psywork1

Individual and situational predictors of workplace bullying: Why do perpetrators engage in the bullying of others?

Lars Johan Hauge*, Anders Skogstad and Ståle Einarsen

Faculty of Psychology, University of Bergen, Norway

Stressful working environments are often assumed to create conditions that may lead to

bullying. However, few studies have investigated how factors experienced in the work

environment may trigger perpetrators to engage in bullying of others. Drawing on Spector

and Fox’s (2005) stressor�emotion model of counterproductive work behaviour, the present study investigated the predictive effects of both individual and situational factors as predictors

of being a perpetrator of workplace bullying, as applied to a representative sample of the

Norwegian workforce (N�2359). Results from logistic regression analysis show that being oneself a target of bullying, regardless of the frequency, and being male strongly predicted

involvement in bullying of others. Among the situational factors, only role conflict and

interpersonal conflicts significantly predicted being a perpetrator of bullying. The present

findings support the notion that bullying will thrive in stressful working environments and

thus yield an important contribution in identifying antecedent conditions to counteract the

development of bullying at workplaces.

Keywords: bullying; harassment; aggression; job stress; perpetrator

Introduction

Stressful working environments have long been assumed to create conditions that

may lead to the development of bullying at work (Leymann, 1996). Although the

reasons for why bullying develops at workplaces may be many and interwoven, and

be related to characteristics of both the targeted individual and the perpetrator,

prevailing explanations emphasize the importance of problematic organizational and

work-related conditions as underlying factors in this process (cf. Bowling & Beehr,

2006; Einarsen, Hoel, Zapf, & Cooper, 2003; Leymann, 1996). A growing body of

research during the last couple of decades has shown a range of work-related factors

to be related to exposure to workplace bullying, while considerably less attention has

been devoted seeking to explain why perpetrators engage in bullying. The present

study aims to address this gap in workplace bullying research by investigating

individual and situational factors that may encourage individuals to become

perpetrators of bullying at work. After reviewing the relatively limited empirical

knowledge that exists on perpetrators of workplace bullying, we will investigate the

effects of the situational factors decision authority, role ambiguity, role conflict and

interpersonal conflicts as predictors of engaging in bullying of others, drawing on the

*Corresponding author. Email: [email protected]

ISSN 0267-8373 print/ISSN 1464-5335 online

# 2009 Taylor & Francis DOI: 10.1080/02678370903395568

http://www.informaworld.com

Work & Stress

Vol. 23, No. 4, October�December 2009, 349�358

stressor�emotion model of counterproductive work behaviour as a theoretical backdrop (Spector & Fox, 2005). Their predictive effects will be investigated after

also taking into account the effects of four individual factors, namely target status,

gender, hierarchical position and age.

The phenomenon of workplace bullying refers to a gradually evolving process,

whereby an individual ends up in an inferior position and becomes the target of systematic negative social acts by one or more perpetrators (Brodsky, 1976).

Workplace bullying consists of repeated and prolonged exposure to predominantly

psychological mistreatment, directed at a target who is typically teased, badgered

and insulted, and who perceives himself or herself as not having the opportunity to

retaliate in kind (Einarsen et al., 2003). Workplace bullying can take the form of

direct acts, such as verbal abuse, accusations and public humiliation, but it can also

be of a more subtle and indirect nature in the form of gossiping, rumour spreading

and social exclusion (Einarsen, Hoel, & Notelaers, 2009). However, when frequently

and persistently directed at the same individual, even such subtle and indirect

behaviour can be experienced as an extreme source of social stress at work (Zapf,

1999). Exposure to workplace bullying has repeatedly been shown to have

detrimental consequences for affected individuals and to have wide-ranging negative

consequences for organizations at large (cf. Aquino & Thau, 2009; Bowling & Beehr,

2006).

