Bullying in a large Danish municipality:
The moderating role of social support
Ann-Kristina Løkke & Rune Bysted
Abstract
The objective of the paper is to investigate how social support from colleagues and supervisors moderates the outcome of bullying in the form of job satisfaction and psychological distress among 4,162 Danish municipal employees.
Partial Least Square regression is used as statistical technique.
5.3 per cent of the municipal respondents have been bullied in the past year. Bullied employees are less satisfied with their job and more worn out, emotionally exhausted, testy, and stressed than employees who have not experienced bullying. Employees bul- lied by supervisors are less satisfied with their jobs, with emotional demands, influ- ence, and support from supervisors, than employees bullied by colleagues.
Unexpectedly, the social support of supervisors and colleagues does not moderate the negative relationships of bullying on job satisfaction and psychological distress.
1. Introduction
Bullying is a problem in many European countries (Eurofound, 2012). It can take various forms such as verbal hostility, social isolation, highly stressful changes in tasks and work situation, being given unreasonable deadlines, etc. (Einarsen & Nielsen, 2015).
The impact of a poor work environment on bullying was identified early in the bully- ing literature (e.g., Leymann, 1996), and has also been empirically supported (e.g., Bail- lien, et al., 2011; Van den Broeck et al., 2011). This paper focuses on the importance of social support in the workplace for overcoming the negative consequences of bullying.
Workplace bullying has a number of negative outcomes. Previous longitudinal studies confirm that bullying has a negative effect on job satisfaction (Quine, 1999; Rodríguez- Muñoz et al., 2009) and psychological distress (e.g., Einarsen & Nielsen, 2015; Finne, et al., 2011; Rugulies et al., 2012). Mental health problems are not only a consequence of workplace bullying in the short term. They also occur many years after the bullying
has stopped (Einarsen & Nielsen, 2015). This supports the need to address the prob- lems for researchers and practitioners.
In the EU directive and in the legal frameworks of many European countries, it is stated that employers are obliged to provide work conditions that secure the health and safety of their employees (Milczarek, 2010); Denmark is no exception. In the last decade, bullying has received ever-increasing attention in Denmark (e.g., Conway et al., 2016; Gullander et al., 2014; Hansen et al., 2011; Høgh, 2013; Rugulies et al., 2012).
The increased research focus on bullying is particularly a result of an amendment of the Health and Safety Legislation in 2000. Here it is specified that supervisors are re- sponsible for ensuring “that the work does not involve a risk of physical or psycholog- ical health as a result of bullying” [Executive order on work environment, sect. 9]).
Supervisors therefore play a crucial role in relation to bullying. Management quality is crucial to the occurence of bullying, no matter whether managers are directly involved or not (Høgh, 2013). Hence a central aspect of bullying is that it will only take place if the offender feels his/her bullying behaviour is deemed legitimate by his/her supervi- sor (Einarsen, 1999).
In spite of its significance, only limited research has focused on how supervisors and colleagues can help overcome some of the negative effects of bullying; and none of them focus on municipalities. One study shows that support from supervisors and colleagues is related to lower psychological strain from bullying (Gardner et al., 2013).
Another study shows that social support and satisfaction with social relationships at work can moderate the correlation between bullying and health complaints (Einarsen et al., 1996).
Given the limitations in existing literature, it remains unclear whether social support can moderate the negative aspects of bullying in a municipality. Thus, the primary objective of our study is to investigate how social support from colleagues and super- visors moderates the outcome of bullying in terms of job satisfaction and psychologi- cal distress in a municipality. Another purpose of the study is to test the correlation between bullying and job satisfaction and psychological distress respectively, as well as between job satisfaction and psychological distress.
The estimated model, including a number of psychosocial work characteristics as con- trol variables, is depicted in Figure 1:
Figure 1: Model for the moderating role of social support on consequences of bullying in a municipality
Job sasfacon
Internal social support from:
● Supervisor
● Colleagues
Control variables:
● Job demands
● Emoonal demands
● Influence
Bullying Psychological
distress
This study also provides a methodological contribution to the existing literature. Salin and Hoel (2011) suggest that future research should focus on multivariate analyses to investigate the relative strength of bullying and other job stressors. Additionally, many previous studies on bullying are based on questionnaires, and it is not uncommon to see rather low response rates of 30-50 per cent (e.g., Einarsen et al., 1994; Finne et al., 2011; Hodgins, 2014; Hoel & Cooper, 2000; Niedl, 1996), with some exceptions with higher rates (Rugulies et al., 2012). The low response rates might be explained by the fact that bullying is perceived as a sensitive topic. However, some of the conclusions are questionable thus increasing the need for further large-scale studies on the subject.
Our contribution responds to this by demonstrating a multivariate setup based on a survey in a large Danish municipality with a response rate of 74.3 per cent. We know that bullying is a complex phenomenon resulting from a complex of individual, or- ganisational, and cultural factors (Hodgins, 2014). In applying a case study approach, like the one presented here, we hope to reduce some variation caused by individuals belonging to different organizations and thus being subject to, for instance, various common personnel policies, including initiatives to prevent bullying.
1.1. The Danish municipal context of this study
The context of this study is interesting for several reasons. First of all, research shows that Danish public employees demonstrate a general concern for serving public inter- ests (Andersen & Kjeldsen, 2013). Consequently, many public employees have a high degree of personal involvement in their job, which often makes them more vulnerable
to personal attacks (Zapf et al., 2011). Overall, this suggests that the feeling of being bullied might be high among public employees.
Another aspect worth mentioning is that the Danish public sector is under pressure due to more intense competition, organisational changes, and increased measuring of efficiency and control of performance (Olesen et al., 2008). All of these might imply a generally higher level of bullying in the public work environment. Such an increas- ing demand for efficiency and downsizing may contribute to an escalation of stress, frustration, and insecurity (Salin, 2001), which is a breeding ground for co-worker bullying behaviour.
