The impact of workplace bullying on individual wellbeing : The moderating role of coping

The key focus of this study was to explore workplace bullying within the South African context and the key coping methods that bullied individuals utilise to manage bullying interactions. The need to examine bullying has been motivated by research conducted over the past three decades that has indicated that it is a crippling and severe social problem for employees and organisations (Nielsen & Einarsen, 2012). While prevalence rates vary from 2% to 27%, the mental and physical implications for individuals may be so devastating that the organisations in which such bullying is taking place are duty-bound to address them (Nielsen & Einersen, 2012).


Introduction
The key focus of this study was to explore workplace bullying within the South African context and the key coping methods that bullied individuals utilise to manage bullying interactions.The need to examine bullying has been motivated by research conducted over the past three decades that has indicated that it is a crippling and severe social problem for employees and organisations (Nielsen & Einarsen, 2012).While prevalence rates vary from 2% to 27%, the mental and physical implications for individuals may be so devastating that the organisations in which such bullying is taking place are duty-bound to address them (Nielsen & Einersen, 2012).and/or non-verbal communication that is characterised by negative and aggressive elements directed towards the targeted individual.Typical workplace bullying behaviours entail exposure to verbal aggression, physical intimidation, being attacked personally or professionally, having one's work obstructed, being socially isolated from the rest of one's work group, having rumours spread about oneself, or being made the 'laughing stock' by being subjected to verbal or physical acts of humiliation and denigration (Nielsen & Einersen, 2012;Nielsen & Knardahl, 2015).Nielsen and Knardahl (2015) distinguish between targets and victims with those that are targets being exposed to bullying behaviour without necessarily feeling threatened and victimised while those that are victims do feel threatened beyond their ability to cope with and defend themselves against the threat.Workplace bullying is conceptualised to take place relatively often, and over a period of time, and is thus a chronic stressor, with persistent exposure leaving the targeted individual feeling unable to defend himself or herself from the menace of such actions (Einarsen, Hoel, Zapf & Cooper, 2003).However, it has been argued that bullying can be a single event or a few events, the severity of which may be sufficient to ensure that the individual exposed suffers the same extent of negative outcomes as those bullied on a more regular basis (Capponecchia & Wyatt, 2009).Agervold and Mikkelson (2004) note that bullied individuals suffer from impaired psychological well-being, increased levels of anxiety and fear, lowered self-esteem, lowered selfefficacy and lowered belief in their professional competence.Bullying has also been implicated in severe mental health problems such as major depressive disorder, symptomology that resembles post-traumatic stress disorder and even suicide (Rugulies et al., 2012).Physiological outcomes may manifest in sleep disorders and musculoskeletal problems (Hoch, Mikkelsen & Hansen, 2011).Some longitudinal studies have gone so far as to implicate chronic bullying in the development of coronary heart disease (Nielsen, Hetland, Mathhiesen & Einersen, 2012;Notelaeres, Baillien, De Witte, Einersen & Vermunt, 2012;Matthiesen & Einarsen, 2004).Such severe individual implications, in turn, have serious organisational outcomes as victims experience reduced job satisfaction and increased intention to leave the organisation.To the extent that the individual is bullied by co-workers and supervisors, creating a situation in which such relationships become fraught with aggressive interactions, so can this have a negative impact on bullied individual's experience of job satisfaction.This outcome is further exacerbated by victims' perception of a lack of protective conditions provided by the organisation in which bullying is being experienced (Nielsen & Einersen, 2012).In this regard, victims of bullying often report little or no support from the human resource (HR) departments in their organisation, experiencing being pushed from person to person when they complain.Feelings of shame, alienation and possible relocation because of HR, in some instances, supporting the bully instead of the victim are also reported.In such untenable situations, victims often resign from their positions.Bullied individuals may also engage in increased absenteeism because of physical or mental illness or in order to avoid exposure to bullying incidents.They may also show lower organisational commitment, especially in the event of their feeling that the organisation has done nothing or has not done enough to protect them (Balducci, Allfano & Fraccaroli, 2009;Nielsen, Tangen, Idsoe, Matthiesen & Mageroy, 2015).Lower job satisfaction and organisational commitment, and increased absenteeism, in turn, can affect the organisation negatively by reducing employee productivity.Victims may put in less effort because of feelings of inadequacy and continued criticism.Similarly, turnover affects productivity.Loss of production time because of absenteeism and turnover and possible legal costs because of unfair or constructive dismissals can be seen as severe economic costs to the organisation as a whole (Turney, 2003).The list of consequences is thus enormous and the negative effects of bullying serve as further evidence that both individuals and organisations suffer from bullying in the long term.