Several explanations have been put forward to account for why individuals engage in bullying of others at work. Engaging as a perpetrator of workplace

bullying has been proposed to be a consequence of oneself being exposed to bullying

and as a problem-focused coping strategy in defending oneself against further acts of

mistreatment (cf. Aquino & Thau, 2009; Lee & Brotheridge, 2006). Others have

proposed that bullying develops as a result of lack of social competencies and as a

result of micro-political behaviour within organizations, and further as a self-

regulatory process with regard to protection of ones self-esteem (see Zapf &

Einarsen, 2003 for a comprehensive discussion). However, although Zapf and

Einarsen argue that individual and personality factors on the part of the perpetrator

probably do play a role in the development of workplace bullying, they strongly

argue against one-sided and mono-causal explanations. Explanations for why such

behaviour takes place within workplaces will probably be too simplistic without also

taking into account work-related and organizational factors. In this sense, the

stressor�emotion model of counterproductive work behaviour may prove useful (cf. Spector & Fox, 2005).

According to the stressor�emotion model, stressors experienced in the work environment may induce negative emotions in some individuals, which, in turn, may

lead them to engage in aggressive behaviour towards others. Processes leading up to

aggressive behaviour are further related both to individual characteristics and to

whether the individual perceives him or herself to be in control of the situation

inducing the experience of stress and negative emotions. While several studies have

shown a range of work-related factors to be related to being exposed to workplace

bullying (see e.g. Hauge, Skogstad, & Einarsen, 2007 for a review), few studies have

so far explored how such factors may relate to being a perpetrator of bullying. Still,

reviewing studies relating to counterproductive work behaviour, Spector and Fox

(2005) identified experienced role ambiguity, role conflict and interpersonal conflicts

as important precipitating conditions for engaging in aggressive behaviour targeted

350 L. J. Hauge et al.

towards other individuals in the organization, thus corresponding to work-related

factors that are normally found as strong correlates of exposure to workplace

bullying (cf. Bowling & Beehr, 2006).

Due to difficulties in collecting and obtaining valid and reliable information,

existing empirical knowledge on perpetrators and perpetrator characteristics is

scarce and has mainly been obtained from self-reports of targets of bullying, while less evidence has been presented reflecting self-reports from perpetrators themselves

(Matthiesen & Einarsen, 2007; Zapf & Einarsen, 2003). Yet, the existing evidence

shows perpetrators to be males more often than females (e.g. De Cuyper, Baillien, &

De Witte, 2009; Hershcovis et al., 2007; Rayner, 1997), and to be supervisors and

managers more often than subordinates (e.g. Hoel, Cooper, & Faragher, 2001),

although Scandinavian studies in general report approximately equal numbers of

perpetrators among supervisors and subordinates (cf. Zapf, Einarsen, Hoel, &

Vartia, 2003). Engaging in aggressive behaviour is also frequently associated with

age. It is assumed that with increasing age, individuals better understand the

consequences of their behaviour and that they therefore are more likely to exert

control over their anger (Barling, Dupré, & Kelloway, 2009). However, studies have

shown mixed results in relation to age, with some studies reporting a negative

relationship (e.g. De Cuyper et al., 2009; Inness, Barling, & Turner, 2005), while

other studies have found no significant relationship between age and engaging in

aggressive behaviour at work (e.g. Glomb & Liao, 2003). Only a few studies have reported prevalence rates of perpetrators of bullying. In a

UK study, prevalence rates varied largely from 19.3% when applying a sole self-

report measure to 2.7% when applying a more stringent criterion reflecting both self-

and peer-reported behaviour (Coyne, Chong, Seigne, & Randall, 2003). None of the

perpetrators in the UK study reported being both a target and a perpetrator of

bullying. In a Norwegian study, Matthiesen and Einarsen (2007) found perpetrators

to yield a total prevalence of 7.5%. In addition to non-involved individuals, they

further distinguished between respondents who were perpetrators only, and

perpetrators who were both perpetrators and targets of bullying, constituting 5.4%

and 2.1% of the sample, respectively. This latter group of targets, who also engage in

bullying of others, has been characterized by a combination of both anxious and

aggressive reaction patterns. Their behaviour may cause irritation and tension in

their surroundings, and corresponds to that of those who, in school research, have

been labelled provocative victims or bully/targets (cf. Olweus, 2003). Matthiesen and

Einarsen (2007) showed that the bully/target group reported significantly lower levels

of self-esteem and higher levels of role stress than did both the perpetrator only

group and non-involved individuals. Moreover, Hauge and colleagues (2007) showed bully/targets to report significantly lower levels of job satisfaction and higher levels

of job stress than did non-involved individuals. However, applying a strict criterion

for significance, no significant differences were identified between the perpetrator-

only group and non-involved individuals.

