained by the different power positions of men and women in organizations.

The Number of Bullies

Although bullying can be a conflict between two people, some victims report that everyone in the organization is bullying them. Data on the number of bullies in various studies are summarized in Table 3.3. Weighted by sample size (see footnote c), in 48% of all cases, there was only one bully involved, in 35% there were 2-4 bullies, and in 17%, there were more than four bullies involved. In the German studies of bullying victims by Zapf (1999a), being bullied by only one person was much rarer. In fact, in these studies, in more than 50% of all cases more than four bullies were involved. These differences may be explained as follows: As described above, samples made up of bullying victims usually consist of more serious bullying cases, which, for example, show a longer mean duration of the bullying conflict. There is some evidence that bullying becomes more and more severe the longer it lasts. Studies by Einarsen and Skogstad (1996) and Zapf and Gross (2001) showed that bullying incidents/negative acts occurred more often the longer it lasted. In the study by Zapf (1999a), the duration of bullying correlated positively with the number of bullies. The average duration of bullying of those who were bullied by only one person was 28 months, for those who were bullied by 2-4, and by more than four people, the duration was 36 months and 55 months respectively. These data suggest that it is getting increasingly difficult to remain a neutral bystander the longer

Table 3.3 The Number of Bullies (%)

Notes:

N Sample size

^a The third category of this study was ’4 and more bullies’

^b The middle category of this study was ’2-5 bullies’

^c Because this sample represents almost 50% of the sample size and would therefore have a very strong impact on the overall result we decided to weight this sample by N = 500. If we would use the real sample size the respective numbers would be 61%, 28% and 11%. bullying goes on (see also Niven et al., this volume). Therefore, more and more people may become involved as bullies in the course of time. This may explain the higher mean number of bullies in the pure victim samples which show a higher mean duration of bullying.

Some studies, especially the British ones (Hoel and Cooper. 2000; Rayner, 1997), report that many victims share their experience with other colleagues. For example, in the study by Hoel and Cooper, as much as 55% of the bullying victims reported that they shared their experience with other work colleagues, and 15% reported that everyone in the w'ork group was bullied. Similar results were reported in the UNISON studies (1997, 2000). In other countries, such as Austria (Niedl, 1995) or Germany (Zapf, 1999a, 1999b), this is reported only occasionally. This may be a country-specific phenomenon; however, it may also have to do with the definition of bullying. The more stringent the definition of bullying, the more likely it is that it involves only one victim. While a perpetrator may occasionally bully everyone in the work group for months and years, it seems much more unlikely that he or she can bully to such intensity that everyone in the work group is exposed to bullying at least on a weekly basis.

The Organizational Status of Bullies and Victims

In the following, we review research findings on the organizational status of bullies and victims. Organizational status in this respect refers to the formal position within the organizational hierarchy.

The Status of the Victim

Relatively little has been reported about the status of the victim. Einarsen and Raknes (1997). in a study of male employees at a Norwegian engineering plant, found no difference between the experience of negative behaviours for workers, on the one hand, and supervisors/managers on the other. Similar results were found by Hoel et al. (2001). They found the risk of being bullied to be similar for workers, supervisors and middle or senior management. A representative sample of Finnish employees showed that white-collar employees in higher ranks experienced bullying somewhat more often than low'er-ranked whitecollar employees or workers (Piirainen et al., 2000). Salin (2001), however, found less bullying at the higher levels of the organization. Skogstad et al. (2008), in a representative sample of the Norwegian workforce, showed that although managers reported the same level of exposure to bullying behaviours, they labelled their experiences less as bullying compared to non-managers. Hoel et al. (2001) report some interesting interaction effects with gender: Whereas male workers and supervisors were bullied more than women at these levels, this was the other way round at the management level. The largest differences occurred for the senior management level, where 16% of the female senior managers reported having been bullied. This finding may be due to the visibility of women at this male-dominated hierarchical level and may reflect widespread prejudice against women in leadership positions (see also Davidson and Cooper, 1992).

All in all. the findings of Hoel et al. (2001) question a common assumption in various European countries that the weak and defenceless, in terms of organizational status, become the primary victims of bullying. Rather, there seem to be similar risks at all organizational status levels. Supervisors and senior managers may also experience a power imbalance relative to their colleagues and superiors.