In contrast to the above, the public sector is also characterized by a high sense of community and strong collectivism (Boyne, 2002; Hartley, 2005). If employees share a stronger collectivistic feeling, they might bully each other less and support each other more when bullying takes place. Findings support that bullying is lower in the public sector than in the private (Einarsen & Skogstad, 1996), maybe due to more competi- tion and lower job security in the private sector.
Altogether, the public sector, including municipalities, shares certain characteristics that might influence bullying behaviour and the possibility of providing social sup- port to moderate the negative outcome of bullying. This makes it interesting to study further.
1.2. The role of social support
Only a small body of research has focused on how social support can moderate the negative consequences of bullying; all findings point in the same direction.
One study among 1,733 employees from 36 organizations from the education, health, hospitality and travel sectors from Aotearoa/New Zealand shows that social support from colleagues and supervisors is related to lower psychological strain from bullying (Gardner et al., 2013).
Another study among 2,215 Norwegian union members from public and private organizations shows that social support and satisfaction with social relationships at work can moderate the correlation between bullying and health complaints (Einarsen et al., 1996). This buffer effect of social support is here explained as it simply makes the individual feel less vulnerable, because he or she knows where to find support if necessary (Einarsen et al., 1996).
A study among 335 Australian school teachers from government and non-government high schools shows a relation between perceived organizational support and lower intention to leave the organization (Djurkovic et al., 2008).
A qualitative study among Danish victims of bullying shows that one of the most commonly employed coping strategies is to seek support among colleagues and, in some cases, their superiors; the victims did, however, not succeed in putting an end to the bullying (Mikkelsen, 2004).
Another Danish study shows that witnesses to bullying react according to whether they categorise the victims’ behaviour as within the social norms of the workplace or not (Bloch, 2013). In the first category, witnesses react by showing sympathy and in- terest in the person by listening to her/his problems, offering advice etc.; in the other category, they ignore the victim and slander him or her (Bloch, 2013). Thus, it might be difficult for bullied employees to find support at the workplace, which can result in isolation in the organization.
The above leads to the following four hypotheses to be tested:
H1: Bullied employees experiencing social support from colleagues will have a higher level of job satisfaction than those who do not experience social support.
H2: Bullied employees experiencing social support from colleagues will have a lower level of psychological distress than those who do not experience social support.
H3: Bullied employees experiencing social support from a supervisor will have a higher level of job satisfaction than those who do not experience social support.
H4: Bullied employees experiencing social support from a supervisor will have a lower level of psychological distress than those who do not experience social support.
1.3. Other relationships in the model
A number of longitudinal studies show that bullying is related to low job satisfac- tion (Quine, 1999; Rodríguez-Muñoz et al., 2009). Findings confirm that bullying in one time period predicts job satisfaction in the following period, which supports that bullying should be seen as a cause rather than a consequence of job satisfaction (Rodríguez-Muñoz et al., 2009).
Previous studies also confirm that bullying leads to increased psychological distress (Einarsen & Nielsen, 2015; Finne et al., 2011; Gullander et al., 2014; Kivimäki et al., 2003; M.B. Nielsen et al., 2014; Rugulies et al., 2012; Verkuil et al., 2015).
Job satisfaction is correlated with reduced psychological distress and poor mental health (Amati, Tomasetti et al., 2010; Nagai, Tsuchiya et al., 2007).
Social support at work might have a direct positive relation to job satisfaction (Cor- tese et al., 2010; Lewig & Dollard, 2003; Macklin et al., 2006) as well as to health and psychological distress (e.g., Bech et al., 2002; Godin & Kittel, 2004; Whitehall, 2004), although this is not always confirmed (Macklin et al., 2006).
The estimated model will be controlled for job demands, emotional demands, and influence which have proven to be in line with job satisfaction (e.g., Lewig & Dol- lard, 2003; Macklin et al., 2006), and psychological distress (Barnett & Brennan, 1995;
Godin & Kittel, 2004; Karasek & Theorell, 1990; Macklin et al., 2006). Social support is included for model completeness.
2. Methodology
2.1. Participants and study design
Data used in this cross-sectional study is based on a yearly electronic employee survey in a Danish municipality; this one was conducted October to December 2012. The target population of the survey is all employees in the municipality. The number of re- spondents in the survey is 4,162 individuals. It was distributed to all employees, i.e., a total of 5,672 individuals, leading to a response rate of 73.4 per cent. Employees were guaranteed anonomity in the survey.
2.2. Measurements
All variables used in the analysis are assessed based on the medium length version of the Copenhagen Psychosocial Questionnaire II, COPSOQ II (National Research Center for the Working Environment, 2005; Pejtersen et al., 2010). The COPSOQ II was devel- oped by the Danish National Research Centre for the Working Environment and has proven to be a reliable and valid tool for measuring psychosocial work environments (Bjorner & Pejtersen, 2010; Thorsen & Bjorner, 2010). A selection of questions from the COPSOQ II is included in the survey and used to describe each construct.
The study definition of bullying: “When a person is pestered regularly and over a long period of time, or is repeatedly and seriously exposed to offensive actions that he/she perceives as hurtful or degrading. These infringing acts, however, only be- come bullying when the victim is unable to defend him/herself against them. Teasing which both the ones involved regard as friendly does not classify as bullying. Neither do solitary conflicts.” This definition resembles other definitions applied (e.g. Einarsen et al., 2011; Ortega et al., 2009; Pejtersen et al., 2010).
Respondents are subsequently asked the following question: “Have you been exposed to bullying at work during the last 12 months?” The question is answered on a scale with the options ‘yes, daily’; ‘yes, weekly’; ‘yes, monthly’; ‘yes, sometimes’; and ‘no’.