Based on the relatively limited systematic knowledge that exists on perpetrators

of bullying, the present study will investigate both individual and situational factors

as predictors for engaging in bullying of others. Although some evidence has been

provided for why individuals engage in aggressive behaviour at work, drawing

conclusions based on zero-order correlations may capitalize on chance and

potentially lead to erroneous conclusions (cf. Barling et al., 2009). Drawing on

Work & Stress 351

situational factors that previously have been found to be of importance in relation to

a stressor�emotion framework (cf. Spector & Fox, 2005), the predictive effects of decision authority, role ambiguity, role conflict and interpersonal conflicts will be

investigated applying a multivariate design. In line with the existing empirical knowledge, the predictive effects of these situational factors will be investigated after

also taking into account the effects of the individual factors target status, gender,

hierarchical position and age.

Method

Sample and procedure

This study constitutes an extension and reanalysis of data as employed by Hauge and

colleagues in a previous study (2007). The study sample is based on a representative

sample of the Norwegian workforce, collected through anonymous self-report

questionnaires by Statistics Norway (SSB). A total of 2539 questionnaires were

returned, yielding a response rate of 56.4%. To be able to retain as many respondents as possible for the analyses and to avoid loss of power, responses with minor number

of missing values were imputed by the EM algorithm in SPSS, yielding a sample of

2359 cases to be analysed. Thus, some minor differences existed as compared to the

overlapping results presented by Hauge et al. (2007). Males yielded 48.5% of the

analysed sample, with 19.8% reporting to be supervisors. Mean age was 43.7 years.

As previous studies have identified the individual factors target status, gender,

hierarchical position and age as possible important factors in relation to being a

perpetrator of bullying, the effects of these variables were taken into account in the analyses. The situational factors investigated in the present study refer to decision

authority, role ambiguity, role conflict and interpersonal conflicts (cf. Rizzo, House,

& Lirtzman, 1970; van Veldhoven & Meijman, 1994). To measure perpetration of

workplace bullying, respondents were asked to indicate if they themselves had

exposed others to bullying at their workplace during the last six months. Response

categories were ‘‘no,’’ ‘‘yes, to some extent’’ and ‘‘yes, to a large extent,’’ where the

last two categories were combined to avoid too small groups for the analyses.

Exposure to workplace bullying was measured by a self-labelling measure, whereby respondents were asked to indicate whether they considered themselves to have been

exposed to bullying at work during the last six months. Response categories were

‘‘no,’’ ‘‘rarely,’’ ‘‘now and then,’’ ‘‘once a week’’ and ‘‘several times a week.’’

Both exposure to and perpetration of workplace bullying were measured

according to the following definition: ‘‘Bullying takes place when one or more

persons systematically and over time feel that they have been subjected to negative

treatment on the part of one or more persons, in a situation in which the person(s)

exposed to the treatment have difficulty in defending themselves against it. It is not bullying when two equally strong opponents are in conflict with each other’’ (for a

full description of the measurement instruments applied, see Hauge et al., 2007).

Analyses

To obtain zero-order correlations between the study variables, Pearson’s product

moment correlations were computed. For the correlational analysis, exposure to

workplace bullying was analysed as a binary variable (i.e. ‘‘no’’ and ‘‘yes’’). To

352 L. J. Hauge et al.

predict being a perpetrator of workplace bullying, logistic regression analysis was

conducted. Only variables correlating significantly with being a perpetrator of

bullying were included in the regression analysis. The predictor variable exposure to

workplace bullying was recoded into two dummy variables for the regression

analysis, reflecting being exposed to bullying occasionally (i.e. ‘‘rarely’’ and ‘‘now

and then’’) and on a weekly basis (i.e. ‘‘once a week’’ and ‘‘several times a week’’),

with no exposure to bullying as the reference group. The odds ratio obtained from

the logistic regression analysis is to be interpreted as the likelihood of being a

perpetrator of workplace bullying with an increase in a predictor variable by one unit

(cf. Tabachnick & Fidell, 2007).