The Status of the Bully

By contrast, the issue of perpetrator status has received considerable attention. Interestingly, the findings vary across countries. Leymann (1993b) introduced ‘mobbing’ as the definition of a lasting conflict among colleagues. Yet even in his study, there were only marginally more colleagues among the bullies than there were supervisors. However, taking the Scandinavian studies as a whole, people in superior positions were identified as offenders in approximately equal numbers to peers, with only a small number bullied by a subordinate (Einarsen and Skogstad, 1996; Leymann, 1992, 1993b). In contrast, British studies have consistently identified people in superiory positions as perpetrators in an overwhelming majority of cases (Cowie et al., 2000; Hoel et al., 2001; Rayner, 1997). Analysing the available samples listed in Table 3.4 (total N = 17,919 victims, k = 60 samples), the percentages weighted by sample size were as follows: 50.0% were bullied by supervisors, 42.5% by colleagues and 7.5% by subordinates respectively. For the last analysis, we took into consideration that in samples where no subordinates were reported as bullies, this category might not have been offered as a possible response. Therefore, these studies were excluded in the computation of the percentage of bullying by subordinates. Thus, the overall picture across countries is—given that in organizations the number of supervisors is much smaller than the number of colleagues— supervisors are more often identified as the bullies than are colleagues. However, the numbers for colleagues and subordinates involved clearly speak against the view that bullying is primarily a top-down process. It is interesting to note that compared to our summary in Zapf et al. (20H), there was a notable drop of supervisor involvement from 65.4% to 50% and an increase in colleague involvement from 39.4% to 42.5%. One can only speculate why this is so. One possible reason could be that in various Anglo-American countries, a belief prevailed that bullying was associated with the behaviour of supervisors and managers, maybe due to the influencing book of Andrea Adams (1992). This view might have changed in the meantime and colleagues are now equally perceived as potential perpetrators in these countries.

Einarsen (2000), referring to Hofstede (1993), argued that some cultural differences between the Nordic and the central European countries may explain some of the different findings with regard to the organizational status of the bullies. Hofstede’s studies suggest that low power differentials and feminine values prevail in the Scandinavian countries. The abuse of formal power is much more sanctioned in such countries. Power differences between immediate supervisors and their colleagues are small, hence producing more similar numbers of perpetrators for supervisors and colleagues. As far as Sweden is concerned, the predominance of horizontal bullying is also explained by reference to country-specific factors, such as strong emphasis on group loyalty and conformity, and a belief in consensus, or collective understanding, with social exclusion for perceived norm-breaching as a common feature (see Beale and Hoel, 2010). In a Danish study by Ortega et al. (2008), peer bullying was found to be the most typical kind of bullying, with colleagues being the main perpetrators in more than 70% of the cases.

Generally, superiors are seldom bullied by subordinates. In particular, there are only a small number of cases reported where superiors were exclusively bullied by their subordinates. Usually, subordinates bully a superior in conjunction with other supervisors or managers. The reason for this is, of course, that it is not easy to overcome the formal power of a superior using informal power. Although it is possible if the superior is socially isolated (which points at tensions or conflicts within management), but it is almost impossible if the superior is backed up by superiors at the same level and/or by senior management. One can certainly say that only superiors, who have lost the support of their colleagues and of senior management or are considered a threat by fellow managers (Hoel et al., 2001), carry the risk of becoming the victims of bullying by subordinates. A detailed discussion of upward bullying can be found in Branch et al. (2018).

Whereas in the previous edition of this chapter (Zapf et al.. 2011) we had to state that little is knowm about patterned negative supervisory behaviour, and that leadership studies have focused almost exclusively on the positive aspects of leadership, this has substantially changed in recent years. Although some of the common leadership questionnaires, such as the leader behaviour description questionnaire LBDQ (Fleishman, 1953), contain some items similar to those which appear in workplace bullying questionnaires, negative leadership behaviour has not really been investigated within this tradition. Humiliating, yelling or threatening somebody is, however, not simply the absence of positive leadership characteristics such as consideration or employee orientation. Bullying by superiors is, therefore, an issue for research into leadership in its own right (see Aasland et al.. 2009; Einarsen et al., 2007) and destructive leadership and abusive supervision (Tepper, 2000. 2007) which come close to or resemble supervisor bullying, have received much attention (see the meta-analyses of Mackey et al., 2017 and Schyns and Schilling, 2013). In a study employing a large scale sample of UK workers, Hoel and colleagues (2010) showed that both authoritarian, laissez-faire and inconsistent leadership were associated with experiences of bullying as reported by victims and observers alike. Yet, while observers regarded authoritarian leadership to be most strongly associated with bullying, reports by victims about bullying were most closely related to inconsistent leadership in the form of unpredictable punishment. In a representative study of Norwegian workers, reports of bullying, as made by both victims and observers, were strongly correlated with reports of tyrannical leadership from one’s immediate supervisor (Hauge et al., 2007), while Skogstad and colleagues (2007), employing the same data source, showed that laissez-faire leadership was associated with reports of bullying through its effect on role stressors and interpersonal conflicts in the work environment. Hence, there appears to be strong support for a view that leadership styles are related to experiences of bullying among targets as well as among observers.