This is in accordance with the COPSOQ II (National Research Center for the Working Environment, 2005; Pejtersen et al., 2010). For the purpose of estimating the model, bullying is dichotomized by grouping the response score ‘no’ into 0 and the other scores into 1; this is a previously used method (e.g., Stapelfeldt et al., 2013).
Participants are also asked to indicate whether they had been subjected to bullying by colleagues, a superior, subordinates, and/or clients/patients/customers. For this ques- tion, they could choose all options.
The method described is a common technique applied when asking about bullying (cf.
e.g., Ortega et al., 2009).
Psychological distress is measured by the respondents’ experienced symptoms using four questions, i.e., how often have you felt worn out, been emotionally exhausted, been testy, or been stressed? These four questions are answered on a scale with the fol- lowing options (all the time, much of the time, part of the time, occasionally, not at all).
Job satisfaction is measured by the item: “How pleased are you with your job as a whole, everything considered? This question is answered on a scale with the options:
very satisfied, satisfied, unsatisfied, very unsatisfied.
Social support from supervisors is measured by the following two questions: “How often is your immediate superior willing to listen to your problems at work?” and
“How often do you get help and support from your immediate superior?” Both ques- tions are answered on a scale with the following options: always, often, sometimes, seldom, never/hardly ever.
Social support from colleagues is measured by three questions: “How often do you get help and support from your colleagues? How often are your colleagues willing to listen to your problems at work? How often do your colleagues tell you how well you carry out your work?” All three questions are answered on a scale with these options:
always, often, sometimes, seldom, never/hardly ever.
Job demands are measured by the following three questions focusing on quantitative job demands: “Do you get behind in your work? Is it necessary to keep working at a high pace? Is your work pace high throughout the day?” All questions are answered on a scale with these options: to a very large extent, to a large extent, somewhat, to a small extent, to a very small extent.
Emotional demands are measured by two questions: “Does your work put you in an emotionally disturbing situation?” and “Do you have to relate to other people’s per- sonal problems as part of your work?” The questions are answered on a scale with the options: always, often, sometimes, seldom, never/hardly ever.
Influence is measured by two questions: “Do you have a large degree of influence on your work?” and “Can you influence the amount of work assigned to you?” The scales of both questions comprise: always, often, sometimes, seldom, never/hardly ever.
2.3. Statistical procedure
As the aim of this study is to estimate a model of latent variables, i.e. job satisfaction and psychological distress, Partial Least Square (PLS) regression is an appropriate method (Jöreskog & Wold, 1982).
Furthermore, the model contains both reflective latent constructs (job demands, emo- tional demands, influence, and social support from supervisors and colleagues) and formative structures (bullying, job satisfaction, and psychological distress). In addi- tion we include the moderating effects of social support. The implication is that the model cannot be handled by traditional structural equation modelling (Gefen et al., 2000). SmartPLS is used to estimate the model (Ringle et al., 2005).
A number of measures can be used to evaluate the fit of a PLS model. Usually the measurement model and the structural model are assessed sequentially (Hulland, 1999). The measurement model is assessed through item reliability, convergent valid- ity, and discriminant validity (Hulland, 1999). Item reliability, assessed by examining the loadings of the individual items with their respective construct, is high (> 0.67).
Convergent validity is measured by composite reliability and Cronbach Alpha, and both should preferably exceed 0.7 (Hulland, 1999). In this study composite reliability is between 0.83 and 0.93, and Cronbach Alpha is between 0.64 and 0.87.
The discriminant validity represents the extent to which measures of a given con- struct differ from measures of other constructs in the model. One way of examining this is to evaluate the root average variance extracted, i.e., the level of variance inter- nally in the constructs (calculated for each of the constructs along the diagonal cf. ta- ble 1). For adequate discriminant validity, the diagonal elements should be higher than the off-diagonal elements in the corresponding rows and columns (Hulland, 1999). In this case, the root AVE lies between (0.796 and 0.931) and is higher than the shared variance across the constructs. Thus discriminant validity is high.
Regarding the structural model there is no proper overall goodness-of-fit measure for the PLS model. Reporting R square values for the dependent constructs is, however, recommended (Hulland, 1999). The R square for job satisfaction is 24.6 percent, and for psychological distress it is 35.3 percent, which is acceptable. The conclusion is that the model fulfils the requirements for PLS models (Hulland, 1999). The following results are therefore statistically valid.
Table 1: Average Variance Extracted (AVE), correlation and Cronbach Alpha
(1) (2) (3) (4) (5) (6) (7) (8)
Influence (1) ,863*
Bullying (2) -,077
Emotional demands (3) ,133 -,097 ,841*
Psychological distress (4) ,228 -,119 ,433
Job satisfaction (5) ,197 -,084 ,187 ,251
Social support from supervisors (6) ,295 -,140 ,193 ,278 ,265 ,931*
Social support from colleagues (7) ,151 -,165 ,025 ,157 ,176 ,292 ,822*
Job demands (8) ,251 -,043 ,340 ,395 ,166 ,174 ,080 ,796*
AVE ,745 ,708 ,867 ,676 ,634
Cronbach alpha ,635 ,650 ,664 ,778 ,639
*Root AVE
3. Results
3.1. Level of bullying
Descriptive results show that 5.3 per cent of the respondents in the municipality have been bullied in the past year. For comparison, a meta-analysis of the impact of meth- odological moderators on prevalence rates of workplace bullying estimated the preva- lence rate at 4.6 per cent in Scandinavia (Nielsen et al., 2010) when applying the same self-labelling method combined with a definition of bullying. This indicates that the bullying level is only slightly higher in this particular municipality. We will elaborate more on this in the discussion section. Figures shows that 12 per cent of bullied work- ers experience bullying daily or weekly, 8 per cent experience bullying monthly, and 80 per cent only sometimes.