Results and discussion

Perpetrators yielded 2.9% of the sample, with 1.9% being perpetrators only while a

further 1% of the perpetrators reported also being bullied themselves. The prevalence

rate found in the present sample is thus somewhat lower than what has been reported

in previous studies reporting prevalence rates on perpetrators of bullying.

Furthermore, results from the correlational analysis revealed moderate-to-weak

associations between being a perpetrator and most of the study variables. The

relative substantial correlation between being a perpetrator and being a target of

bullying (r .27) indicates a reasonable degree of overlap between the two, and thus

demonstrates the importance of taking into account target status when predicting

perpetration of bullying (cf. Glomb & Liao, 2003). Except for target status, only role

conflict and interpersonal conflicts showed relationships of any particular magnitude

with being a perpetrator of bullying (Table 1). In line with what has been found in

previous Scandinavian studies (cf. Zapf et al., 2003), no significant differences were

found with regard to hierarchical position of perpetrators (x2 0.03; df 1). In addition, the assumption that age is related to engaging in aggressive behaviour was not

supported in the present study. Hierarchical position and age were thus left out of the

following regression analysis.

The results from the logistic regression analysis showed target status to be the

most important predictor of being a perpetrator of workplace bullying. Being

exposed to bullying occasionally (OR 9.38) and on a weekly basis (OR 11.11), both

proved to be strong predictors of being a perpetrator, as compared to not being

exposed to bullying. The findings also show an increasing probability of being a

perpetrator with intensified frequency of exposure to bullying. In line with previous

findings (cf. Zapf et al., 2003), males were found to have a significantly higher

probability of being a perpetrator of bullying as compared to women (OR 2.09).

Among the situational factors, however, only role conflict (OR 1.37) and

interpersonal conflicts (OR 1.26) were able to significantly predict being a

perpetrator of bullying, although rather weakly (Table 2). Neither decision authority

nor role ambiguity was able to predict being a perpetrator of workplace bullying in

the present sample when taking into account the effects of the other variables. The large overlap between being a target and being a perpetrator of bullying

highlights the need to institute preventive measures against bullying at workplaces.

Regardless of the frequency of the bullying exposure, individuals who were being

exposed to bullying both occasionally and on a weekly basis all showed a substantial

propensity to engage as perpetrators of bullying. Perceptions of unfair treatment

Work & Stress 353

Table 1. Descriptive statistics and correlations.

Mean SD 1 2 3 4 5 6 7 8 9

1. Bullyinga (Perpetrator) 0.03 � � 2. Bullyinga (Target) 0.05 � .27** � 3. Genderb 0.49 � .06** .01 ns � 4. Hierarchical positionc 0.20 � .01 ns �.02 ns .12** � 5. Age 43.66 11.32 �.03 ns .01 ns .01 ns .07** � 6. Decision authority 2.70 0.57 �.07** �.16** .12** .29** .04 ns (.84) 7. Role ambiguity 2.17 0.92 .06** .14** .03 ns �.07** �.09** �.30** (.85) 8. Role conflict 3.12 1.29 .12** .19** .06** .07** �.08** �.27** .35** (.83) 9. Interpersonal conflicts 1.30 0.49 .16** .29** .01 ns .02 ns �.07** �.20** .28** .38** (.75)

Note: a coded 0 (No), 1 (Yes); b coded 0 (Female), 1 (Male); c coded 0 (Subordinate), 1 (Supervisor); Cronbach’s a coefficients are presented on the diagonal in parentheses. **pB.01; ns�non-significant.

3 5

4 L

. J.

H a

u g e

et a

l.

have previously been found to be an important antecedent for engaging in aggressive

behaviour at work (cf. Lee & Brotheridge, 2006; Neuman & Baron, 2003). As being

exposed to bullying is likely to be perceived as unfair by most individuals, there exists

a possibility that individuals will respond in kind to the type of treatment they

receive, thus becoming perpetrators of bullying. Such a line of thinking is also in

accordance with frameworks describing a spiralling effect from relative mild forms of

uncivil behaviour into increasingly more intense and aggressive behaviour (cf.