The Frequency of Bullying in Various Sectors

In this section, we summarize some findings on the frequency of bullying in various sectors. Leymann (1993a, 1993b) reported an over-representation of bullying in the educational (approximately 2:1) and administrative (1.5:1) sectors, and an under-representation in the trade and retailing, production and health-service sectors. The prevalence of bullying in Swedish public administration was 1% higher than the average score of 3.5% (Leymann. 1993a). However, in other studies, Leymann also found a high level of occurrence in the health-service sector. In another study by Leymann and Gustafsson (1996), public administration, the social and health sectors, as well as religious organizations showed higher prevalence, whereas trade and industry reported lower levels of bullying. Meschkutat et al. (2002), Niedl (1995), Piirainen et al. (2000) and Vartia (1993, 1996) also report high levels of bullying in the health and social sector. In Leon-Pérez et al.’s (2019) study, most empirical studies reviewed come from this sector and many of them report high prevalence rates. Examples are the studies of Bambi et al. (2014), Hpgh et al. (2018) or Stapelfeldt et al. (2013). In the studies by Einarsen and Skogstad (1996), the highest rate of frequent bullying (weekly or more often) was found among clerical workers (3.9%) and within trade and commerce (3.5%). For occasional bullying, the results were different. Here, in contrast to Leymann’s Swedish study, there were significantly fewer respondents from public sector organizations who reported bullying than from private enterprises. The highest prevalence rate was found among industrial workers, where 17.4% reported having been occasionally bullied during the last six months. Bullying was also frequent among those who did graphical work, and hotel and restaurant w'orkers. The lowest rate of bullying was found among psychologists and university employees.

In Germany, analyses based on almost 400 victims of serious bullying (Zapf, 1999a) showed that employees within the health and social services sector had a seven-fold risk of being bullied. Other occupational sectors where the risk of bullying where high or elevated were: public administration, 3.5:1, and the educational sector, 3:1. Moreover, there was also an increased risk of being bullied in the banking and insurance sectors. In contrast. the risk was relatively low in the areas of transport, trade and fanning, in the hotel and restaurant sector, as well as in the building industry. Hubert and van Veldhoven (2001) found increased risks of aggressive and unpleasant behaviour in service organizations, in industry and in education. Salin (2001) reported more frequent bullying in the public sector than in the private sector, as did Hoel et al. (2001), and Piirainen et al. (2000) in the municipal sector rather than the private sector or the civil service. More recent studies from Eisermann and de Costanzo (2011) and Venetoklis and Kettunen (2016) report similar results.

Taking the studies together, a higher risk of being bullied is reported for the social and health, public administration and education sectors, which all belong to the public sector in many countries. There may be various reasons which explain the differences between sectors. First, one may assume that bullying is less frequent in small family enterprises such as the hotel and restaurant business as well as in the building sector. Here personal relationships can be expected to develop between employees and between employers. If severe conflicts arise, one party may leave the ‘family’, as mobility within these sectors are generally high. Moreover, in these areas, short-term job contracts prevail; thus prolonged conflicts lasting several years are almost impossible because the employees would find it relatively easy to leave their jobs.

On the other hand, in many European countries—for example Germany, Norway and Sweden—working in public administration means having a secure, lifelong job which usually compensates for a somewhat lower-than-average salary. In this case, it is much more complicated to give up one’s job when bullying occurs, because this would involve giving up the high job security which is among the most important aspects of these jobs. Frequently the specific knowledge gained in such jobs cannot easily be applied in the private sector. Moreover, moving to another job within the public sector may not resolve the problem because one still finds oneself within the same organization. A typical example would be the case of a bullied police officer. In a unitary organization such as the police force, rumours may spread fast and, in case of a requested move the officer’s potential new superior might receive biased information, and, to be on the safe side, possibly reject the bullied officer’s application (cf. Leymann, 1993b).

Yet another aspect may be inherent in the very nature of the job itself. Some jobs in the service sector, and in particular in the social and health service sector, require a high level of personal involvement, i.e., a form of emotional labour (Hochschild, 1983; Zapf, 2002), which means sensing and expressing emotions and building personal relationships. In other jobs, such as manufacturing work, a much more instrumental job attitude may suffice. The higher the level of personal involvement, the more personal information is out in the open, and the more possibilities for being attacked would therefore exist. Moreover, it is much more difficult to objectively evaluate or appraise these jobs which offer a lot of opportunity for attacking or unfairly criticizing someone. If a production worker is accused of doing a bad job, they can more easily defend themselves by referring to their job description compared to a teacher or a nurse, who may have much greater difficulties proving that they are doing a good job.