Bullying predominantly occurs from colleagues. Thus 55.8 per cent are bullied by col- leageus, whereas 9.7 per cent are bullied by supervisors, 1.8 per cent by subordinates, and 16.1 per cent by clients/patients/customers. A total of 8.8 per cent are bullied by both supervisor and colleagues, and other combinations also exist as shown in table 2.
Table 2: Bullied employees distributed on type of offenders
Bullied % (absolute number)
Supervisor 9.7 % (21)
Colleagues 55.8 % (121)
Subordinates 1.8 % (4)
Clients/patients/customers 16.1 % (35)
Supervisor and colleagues 8.8 %(19)
Colleagues and subordinates 0.9 % (2)
Colleagues and clients/patients/customers 6.9 % (15)
Total 100% (217)
3.2. Differences between bullied and non-bullied employees
As shown in table 3, bullied employees are less satisfied with their jobs, and they are more worn out, emotionally exhausted, stressed, and testy than employees who have not experienced bullying in the last 12 months. The mean differences between bullied and non-bullied individuals are all significant at p < 0.001.
Table 3: Mean differences in job satisfaction and psychological distress between bullied and non- bullied employees (standard deviation in brackets)
Job satisfaction Felt worn outa) Emotionally
exhausteda) Stresseda) Testya)
Bullied 3.07 (0.56) 3.24 (0.89) 3.34 (0.97) 3.44 (0.97) 3.50 (0.88)
Non-bullied 3.36 (0.58) 3.60 (0.83) 3.81 (0.87) 3.74 (0.87) 3.89 (0.75)
a) All are significant at p<0.001
b) Note, a high score is a positive evaluation, i.e., subjects are less distressed
There is also a difference in the assessment of the psychosocial work characteristics between bullied and non-bullied employees. As can be seen from table 4, non-bullied employees evaluate job demands, emotional demands, influence, and support from su- pervisors and colleagues more positively than bullied employees. All mean differences between bullied and non-bullied employees are significant at p < 0.001.
Table 4: Mean differences in psychosocial work characteristics between bullied and non-bullied employees (standard deviation in brackets)
Job demands Emotional demands
Influence Social support from supervisor
Social support from colleagues
Bullied 2.62 (0.66) 2.58 (0.86) 2.95 (0.86) 3.44 (1.0) 3.47 (0.8)
Non-bullied 2.76 (0.64) 2.94 (0.9) 3.25 (0.82) 4.04 (0.9) 3.96 (0.68)
All are significant at p<0.001
3.3. Differences between employees bullied by supervisor and by colleagues
As shown in table 5, employees who are bullied by their supervisor are less satisfied with their jobs and more worn out than employees who are bullied by colleagues (p
< 0.05). There are no significant differences among the two groups regarding being emotionally exhausted, stressed, and testy.
Table 5: Mean differences in job satisfaction and psychological distress between bullied by supervisor and bullied by colleagues (standard deviation in brackets)
Job satisfactiona)
Felt worn outa)
Emotionally exhausted
Stressed Testy
Bullied by supervisor 2.77 (0.53) 2.86 (0.83) 3.09 (0.68) 3.14 (1.04) 3.73 (0.83) Bullied by colleageus 3.10 (0.57) 3.39 (0.92) 3.38 (0.99) 3.48 (0.96) 3.48 (0.87) a) Significant at p<0.05
b) Note, a high score is a positive evaluation for the psychological distress items , i.e., subjects are less distressed
Assessment of the psychosocial work characteristics differ between employees bullied by the supervisor and those bullied by colleagues. Table 6 shows that employees being bullied by supervisor are significantly less satisfied with emotional demands, influ- ence, and support from supervisors than employees being bullied by colleagues. There are no significant differences in the evaluation of job demands and social support from colleagues between the two groups.
Table 6: Mean differences in psychosocial work characteristics between bullied by supervisor and bullied by colleagues (standard deviation in brackets)
Job demands Emotional demandsb)
InfluenceC) Social support from
supervisora)
Social support from
colleagues Bullied by supervisor 2.49 (0.58) 2.14 (0.59) 2.59 (0.85) 2.34 (1.07) 3.56 (0.98) Bullied by colleagues 2.67 (0.65) 2.67 (0.85) 2.95 (0.86) 3.59 (0.83) 3.35 (0.73) a) Significant at p<0.001, b) Significant at p<0.01, c) Significant at p<0.1
3.4. The moderating role of social support
Table 7 presents the standardized coefficients of the estimated model for municipal employee job satisfaction and psychological distress. The coefficients express the ef- fect of a change in the predicting variable on the depending variable, ceteris paribus.
Table 7: Standardized coefficients in the estimated model
Job satisfaction (1)low-high(4)
Psychological distress 1(high)-low(5) Control variables
Job demands (1) high – low (5) .082*** .214***
Emotional demands (1) high – low (5) .143*** .300***
Influence (1) low – high (5) .176*** .037**
Social support from supervisors (1) low – high (5) .256*** .081***
Social support from colleagues (1) low – high (5) .126*** .046***
Bullying (bullied=1) -.028* -.035**
Job satisfaction (1) low – high (4) .214***
Hypothesized effects
Social support from supervisors x Bullying -.012 .006
Social support from colleagues x Bullying -.009 -.005
R2 .246 .353
*p < .10; **p < .01; ***p < .001
The results indicate that no moderation on the negative outcome of bullying takes place from social support neither from colleagues nor supervisors, i.e., the hypotheses H1-H4 are not supported.
Regarding the other associations in the tested model, there are significant correlations between bullying and lower job satisfaction (-.028) and higher psychological distress, i.e., lower psychological well-being (-.035).
As table 7 also shows, social support from supervisors and colleagues is related to higher job satisfaction by (.256) and (.126) respectively. Furthermore, social support from supervisors and colleagues is associated with lower psychological distress by (.081) and (.046).
High job satisfaction is also connected to lower psychological distress (.214).