Andersson & Pearson, 1999; Zapf & Gross, 2001). Although a cross-sectional design

cannot determine whether being a perpetrator is a consequence of being a target of

bullying or vice versa, the fact that one-third of targets at one point or another also

engages in bullying of others indicates an escalating work environment problem in

which an increasing number of individuals take the role of perpetrator or target as

time goes by. Not to forget the severe individual consequences of such exposure,

there is no doubt that organizations should be interested in terminating such

processes at an early stage, regardless of their causes.

The findings further indicate that situational factors relating to role conflict and

interpersonal conflicts at work can instigate tension and frustration in individuals,

which, in turn, may be projected onto others in the work environment (cf. Tedeschi &

Felson, 1994; Thylefors, 1987). Although these are relatively modest contributions,

identifying that being a perpetrator of bullying is significantly related to both

exposure to bullying and experienced stress at work indicates that preventive

measures should be implemented. These should involve a focus on both improving

the overall work environment to avoid individual stress resulting in the bullying of

others, but also on actively managing cases of bullying that may occur. However,

interventions are not likely to be effective by solely focusing on rehabilitation of

targets of bullying and on general work environment improvements. Active steps in

preventing individuals engaging in bullying of others must also be taken. Findings

from research on school bullying may prove such direct action to be the most

effective way to proceed in reducing bullying at work (cf. Olweus, 2003).

Organizations are likely to benefit from developing and implementing sound anti-

bullying policies and practices and taking active steps to prevent interaction among

individuals from escalating into bullying behaviour (cf. Einarsen & Hoel, 2008).

Workplace bullying will only be able to develop within the organizational contexts

that allow such behaviour to take place (Brodsky, 1976).

Table 2. Results of logistic regression analysis.

Dependent variable: Bullying (Perpetrator) 95% CI for OR

Independent variables B SE Wald test Odds ratio Lower CI Upper CI

1. Bullied (Occasionally) 2.24 0.33 45.67 9.38** 4.90 17.94

2. Bullied (Weekly) 2.41 0.68 12.57 11.11** 2.94 42.02

3. Gender 0.74 0.27 7.23 2.09** 1.22 3.57

4. Decision authority �0.03 0.14 0.05 0.97 ns 0.74 1.28 5. Role ambiguity �0.05 0.13 0.16 0.95 ns 0.74 1.22 6. Role conflict 0.32 0.14 4.81 1.37* 1.03 1.82

7. Interpersonal conflicts 0.23 0.11 4.14 1.26* 1.01 1.56

Note: *p B.05; **pB.01;ns�non-significant.

Work & Stress 355

Limitations

Although this study has demonstrated relationships between individual and

situational factors as predictors of engaging in bullying of others in a large and

representative sample, some limitations need to be considered. Relying solely on self-

report methodology is always problematic and probably especially so with regard to

self-report of being a perpetrator of bullying due to the detrimental nature of such

behaviour. Although anonymity is ensured, there exists a significant possibility that

individuals will underreport engagement in such behaviour. Such underreporting

may attenuate correlations between bullying and other variables because some individuals will be less honest in their reporting than others and thus introduce error

in the observed relationships (cf. Spector & Fox, 2005). Triangulation with other

sources of information such as peer-reported behaviour may prove useful in reducing

possible mono-method bias.

Moreover, although rather weak relationships between the situational factors

and being a perpetrator of bullying were identified in the present study, thus

suggesting possible ignorable associations between the factors of interest, one

should keep in mind the strength of relationships obtained from self-labelling single- item measures as compared to multiple-item behavioural checklists in reflecting

different kinds of behaviours to assess engagement in bullying behaviour. In the case

of exposure to workplace bullying, correlations between situational factors and

behavioural multiple-item measures aimed at capturing workplace bullying are

normally of a considerably stronger nature than are correlations with self-labelling

measures (cf. Hauge et al., 2007). Thus, it is likely that the strength of observed

relationships would be of a non-ignorable nature if multiple-item measures were

applied, as is typically the case in studies of counterproductive behaviour towards other individuals at work (cf. Spector & Fox, 2005).