All in all, looking at the distribution of bullying across sectors, bullying seems to be a greater problem among white-collar workers, service employees and employees in supervisory positions than among blue-collar workers. Still, a representative study of the Norwegian workforce from 2005 showed that bullying prevailed in all kinds of organizations with no particular sector being ‘bully proof’ (Einarsen et a!., 2007; Rayner etal., 2002).

Categories of Bullying

The final question addressed in this chapter is: Is bullying a homogeneous construct or are there specific types or categories of bullying which can be identified? Homogeneity of bullying would imply that all bullying actions show similar frequencies, have similar causes and consequences and occur under the same circumstances (Zapf, Knorz et al.. 1996). Leymann (1996) differentiated between five classes of bullying behaviour, which he referred to as the manipulation of: (1) the victim’s reputation;

• (2) the victim's possibilities of communicating with coworkers;
• (3) the victim’s social relationships; (4) the quality of a person’s occupational and life situation; and (5) the victim’s health. In an empirical study, Leymann (1992) found factors which he labelled as negative communication: humiliating behaviour, isolating behaviour, frequent changes of tasks to punish someone, and violence or threat of violence. Using factor analyses.

Zapf, Knorz et al. (1996) found seven factors in two samples: Organizational measures consisting of behaviours initiated by the supervisor or aspects directly related to the victim’s tasks. ‘Social isolation' is related to informal social relationships at work. The third factor is related to individual attributes of the victim and the victim’s private life. ‘Physical violence' includes two items of sexual harassment as well as general physical violence or threat of violence. ‘Attacking the victim’s attitudes’ is related to political, national and religious attitudes. The factor ‘verbal aggression’ consists of items related to verbal attacks. Finally, there was a factor consisting of two items related to spreading rumours (for comparable results see Niedl, 1995; Vartia, 1991, 1993; zur Mühlen et al., 2001). More recent studies have come to similar results employing different questionnaires (cf. also Nielsen et al., this volume. Table 6.1; Escartin et al., 2019).

Factor analysis of the Negative Acts Questionnaire (NAQ) (Einarsen and Raknes, 1997) identified five factors, four of which appear to overlap with attacking the private person, social isolation, work-related measures and physical violence. Based on a revised version of the NAQ-R applied to a random sample of 5,288 UK employees, Einarsen et al. (2009) found three major factors: Person-related bullying, work-related bullying and physical intimidation. In the most recent version, the short form SNAQ (Notelaers et al., 2019), items relating to work-related, person-related bullying and social exclusion were included in the nine-item scale.

Taking the existing studies together, most researchers have suggested differentiating between work-related bullying and person-related bullying. For work-related bullying researchers used a general factor in most of the cases, as in the case of Einarsen et al. (2009) for the much used NAQ-R, and in some cases authors suggested various categories such as control and manipulation of information, and control and abuse of working conditions (Escartin et al., 2010). With regard to person-related bullying, a variety of sub-categories were suggested such as verbal aggression (e.g., Zapf, Knorz et al., 1996), isolation or social exclusion (e.g., Einarsen and Raknes, 1997; Einarsen et al., 2009; Escartin et al., 2009; Notelaers et al., 2019; Yildirim and Yildirim, 2007; Zapf, Knorz et al., 1996), emotional abuse (Keashly, 1998) and humiliation (e.g., attacks towards self-esteem: Ozturk et al., 2008; attacking the victim's private life: Moreno-Jiménez et al., 2007; Zapf, Knorz et al., 1996; and personal derogation: Einarsen and

Raknes, 1997). Moreover, most bullying categories can be considered to be active forms of aggression (most kinds of work-related bullying, verbal aggression, emotional abuse). Occasionally, passive forms of aggression are used such as withholding or not passing on information. Moreover, both direct forms of aggression (e.g., verbal aggression and most forms of emotional abuse) and indirect forms of aggression (e.g., spreading rumours: Zapf, Knorz et al., 1996; most forms of work-related bullying) occur. Finally, physical and psychological bullying can be distinguished, as can social exclusion and ostracism. In the shipyard study by Einarsen and Raknes (1997), physical violence was reported by 2.4%, whilst in the various studies reported by Zapf (1999a) physical aggression occurred in between 3.6 and 9.1% of the bullying cases. Thus, the results underline that, in the first instance, bullying is primarily a form of psychological rather than physical aggression although some cultural differences exist (Escartin, Zapf et al., 2011).