Regarding the control variables, low job demand and emotional demand are associated with higher job satisfaction by (.082) and (.143) respectively as well as lower psycho- logical distress by (.235) and (.300), just as expected. Furthermore, higher job influence is correlated with higher job satisfaction (.176) and lower psychological distress (.037).
4. Discussion
This study shows that bullying mainly occurs from colleagues, as suggested in pre- vious findings (Ortega et al., 2009). This supports the argument that bullying is an internal organisational matter.
The investigation could not identify any moderating role of social support from col- leagues and supervisors on the negative consequences of bullying. This finding is sur- prising as previous studies confirm such a relationship (Einarsen et al., 1996; Gardner et al., 2013).
The findings also showed that bullied employees experienced a lower level of social support than non-bullied employees. This may indicate that bullied employees find themselves more isolated in the organisation than non-bullied. This isolation might come about as a consequence of colleagues who are turned against the bullied employ- ees by the offender (Gardner et al., 2013). Or colleagues ignore the bullied employee because her/his behaviour is categorised as outside the social norms of the workplace (Bloch, 2013). Or the isolation arises due to fear of becoming a target oneself (Gardner et al., 2013). Thus, this study’s findings could not support the previous study which found that social support reduces employees’ feeling of vulnerability as a consequence of knowing where to find support if necessary (Einarsen et al., 1996).
Supervisors’ supporting role did not seem to have any influence on the relationship between bullying and job satisfaction and psychological distress respectively. This was unexpected as a weak and absent supervisor was seen as the explanation for bullying in other public sector organisations (Strandmark & Hallberg, 2007; Vartia, 1996).
Our analysis also shows that employees bullied by their supervisor report less satis- faction with, for instance, emotional demands and influence compared to those bul- lied by colleagues. Furthermore, they have a lower general job satisfaction compared to those bullied by colleagues This indicates that the supervisor might behave in a harassing way in other areas as well. In some countries this can be explained by the fact that it is difficult to lay off employees in the public sector and bullying might be used to get rid of unwanted staff (Salin, 2001). The Danish labour market, however, is based on the well-known “Flexicurity” concept, combining flexibility in hiring and fir- ing with security in employment and income. So even though some supervisors bully employees to get rid of them, it may not be that common in Denmark compared to other countries. Still, there might be a larger resistance in the public compared to the private sector against laying off employees; this is part of being socially responsible.
These issues require further investigation, especially studies that compare public and private organizations.
The fact that employees bullied by their supervisor are less satisfied with emotional demands and influence compared to those bullied by colleagues might be explained by the fact that supervisors have the power to increase job demands, reduce influence, and in other respects hamper the work climate.
Overall, the study shows that bullying take place in the public sector. Bullying is more or less at the same level as shown in other studies in Scandinavia using the same defi- nitions of bullying (Nielsen et al., 2010). Further studies need to be conducted to iden- tify whether the reason for bullying is context dependent. Previous empirical findings cannot shed light on this because results point in different directions; two studies show that bullying is more pronounced in the public sector (Hodgins, 2014; Salin, 2001), whereas other findings show the opposite (Mikkelsen & Einarsen, 2001). There is, of course, also a possibility of selection bias, i.e., those who have been exposed to bullying to a high degree have left the organization. This influences both the level of bullying and possibly also some of the other results.
This study also shows that bullied employees evaluate their psychosocial work char- acteristics more negatively than their non-bullied counterparts, just like previous findings show (Agervold & Mikkelsen, 2004). One explanation for this might be that bullied employees project their dissatisfaction and sadness over being bullied to other parts of the psychosocial work environment. In other words, being bullied might be so physically draining that it rubs off on the assessment of other work conditions.
Turning to other findings in the model than the moderating effect of social support, this study demonstrates a relationship between bullying and psychosocial distress just like in previous studies (e.g., Finne et al., 2011; Gullander et al., 2014) as well as between bul- lying and job satisfaction (Quine, 1999; Rodríguez-Muñoz et al., 2009); The correlations are, however, very low. Compared to the other psychosocial work characteristics in the model, bullying has a very limited association to the outcomes. Thus, the results indicate that a work environment characterized by bullying might impair municipal employee well-being and satisfaction on the job. However, dissatisfaction with demands, influence and social support might impair even more.
4.1. Strengths and limitations
The strength of this study is the very large number of participants and the high response rate, which increase the reliability of the statistical conclusions. Previous research has often been based on different workplaces which harms inferences about the work environment for bullied and non-bullied employees. Therefore, the strength of this study is the investigation of bullying and its consequences within a single, relatively homogeneous municipality. This might eliminate some variation regarding being subject to various common staff policies in different organizations which may have an impact on bullying and other psychosocial work characteristics.
A major limitation of the study is that it relies solely on self-reported cross-sectional
often applied method (c.f. e.g., Agervold & Mikkelsen, 2004; Cooper-Thomas et al., 2013; Einarsen et al., 1994; Ortega et al., 2009; Quine, 1999; Vartia, 2001). Yet, future research based on longitudinal data in municipalities is needed to evaluate the causal effects and the generalizability of the study results.
In this study bullying is assessed using a self-labelling method. One advantage of this method is high face validity because it explicitly asks respondents whether they have been exposed to bullying; furthermore construct validity is high because respondents are presented with a precise and easy-to-understand theoretical definition of bullying (Nielsen et al., 2011). The method has, however, also some disadvantages. One of them is that it is a subjective approach where respondents might have different thresholds for labelling themselves as bullied (Nielsen et al., 2011). This subjectivity might lead to an underestimation of the level of bullying. In this study a definition was provided, hence subjectivity bias is less problematic (Nielsen et al., 2011). There is, however, no guarantee that respondents actually read the definition or do not use their own defini- tions when answering (Nielsen et al., 2011). As respondents have to label themselves as victims, there is an unavoidable risk that some refuse to do so because they are ashamed (Nielsen et al., 2011).