Conclusions

To summarize, we believe that the findings of this study make an important

contribution to research on workplace bullying in identifying factors of importance

for being a perpetrator of bullying at work. Research into this field will probably gain

considerably from bridging the gap that has existed between perpetrator and target-

oriented approaches towards bullying and other related constructs. It is likely that a

sound and thorough integration of these approaches will provide valuable knowledge

in attempting to successfully eliminate bullying from workplaces.

References

Andersson, L.M., & Pearson, C.M. (1999). Tit for tat? The spiraling effect of incivility in the

workplace. Academy of Management Review, 24(3), 452�471. Aquino, K., & Thau, S. (2009). Workplace victimization: Aggression from the target’s

perspective. Annual Review of Psychology, 60, 717�741. Barling, J., Dupré, K.E., & Kelloway, E.K. (2009). Predicting workplace aggression and

violence. Annual Review of Psychology, 60, 671�692. Bowling, N.A., & Beehr, T.A. (2006). Workplace harassment from the victim’s perspective: A

theoretical model and meta-analysis. Journal of Applied Psychology, 91(5), 998�1012. Brodsky, C.M. (1976). The harassed worker. Toronto, ON: Lexington Books.

356 L. J. Hauge et al.

Coyne, I., Chong, P. S.-L., Seigne, E., & 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. De Cuyper, N., Baillien, E., & De Witte, H. (2009). Job insecurity, perceived employability and

targets’ and perpetrators’ experiences of workplace bullying. Work & Stress, 23(3), 206�224. Einarsen, S., & Hoel, H. (2008). Bullying and mistreatment at work: How managers may

prevent and manage such problems. In A. Kinder, R. Hughes & C. L. Cooper (Eds.),

Employee well-being support: A workplace resource (pp. 161�173). Chichester: John Wiley & Sons.

Einarsen, S., Hoel, H., & 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., & Cooper, C. L. (2003). The concept of bullying at work: The

European tradition. In S. Einarsen, H. Hoel, D. Zapf & C.L. Cooper (Eds.), Bullying and

emotional abuse in the workplace. International perspectives in research and practice

(pp. 3�30). London: Taylor & Francis. Glomb, T.M., & Liao, H. (2003). Interpersonal aggression in work groups: Social influence,

reciprocal, and individual effects. Academy of Management Journal, 46(4), 486�496. Hauge, L.J., Skogstad, A., & Einarsen, S. (2007). Relationships between stressful work

environments and bullying: Results of a large representative study. Work & Stress, 21(3),

220�242. Hershcovis, M.S., Turner, N., Barling, J., Arnold, K.A., Dupre, K.E., Inness, M., et al. (2007).

Predicting workplace aggression: A meta-analysis. Journal of Applied Psychology, 92(1),

228�238. Hoel, H., Cooper, C.L., & Faragher, B. (2001). The experience of bullying in Great Britain:

The impact of organizational status. European Journal of Work and Organizational

Psychology, 10(4), 443�465. Inness, M., Barling, J., & Turner, N. (2005). Understanding supervisor-targeted aggression:

A within-person, between-jobs design. Journal of Applied Psychology, 90(4), 731�739. Lee, R.T., & Brotheridge, C.M. (2006). When prey turns predatory: Workplace bullying as a

predictor of counteraggression/bullying, coping, and well-being. European Journal of Work

and Organizational Psychology, 15(3), 352�377. Leymann, H. (1996). The content and development of mobbing at work. European Journal of

Work and Organizational Psychology, 5(2), 165�184. Matthiesen, S.B., & Einarsen, S. (2007). Perpetrators and targets of bullying at work: Role

stress and individual differences. Violence and Victims, 22(6), 735�753. Neuman, J.H., & Baron, R.A. (2003). Social antecedents of bullying: A social interactionist

perspective. In S. Einarsen, H. Hoel, D. Zapf & C.L. Cooper (Eds.), Bullying and emotional

abuse in the workplace. International perspectives in research and practice (pp. 185�202). London: Taylor & Francis.

Olweus, D. (2003). Bully/victim problems in school: Basic facts and an effective intervention

programme. In S. Einarsen, H. Hoel, D. Zapf & C.L. Cooper (Eds.), Bullying and emotional

abuse in the workplace. International perspectives in research and practice (pp. 62�78). London: Taylor & Francis.

Rayner, C. (1997). The incidence of workplace bullying. Journal of …