Correlational analyses of overall samples (e.g., Niedl, 1995) show that the bullying categories are very highly correlated. This means that if people are bullied, they tend to experience a large number of bullying behaviours from different behavioural categories. In Notelaers et al’s (2019) latent cluster analysis, in the cluster representing the severely bullied all bullying categories (work-related, person-related, social exclusion) showed high scores. With regard to gender-specific bullying categories, Leymann and Tallgren (1993) report that women used slander and making someone look a fool, whereas men preferred social isolation. Mackensen von Astfeld (2000) found that women used significantly more strategies affecting communication, social relationships and social reputation, whereas men preferred strategies affecting the victim’s work. In a sense, these results correspond to findings regarding schoolyard bullying. Here Bjôrkqvist et al. (1992) found that boys used physical aggression more often, whereas girls preferred more indirect strategies such as rumours and social exclusion. In Vartia’s (1993) study, women were more often the victims of strategies of indirect aggression such as spreading rumours and social isolation, whereas men were more often the victims of threats and criticism. However, to challenge possible stereotyping, Hoel and Cooper’s (2000) nationwide British study reported that negative rumour and gossiping was particularly widespread in the police service, a highly male-dominated organization.

The meaning of work-related bullying is not always clear. Fevre et al. (2010) and Olafsson and Johannsdottir (2004) pointed out that behaviours such as ‘excessive monitoring of work’ or ‘being given tasks with unreasonable deadlines’ may not necessarily be seen as indicators of bullying but as a (legitimate) part of a manager’s behavioural repertoire. In fact, cluster analyses (Notelaers etal., 2006.2019) showed separate clusters for employees who were only exposed to work-related bullying. However, they can be considered bullying when applied excessively or for personal gain, and especially when they occur together with other types of bullying behaviour (Beale and Hoel, 2011) as is the case in the cluster of the severely bullied in the analyses of Notelaers et al. (2006, 2019).

Work-related strategies including acts such as being given tasks with impossible targets or deadlines, having one’s opinions and views ignored, and being given work clearly below one’s level of competence seem to be experienced more often among persons in superior positions (Hoel et al., 2001; Salin. 2001). In the studies reported by Zapf (1999a) and Zapf. Knorz et al. (1996), coworkers used social isolation and attacking the private sphere more often than the supervisors or managers. Bullying was most frequent when both coworkers and supervisors were among the bullies. If only supervisors were identified as bullies, strategies such as social isolation, attacking the private sphere and spreading rumours occurred less often.

One explanation for these findings may be that some categories, such as social isolation and spreading rumours, only work if several people are involved. Hence, it is far more difficult for a single supervisor to isolate somebody. For other bullying categories, such as attacking the private sphere, personal and private information about the victim is necessary, which may be less often at hand for superiors.

Finally Escartin et al. (2009) were interested in what kind of bullying is experienced as most severe. They found that emotional abuse, a form of person-oriented bullying, was considered to be the most severe category, whereas isolation and devaluating professional roles were perceived as the least severe categories. In the study by Zapf, Knorz et al. (1996), attacking the private person, a kind of person-oriented bullying behaviour that overlaps strongly with the emotional abuse scale of Escartin et al's (2009) study was by far the strongest predictor of psychosomatic complaints and depression, whereas indirect forms of aggression such as isolation were only weakly related to ill-health. Escartin et al. (2009. p. 200) concluded that ‘all in all. there seems to be converging evidence that bullying behaviours such as humiliating someone, treating someone with disdain or ridiculing them premeditatedly, summarized as ‘emotional abuse’ in this paper, are seen as most severe and causing most harm to the target’.

Conclusion

This chapter has summarized empirical findings of bullying studies in European countries over the past 30 years. Although different definitions and measures were used in these studies, and although there may be some cultural differences, a converging picture emerges showing that about 3% of employees may experience serious bullying, and about 10% occasional bullying. Between 10 and 20% (or even higher) of employees may occasionally be confronted with negative social behaviour at work which does not correspond to strict definitions of bullying but which is, nevertheless, still very stressful for the persons concerned. In most countries, there seems to be a tendency for bullying to occur more often in the public sector, although bullying seem to exist in all sectors of working life. Bullying occurs on all organizational levels and finds its targets among young and old and among women as well as men. Yet, men seem to be more often among the perpetrators. Perpetrators for their part may be supervisors or colleagues. Most studies report an average duration of bullying well beyond one year. Bullying can be a conflict between two people; however, very often, there is more than one perpetrator. More and more people seem to become involved the longer bullying lasts. Finally, there is some empirical evidence showing that a variety of bullying behaviours exists. At least some of the variations found in separate studies may be due to cultural differences. It is also important to note that overall findings may mask underlying trends with regard to prevalence as well as the nature of experience, for example with respect to gender and occupational status. Summarizing the existing results on workplace bullying shows that great progress has been made during the last three decades, which, overall, has led to converging results in the various European countries.