There is a possibility that individuals with health problems prior to any experience of bullying are more prone to report themselves as targets of bullying. A further possibil- ity is that individuals with health problems might simply lack the ability to cope with a bullying situation (Hoel & Cooper, 2000).
Another limitation of the present study is that it lacks the possibility of controlling for any personal characteristics since these are not part of the dataset. The respondents are, however, employed in the same Danish municipality which reduces some varia- tion. Including personal characteristics in future research would still be helpful.
The present study should therefore be considered a first step in the study of the mod- erating role of social support on the negative consequences of bullying on municipal employee job satisfaction and psychological distress.
4.2. Theoretical and managerial implications
At the theoretical level this study primarily contributes by showing that social support from colleagues and supervisors does not moderate the negative outcome of being bul- lied among municipal sector employees. This is an important contribution since the limited literature in this area shows the opposite. This study’s findings indicate that resources outside the organisation are needed to re-establish the balance in a munici- pality when bullying has taken place.
Another contribution to literature is that the consequences of being bullied by a supervisor differ from the consequences of being bullied by colleagues. This has not been studied before.
The paper is a first step in explaining bullying in the municipal sector. Further re- search is needed to get a deeper knowledge about the reasons for the bullying behav- iour in the sector and especially to understand why social support does not moderate the negative consequences of bullying in the municipality; this in order to see if it is a general condition in this type of public organization or it is anchored locally.
The managerial implications of the study are first of all that managers have to include a third party to re-establish the balance in the organisation as a consequence of bully- ing, as social support from colleagues and supervisors is not enough.
Another managerial implication of this study is that it confirms managers’ possibility of influencing job satisfaction and psychological distress, not only through psycho- social work characteristics, but indeed through anti-bullying initiatives. Initiatives to prevent and monitor bullying might take the form of written policies, information, bullying surveys, training, and statistical recording of cases of bullying (Salin, 2008).
Anti-bullying policies must also include procedures of how to ensure that the policy will be implemented and evaluated and also what the sanctions of bullying will be (Høgh, 2013). Another relevant initiative for supervisors that has proven successful in avoiding bullying is to perform a transformational, authentic, and constructive leadership style (Cooper-Thomas et al., 2013; Nielsen, 2013) besides creating a good and stress free work climate (Hoel & Cooper, 2000). Managers must build their com- petencies on how to prevent and handle bullying, they must be role models for good behaviour (Høgh, 2013). Furthermore, the organization must focus on the general tone and a perhaps negative culture (Høgh, 2013).
To implement one or more of these initiatives, managers must allocate the necessary resources.
A more general implication for the public sector, including municipalities, is related to the fact that part of the bullying behaviour from supervisors may emanate from the desire to get rid of unwanted employees, and is thus a consequence of the resistance to laying off employees in this sector. Further research is needed to confirm this, but if true, then it might affect policy implications regarding whether this is an appropri- ate aspect of this sector or not.
5. Conclusion
This study could not find a moderating role of social support from colleagues and supervisors on the negative consequences of bullying, and this shows that it is impor- tant to include resources outside the organisation to re-establish the power balance when bullying occurs.
This study shows that bullying is a problem among municipal employees. However, psychosocial work characteristics might impair municipal employee well-being and satisfaction on the job more than bullying. Further research is needed to get a deeper knowledge of explanations for municipal employee bullying and its moderating effects.
References
Agervold, M., & Mikkelsen, E.G. (2004). Relationships between bullying, psychosocial work environment and individual stress reactions. Work & Stress, 18(4), 336-351.
Amati, M., Tomasetti, M., Ciuccareli, M., Mariotti, L., Tarquini, L.M., Bracci, M. & Santarelli, L. (2010).
Relationship of job satisfaction, psychological distress and stress-related biological parameters among healthy nurses: a longitudinal study. Journal of Occupational Health, 52, 31-38.
Andersen, L.B., & Kjeldsen, A.M. (2013). Public service motivation, user orientation, and job satisfaction:
a question of employment sector? International Public Management Journal, 16(2), 252-274.
Baillien, E., Rodriguez-Muñoz, A., Van den Broeck, A., & De Witte, H. (2011). Do demands and resources affect target’s and perpetrators’ reports of workplace bullying? A two-wave cross-lagged study. Work & Stress, 25(2), 128-146.
Barnett, R.C., & Brennan, R.T. (1995). The relationship between job experience and psychological distress:
a structural equation approach. Journal of Organizational Behavior, 16(3), 259-276.
Bech, P., Andersen, M.B., Tønnesen, S., & Agnarsdóttir, E. (2002). Stress hos ledere i Danmark - årsager, udbredelse og konsekvenser (In Danish). Arbejdsmiljørådets Service Center.
Bjorner, J.B., & Pejtersen, J.H. (2010). Evaluating construct validity of the second version of the Copenhagen Psychosocial Questionnaire through analysis of the differential item functioning and differential item effect. Scandinavian Journal of Public Health, 38(3), 90-105.
Bloch, C. (2013). Vidner som aktører i mobning (In Danish). In N.C.M. Nickelsen (Ed.), Arbejdslivets skyggesider (pp. 217-236). Aarhus: Klim.
Boyne, G.A. (2002). Public and private management: What’s the difference? Journal of Management Studies, 39(1), 97-122.
Conway, P.M., Clausen, T., Hansen, Å.M., & Høegh, A. (2016). Workplace bullying and sickness presenteeism:
cross-sectional and prospective associations in a 2-year follow-up study. International Archives of Occupational & Environmental Health, 89, 103-114.
Cooper-Thomas, H., Gardner, D., O’Driscoll, M., Catley, B., Bentley, T., & Trenberth, L. (2013). Neutralizing workplace bullying: the buffering effects of contextual factors. Journal of Managerial Psychology, 28(4), 384-407.