Appendix

Table 3.4 Studies on the Frequency ofWorkplace Bullying

Tôrôk et al. (2016)

General working population

(Continued)

(Continued)

Country

et al. (2017)

Slovenia

Mumel et al. (2015)

General working population

Spain

Kovacic et al. (2017)

Fidalgo and Piñuel

(2004)

General working population

General working population

Camero et al. (2012) Escartin et al. (2012) Losa-Iglesias and de Bcngoa(2012)

General working population

General working population

Nurses

Country

Authors

Sample

Forssell (2016)

Xu et al. (2018)

Switzerland Tong et al. (2017)

Turkey Cemaloglu (2007)

Soylu et al. (2008) Ozturk et al. (2008) Yildirim et al. (2007)

Yildirim and Yildirim (2007)

Bilgel et al. (2006)

Aytac et al. (2011) Gök (2011)

Yapici et al. (2011) Civilidag (2014) Ertürk and Cemaloglu (2014)

Picakciefe et al. (2017)

Yagci and Uluöz (2017)

Minibas-Poussard et al. (2018)

General working population

Multicohort study (Sweden.

Denmark & Finland)

Care workers in nursing homes

School teachers

General working population

Academic Nurses

University Nursing School

Academics

Nurses from the European side of Istanbul province

Public Sector Organizations

General working population

Banking employees

Agriculture industry

Hotel employees

Teachers

Health workers

Teachers

Faculty member participants at junior ranks

'Notes:

• 1. denotes duration of acts: 1 a within the last six months; 1 b over six months; 1c ever in the career
• 2. denotes type of acts included in judgements (it is asked ‘intention to harass’)
• 3. denotes frequency of acts: 3a at least weekly; 3b less frequently than weekly; 3c ever
• 4. denotes victims label themselves as bullied based on a definition
• 5. denotes approximate criterion
• 6. denotes number of negative acts per week: 6a one negative act; 6b two negative acts or more; 6 at least one negative act
• 7. Latent class cluster analysis (LCC)
• 8. denotes victims label themselves (without a definition)

Bibliography

Aasland, M. S.. Skogstad, A., Notelaers. G.. Nielsen. M. B. and Einarsen, S. (2009) The prevalence of destructive leadership behaviour. British Journal of Management. 21(2), 438-452.

Adams. A. (1992) Bullying at work. How to confront and overcome it. London: Virago Press.

Agervold, M. (2007) Bullying at work: A discussion of definitions and prevalence, based on an empirical study. Scandinavian Journal of Psychology, 48. 161-172.

---. (2009) The significance of organizational factors for the incidence of bullying. Scandinavian Journal of Psychology, 50(3), 267-276.

Agervold, M. and Mikkelsen, G. E. (2004) Relationships between bullying, psychosocial work environment and individual stress reactions. Work & Stress, 18(4). 336-351.

Apospori. E. and Papalexandris, N. (2008) Workplace bullying and organizational culture: A multi-level approach. 6th international conference on workplace bullying (pp. 52-54). 4-6 June. Montreal, Canada.

Archer. D. (1999) Exploring “bullying” culture in the para-military organisation. International Journal of Manpower, 20. 94-105.

Arenas. A.. Giorgi. G„ Montani, E. Mancuso, S.. Perez. J. E, Mucci. N. and Arcangeli. G. (2015) Workplace bullying in a sample of Italian and Spanish employees and its relationship with job satisfaction, and psychological well-being. Frontiers in Psychology, 6. 1912.

Ashforth. В. E. (1994) Petty tyranny in organizations. Human Relations, 47, 755-778.

Aytac, S., Bozkurt, V., Bayram, N.. Yildiz, S.. Aytac, M.. Akinci, F. S. and Bilgel, N. (2011) Workplace violence: A study of Turkish workers. International Journal of Occupational Safety and Ergonomics. 17(4), 385-402.

Báguena. M. J., Beleña, M. A., De la Paz Toldos. M. P. and Martínez, D. (2011) Psychological harassment in the workplace: Methods of evaluation and prevalence. The Open Criminology Journal. 4. 102-108.

Baillien. E„ Escartín, J.. Gross. C. and Zape. D. (2017) Towards a conceptual and empirical differentiation between workplace bullying and interpersonal conflict. European Journal of Work and Organizational Psychology, 26(6). 870-881.

Balducci. C„ Alfano. V. and Fraccaroli, F. (2009) Relationships between mobbing at work and MMPI-2 personality profile, post-traumatic stress symptoms, and suicidal ideation and behavior. Violence and Victims. 24(1). 52-67.

Bambi, S.. Becattini. G.. Giusti, G. D.. Mezzetti, A.. Guazzini, A. and Lumini, E. (2014) Lateral hostilities among nurses employed in intensive care units, emergency departments, operating rooms, and emergency medical services: A national survey in Italy. Dimensions of Critical Care Nursing, 33(6). 347-354.

Baruch. Y. (2005) Bullying on the net: Adverse behavior on e-mail and its impact. Information & Management. 42. 361-371.

Beale. D. and Hoel, H. (2010) Workplace bullying, industrial relations and the challenge for management: Britain and Sweden compared. European Journal of Industrial Relations. 16. 101-118.

---. (2011) Workplace bullying and the employment relationship: Exploring questions of prevention, control and context. Work, Employment and Society. 25(1), 5-18.

Bernotaite. L. and Malinauskiene. V. (2017) Workplace bullying and mental health among teachers in relation to psychosocial job characteristics and burnout. International Journal of Occupational Medicine and Environmental Health. 30(4), 629-640.

Bilgel, N.. Aytac, S. and Bayram, N. (2006) Bullying in Turkish white-collar workers. Occupational Medicine. 56. 226-231.

Birks. M.. Cant, R. P., Budden, L. M.. Russell-Westhead, M.. Üzar ÔzçETiN, Y. S. and Tee. S. (2017) Uncovering degrees of workplace bullying: A comparison of baccalaureate nursing students’ experiences during clinical placement in Australia and the UK. Nurse Education in Practice. 25. 14-21.

Bjôrkqvist, K. (1994) Sex differences in aggression. Sex Roles, 30. 177-188.

Bjôrkqvist, K.. Lagerspetz. K. M. J. and Kaukiainen. A. (1992) Do girls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behavior. 18, 117-127.

Bjôrkqvist, К., Osterman. K. and Hjelt-Bâck, M. (1994) Aggression among university employees. Aggressive Behavior. 20,173-184.

Boniol, M.. McIsaac, M., Xu. L., Wuliji, T.. Diallo, K. and Campbell, J. (2019) Gender equity in the health workforce: Analysis of 104 countries. Working paper 1. World Health Organization. Geneva.

Bowling. N. A. and Beehr. T. A. (2006) Workplace harassment from the victim's perspective: A theoretical model and metaanalysis. Journal of Applied Psychology. 91, 998-1012.

Branch. S.. Ramsay, S.. Shallcross. L„ Hedges, A. and Barker. M. (2018) Bosses get bullied too: Exploring upwards bullying to learn more about workplace bullying. In P. D'Cruz, et al. (eds.), Pathways of Job-related negative behaviour. Handbooks of workplace bullying, emotional abuse and harassment (Vol. 2, pp. 1-32). Singapore: Springer Nature.

Buka, M. S. M. and Karaj. T. (2012) Mobbing in the academe: The case of Albanian universities. Proceedings of the 2nd International Conference on Human and Social Sciences, 8, 26-29.

Campanini P.. Gilioli. R.. Punzi. S.. Cassitto, M. G.. Conway. P. M. and Costa. G. (2008) Workplace bullying in a large sample of Italian workers. Sixth International Conference on Workplace Bullying. I Book of Abstracts: Sixth International Conference on Workplace Bullying, 4-6 June 2008, Montreal. Canada.

Carnero, M. A., Marti'nez. B. and Sanchez-Mangas, R. (20I2) Mobbing and workers' health: Empirical analysis for Spain. International Journal of Manpower. 33(3), 322-339.

Carretero, N. and Luciano. J. V. (2013) Prevalence and incidence of workplace bullying among Spanish employees working with people with intellectual disability. Disability and Health Journal, 6(4). 405-409.

Carter. M.. Thompson. N.. Crampton. P„ Morrow. G., Burford. B.. Gray, C. and Illing. J. (2013) Workplace bullying in the UK NHS: A questionnaire and interview study on prevalence, impact and barriers to reporting. BMJ Open. 3(6), e002628.

Catalyst (2019) Women in the workforce - Europe: Quick take (January 9. 2019) Retrieved from www.catalyst.org/research/ women-in-the-workforce-europe/; 5.9.2019.

Ceja. L.. Escarti'n, J. and Rodri'guez-Carballeira, A. (2012) Organizational contexts that foster positive behaviour and wellbeing: A comparison between family-owned firms and nonfamily businesses. Revista de Psicologia Social. 27(). 69-84.

Cemaloglu, N. (2007) Okul yoneticilerinin liderlik stilleri ile yildirma arasindaki iliski. Hacettepe Universitesi Egitim Fakiiltesi Dergisi, 33. 77-87.