Cortese, C.G., Colombo, L., & Ghislieri, C. (2010). Determinants of nurses’ job satisfaction: the role of work- family conflict, job demand, emotional charge and social support. Journal of Nursing Management, 18, 35-43.
Djurkovic, N., McCormack, D., & Casimir, G. (2008). Workplace bullying and intention to leave: the moderating effect of perceived organisational support. Human Resource Management Journal, 18(4), 405-422.
Einarsen, S. (1999). The nature and causes of bullying at work. International Journal of Manpower, 20(1/2), 16-27.
Einarsen, S., Hoel, H., Zapf, D., & Cooper, C. (2011). The concept of bullying and harassment at work: the European tradition. In S. Einarsen, H. Hoel, D. Zapf & C. Cooper (Eds.), Bullying and harassment in the
workplace: Developments in theory, research, and practice (pp. 3-41). Boca Raton: CRC Press Taylor &
Francis Group.
Einarsen, S., & Nielsen, M.B. (2015). Workplace bullying as an antecedent of mental health problems: a five- year prospective and representative study. International Archives of Occupational & Environmental Health [Epub ahead of print], 88(2), 131-142. doi: DOI 10.1007/s00420-014-0944-7
Einarsen, S., Raknes, B.I., Matthiesen, S.B., & Hellesøy, O.H. (1996). Helsemessige aspekter ved mobbing i arbeidslivet - modererende effekter av sosial støtte og personlighet (In Norwegian). Nordisk Psykologi, 48(2), 116-137.
Einarsen, S., Raknes, I., & Matthiesen, S.B. (1994). Bullying and harassment at work and their relationships to work environment quality: an exploratory study. European Work and Organizational Psychologist, 4(4), 381-401.
Einarsen, S., & Skogstad, A. (1996). Bullying at work: epidemiological findings in public and private organiza- tions. European Journal of Work and Organizational Psychology, 5(2), 185-201.
Eurofound. (2012). Fifth European working conditions survey. Dublin: European Foundation for the improve- ment of Living and Working Conditions.
Finne, L.B., Knardahl, S., & Lau, B. (2011). Workplace bullying and mental distress – a prospective study of Norwegian employees. Scandinavian Journal of Work, Environment & Health, 37(4), 276-287.
Gardner, D., Bentley, T., Catley, B., Cooper-Thomas, H., O’Driscoll, M., & Trenberth, L. (2013). Ethnicity, work- place bullying, social support and psychological strain in Aotearoa/New Zealand. New Zealand Journal of Psychology, 42(2), 84-91.
Gefen, D., Straub, D.W., & Boudreau, M.-C. (2000). Structural equation modeling and regression: guidelines for research practice. Communications of Association for Information Systems, 4(7), 1-79.
Godin, I., & Kittel, F. (2004). Differential economic stability and psychosocial stress at work: associations with psychosomatic complaint and absenteeism. Social Science & Medicine, 58, 1543-1553.
Gullander, M., Hogh, A., M., H.Å., Persson, R., Rugulies, R., Kolstad, H.A., . . . Bonde, J.P. (2014). Exposure to workplace bullying and risk of depression. Journal of Occupational and Environmental Medicine, 56(12), 1258-1265.
Hansen, Å. M., Hogh, A., & Persson, R. (2011). Frequency of bullying at work, physiological response, and mental health. Journal of Psychosomatic Research, 70, 19-27.
Hartley, J. (2005). Innovation in governance and public services: past and present. Public Money and Manage- ment, 25(1), 27-34.
Hodgins, M. (2014). Workplace mistreatment: health, working environment and social and economic factors.
Health, 6(5), 392-403.
Hoel, H., & Cooper, C.L. (2000). Destructive conflict and bullying at work. Manchester: University of Manchester Institute of Science and Technology (UMIST).
Hulland, J. (1999). Use of partial least squares (pls) in strategic management research: a review of four recent studies. Strategic Management Journal, 20, 195-204.
Høgh, A. (2013). Ledelse, mobning og negativ adfærd på arbejdspladsen. In N.C.M. Nickelsen (Ed.), Arbejd- slivets skyggesider (In Danish) (1 ed., pp. 237-256). Aarhus: Klim.
Jöreskog, K.G., & Wold, H. (1982). The ML and PLS techniques for modeling with latent variables. In H. Wold (Ed.), Systems under undirect observations (pp. 263-270). New York: North-Holland.
Karasek, R., & Theorell, T. (1990). Healthy work: stress, productivity and the reconstruction of working life. New York: Basic Books.
Kivimäki, M., Virtanen, M., Vartia, M., Elovainio, M., Vahtera, J., & Keltikangas-Järvinen, L. (2003). Workplace bullying and the risk of cardiovascular disease and depression. Occupational and Environmental Medicine, 60(10), 779-783.
Lewig, K.A., & Dollard, M.F. (2003). Emotional dissonnance, emotional exhaustion and job satisfaction in call centre workers. European Journal of Work and Organizational Psychology, 12(4), 366-392.
Leymann, H. (1996). The content and development of mobbing at work. European Journal of Work and Organi- zational Psychology, 5, 735-753.
Macklin, D.S., Smith, L.A., & Dollard, M.F. (2006). Public and private sector work stress: workers compensa- tion, levels of distress and job satisfaction, and the demand-control-support model. Australian Journal of Psychology, 58(3), 130-143.
Mikkelsen, E.G. (2004). Coping with bullying at work: Results from an interview study. Paper presented at the The Fourth International Conference on Bullying and Harrasment in the Workplace, Bergen, Norway.
Mikkelsen, E.G., & Einarsen, S. (2001). Bullying in Danish work-life: prevalence and health correlates. European Journal of Work and Organizational Psychology, 10(4), 393-413.
Milczarek, M. (2010). Workplace violence and harassment: a European picture. https://osha.europa.eu/en/
publications/reports/violence-harassment-TERO09010ENC
Nagai, M., Tsuchiya, K.J., Toulopoulou, T., & Takei, N. (2007). Poor mental health associated with job dissatis- faction among school teachers in Japan. Journal of Occupational Health, 49, 515-522.