Chatziioannidis. I.. Bascialla, F. G.. Chatzivalsama, P.. Vouzas. F. and Mitsiakos. G. (2018) Prevalence, causes and mental health impact of workplace bullying in the Neonatal Intensive Care Unit environment. BMJ Open. 8(2). e018766.

Chirila, T. (2012) Perceived victimization as a consequence of bullying among Romanian employees: Gender differences. Annals of the Al. I. Cuza University. Psychology, 21(2). 85-98.

Civilidag, A. (2014) Hotel employees’ mobbing, burnout, job satisfaction and perceived organizational support: A research on hospitality in Turkey. European Scientific Journal. 10(35).

Conway. P. M.. Clausen, T.. Hansen, A. M. and Hogh. A. (2016) Workplace bullying and sickness presenteeism: Cross-sectional and prospective associations in a 2-year follow-up study. International Archives of Occupational and Environmental Health, 89(). 103-114.

Cowie, H., Jennifer. D.. Neto, C.. Angula. J. C„ Pereira, B.. del Barrio, C. and Ananiadou, K. (2000) Comparing the nature of workplace bullying in two European countries: Portugal and the UK. In M. Sheehan. S. Ramsey and J. Patrick (eds.). Transcending the boundaries: Integrating people, processes and systems. Proceedings of the 2000 Conference (pp. 128-133). Brisbane: Griffith University.

Coyne, I., Craig. J. and Smith-Lee, P. (2004) Workplace bullying in a group context. British Journal of Guidance & Counselling. 32(3), 301-317.

Coyne, I., Smith-Lee, P.. Seigne. E. and Randall, P. (2003) Self and peer nominations of bullying: An analysis of incident rates, individual differences, and perceptions of the working environment. European Journal of Work and Organizational Psychology, 12(3). 209-228.

Cubela, V. and Kvartuc, T. (2007) Effects of mobbing on justice beliefs and adjustment. European Psychologist. 12(4). 261-271.

Da Silva-Joao. A. L. and Saldanha-Portelada. A. F. (2016) Mobbing and its impact on interpersonal relationships at the workplace. Journal of Interpersonal Violence.

Davidson, M. J. and Cooper. C. L. (1992) Shattering the glass ceiling. London: Paul Chapman Publishing.

Dehue, E. Bolman. C.. Völlink, T. and Pouwelse, M. (2012) Coping with bullying at work and health related problems. International Journal of Stress Management. /9(3). 175-197.

Dick. U. and Dulz, K. (1994) Zwischenbericht Mobbing-Telefon für den Zeitraum 23.8.93-22.2.1994 [Intermediate report of the mobbing telephone], Hamburg: AOK.

Di Martino, V.. Hoel, H. and Cooper, C. (2003) Preventing violence and harassment in the workplace. European Foundation for the Improvement of Living and Working Conditions. Luxemburg: Office for Official Publications of the European Communities.

Dobesovä Cakirpaloglu. S.. Cech, T. and Kvintovä. J. (2017) The incidence of workplace bullying in Czech teachers. In P. A. Da Silva Pereira. O. Titrek and G. Sezen-Gultekin (eds.), ICLEL 17 conference proceeding book (pp. 425-431). Sakarya: Sakarya University.

Dogar. N. (2016) Workplace bullying perceptions among health sector employees: A research in a private hospital in Albania. Istanbul Gelifim Üniversitesi Sosyal Bilimler Dergisi, 3(1). 105.

Ege, H. (1998) I numeri del Mobbing. La prima ricera italiana [The frequency of bullying. The first Italian study], Bologna: Pitagora Editrice.

Einarsen, S. (2000) Harassment and bullying at work: A review of the Scandinavian approach. Aggression and Violent Behavior. 4.371-401.

Einarsen, S.. Aasland. M. S. and Skogstad. A. (2007) Destructive leadership: A definition and a conceptual model. The Leadership Quarterly, 18. 207-216.

Einarsen, S„ Hoel. H. and Notelaers. G. (2009) Measuring exposure to bullying and harassment at work: Validity, factor structure and psychometric properties of the Negative Acts Questionnaire-Revised. Work & Stress, 23(1), 24-44.

Einarsen. S.. Hoel. H., Zapf, D. and Cooper, C. L. (eds.) (2003a) Bullying and emotional abuse in the workplace. International perspectives in research and practice.

---. (2003b) The concept of bullying at work: The European tradition. In S. Einarsen. H. Hoel. D. Zapf and C. L. Cooper (eds.), Bullying and emotional abuse in the workplace. International perspectives in research and practice

Einarsen, S„ Matthiesen. S. B. and Skogstad. A. (1998) Bullying, burnout and well-being among assistant nurses. Journal of Occupational Health and Safety, 14, 563-568.