National Research Center for the Working Environment. (2005). The Copenhagen Psychosocial Questionnaire II - medium version. http://www.arbejdsmiljoforskning.dk/~/media/Spoergeskemaer/copsoq/uk/copsoq-ii- medium-size-questionnaire-english.pdf#
Niedl, K. (1996). Mobbing and well-being: economic and personnel development implications. European Jour- nal of Work and Organizational Psychology, 5(2), 239-249.
Nielsen, M.B. (2013). Bullying in work groups: The impact of leadership. Scandinavian Journal of Psychology, 54, 127-136.
Nielsen, M.B., Magerøy, N., Gjerstad, J., & Einarsen, S. (2014). Workplace bullying and subsequent health problems. The Journal for the Norwegian Medical Association, 134(12-13), 1233-1238.
Nielsen, M.B., Matthiesen, S.B., & Einarsen, S. (2010). The impact of methodological moderators on preva- lence rates of workplace bullying. A meta-analysis. Joumal of Occupational and Organizational Psychology, 83, 955-979.
Nielsen, M.B., Notelaers, G., & Einarsen, S. (2011). Measuring exposure to workplace bullying. In S. Einarsen, H. Hoel, D. Zapf & C. Cooper (Eds.), Bullying and harassment in the workplace (pp. 149-174). Boca Raton:
CRC Press Taylor & Francis Group.
Olesen, K.G., Thoft, E., Hasle, P., & Kristensen, T.S. (2008). Virksomhedens sociale kapital - hvidbog (In Dan- ish). http://www.arbejdsmiljoforskning.dk/~/media/Boeger-og-rapporter/hvidbog-socialkapital.pdf Ortega, A., Høgh, A., & Pejtersen, J.H. (2009). Prevalence of workplace bullying and risk groups: a representa-
tive population study. International Archives of Occupational Environmental Health, 82, 417-426.
Pejtersen, J.H., Kristensen, T.S., Borg, W., & Bjorner, J.B. (2010). The second version of the Copenhagen Psy- chosocial Questionnaire. Scandinavian Journal of Public Health, 38(3), 8-24.
Quine, L. (1999). Workplace bullying in NHS community trust: staff questionnaire survey. British Medical Jour- nal, 318, 228-232.
Ringle, C.M., Wende, S., & Will, S. (2005). SmartPLS 2.0 (M3) Hamburg: Beta.
Rodríguez-Muñoz, A., Baillien, E., De Witte, H., Moreno-Jiménez, B., & Pastor, J.C. (2009). Cross-lagged rela- tionships between workplace bullying, job satisfaction and engagement: Two longitudinal studies. Work &
Stress: An International Journal of Work, Health & Organisations, 23(3), 225-243.
Rugulies, R., Madsen, I.E.H., Hjarsbech, P.U., Hogh, A., Borg, V., Carneiro, I.G., & Aust, B. (2012). Bullying at work and onset of a major depressive episode among Danish female eldercare workers. Scandinavian Journal of Work Environment and Health, 38(3), 218-227.
Salin, D. (2001). Prevalence and forms of bullying among business professionals: A comparison of two different strategies for measuring bullying. European Journal of Work and Organizational Psychology, 10(4), 425-441.
Salin, D. (2008). The prevention of workplace bullying as a question of human resource management: meas- ures adopted and underlying organizational factors. Scandinavian Journal of Management, 24, 221-231.
Salin, D., & Hoel, H. (2011). Organisational causes of workplace bullying. In S. Einarsen, H. Hoel, D. Zapf & C.
Cooper (Eds.), Bullying and harassment in the workplace: Developments in theory, research, and practice (2 ed., pp. 227-243). London: Taylor & Francis.
Stapelfeldt, C.M., Nielsen, C.V., Andersen, N. ., Krane, L., Fleten, N., Borg, V., & Jensen, C. (2013). Are envi- ronmental characteristics in the municipal eldercare, more closely associated with frequent short sick leave spells among employees than with total sick leave: a cross-sectional study. BMC Public Health, 13, 578.
Strandmark, M.K., & Hallberg, L.R.-M. (2007). The origin of workplace bullying: experiences from the perspec- tive of bully victims in the public service sector. Journal of Nursing Management, 15, 332-341.
Thorsen, S.V., & Bjorner, J.B. (2010). Reliability of the Copenhagen Psychosocial Questionnaire. Scandinavian Journal of Public Health, 38(3), 25-32.
Van den Broeck, A., Baillien, E., & De Witte, H. (2011). Workplace bullying: A perspective from the job demands resources model. SA Journal of Industrial Psychology, 37, 12., 37(2), 12.
Vartia, M. (1996). The sources of bullying - psychological work environment and organizational climate. Euro- pean Journal of Work and Organizational Psychology, 5(2), 203-214.
Vartia, M. (2001). Consequences of workplace bullying with respect to the well-being of its targets and the observers of bullying. Scandinavian Journal of Work, Environment & Health, 27(1), 63-69.
Verkuil, B., Atasayi, S., & Molendijk, M.L. (2015). Workplace bullying and mental health: a meta-analysis on cross-sectional and longitudinal data. PloS one, 10(8).
Whitehall. (2004). Work stress and health: the Whitehall ll study. Retrieved 02 April, 2013 from http://www.ucl.
ac.uk/whitehallII/pdf/Whitehallbooklet_1_.pdf
Zapf, D., Escartín, J., Einarsen, S., Hoel, H., & Vartia, M. (2011). Empirical findings on prevalence and risk groups of bullying in the workplace. In S. Einarsen, H. Hoel, D. Zapf & C. Cooper (Eds.), Bullying and harass- ment in the workplace. Boca Raton: CRC Press.