Coping with workplace bullying
Although a large body of research has demonstrated that bullying has direct effects on individual and organisational outcomes, the coping capacity of the bullied individual may serve to moderate the relationship between bullying and individual and organisational outcomes (D'Cruz & Noronho, 2010).Nielsen and Einersen (2012) note that the consequences of bullying do vary between individuals, with the nature and severity of responses post-exposure being a 'function of the dynamic interplay between event characteristics, individual appraisal and coping processes' (p.314).Coping is defined as ongoing cognitive and behavioural efforts to manage specific demands or threats that are perceived to be taxing or exceeding the resources of the person (Lazarus, 1993;Lazarus & Folkman, 1984).Coping with bullying consists of a target's cognitive and behavioural efforts to master, reduce or tolerate the demands created by stressful bullying interactions (D'Cruz & Noronho, 2010) (The Qualitative Report,15[3] 507-534).Olafsson and Johannsdottir (2004) examined coping specifically in relation to bullying and described four types of coping with bullying, namely, seeking help, assertiveness, avoidance and doing nothing.Seeking help and assertiveness are active strategies, whereas avoidance and doing nothing are passive strategies.Active strategies are engaged with in order to reduce the extent of the actual threat or eliminate it.In this regard, typical active strategies that could be utilised to deal with bullying would be seeking help, which would require the enlistment and assistance of others and assertiveness, in which the individual would rely on his or her own resources in an attempt to resist the bully by 'standing his or her ground'.
Assertiveness thus takes the form of active coping where the individual will attempt to confront their source of stress.Depending on the degree of assertiveness, it may not always be appropriate.In certain instances, being strongly assertive may exceed the bounds of what is perceived of as assertion and may rather be perceived of as aggression.Although in some instances aggressing against the bully may lead to the perpetrator 'backing down', in other instances it may inflame the bully even further, leading to an interaction between the bully and the target that becomes excessively confrontational and thereby destructive (Upton, 2009).Passive strategies would be doing nothing in the hope that the bully or bullies may eventually tire of their behaviour and stop.By doing nothing, 'pretending that you don't care' and that 'nothing is wrong', that is, by not reacting, there is hope that the bully will lose interest and 'go away' (Dehue, Bolman, Vollink & Pouwelse, 2012).An additional passive strategy could involve avoidance (also known as escape).
Although avoidance involves engaging in some form of action, this action is regarded to be passive in that it entails no resistance.While assertiveness, aggression and avoidance are aligned to the 'fight or flight' response to stress (Selye, 1976), avoidance is considered as an 'inactive' behaviour as it involves running away from, rather than confronting, the problem (Dehue et al., 2012).Murray-Harvey, Skrzpiec and Slee (2012) in their examination of bullying amongst students note that certain coping strategies are productive while others are non-productive with regard to the extent to which they serve to ameliorate the negative impact of bullying on individual outcomes.In this regard, they indicate that seeking help is productive, whereas avoidance or doing nothing is non-productive.Similarly, they also note that although assertiveness can be productive, if it becomes excessive it can deteriorate into aggression which may be counterproductive.
In light of the above discussion, it is apparent that there are a number of coping strategies that targeted individuals may utilise.Consequently, the following research questions were proposed: (1) Do different coping styles moderate the relationship between perceived bullying and individual and organisational health and well-being, that is, the dependent variables of psychological well-being, self-esteem, job satisfaction and intention to leave?(2) Does perceived bullying have a direct effect on psychological well-being, self-esteem, job satisfaction and intention to leave?No predictions were made regarding the extent to which certain types may be more or less productive as the research was exploratory in nature.
Based on the literature review and the aforementioned research questions, the researchers of this study established the following hypotheses: • Hypothesis 1a: Perceived bullying would have a direct effect on psychological wellbeing; that is, the greater the perception of bullying the lower the psychological wellbeing.• Hypothesis 1b: Perceived bullying would have a direct effect on self-esteem; that is, the greater the perception of bullying the lower the self-esteem.• Hypothesis 1c: Perceived bullying would have a direct effect on job satisfaction; that is, the greater the perception of bullying the lower the job satisfaction.
• Hypothesis 1d: Perceived bullying would have a direct effect on intention to leave; that is, the greater the perception of bullying the greater the intention to leave.• Hypothesis 2a: Different coping styles (seeking help, assertiveness, avoidance and doing nothing) would moderate the relationship between perceived bullying and psychological wellbeing; that is, higher levels of coping would reduce the negative impact of bullying on psychological wellbeing.• Hypothesis 2b: Different coping styles (seeking help, assertiveness, avoidance and doing nothing) would moderate the relationship between perceived bullying and self-esteem; that is, higher levels of coping would reduce the negative impact of bullying on self-esteem.• Hypothesis 2c: Different coping styles (seeking help, assertiveness, avoidance and doing nothing) would moderate the relationship between perceived bullying and job satisfaction; that is, higher levels of coping would reduce the negative impact of bullying on job satisfaction.• Hypothesis 2d: Different coping styles (seeking help, assertiveness, avoidance and doing nothing) would moderate the relationship between perceived bullying and intention to leave; that is, higher levels of coping would reduce the negative impact of bullying on intention to leave.
In the following, a discussion of the research design and research method is presented, followed by a presentation and discussion of results.The study will conclude with a discussion of its limitations and theoretical and practical implications.

Research approach
A quantitative non-experimental cross-sectional research design was used.Thus, data were gathered at a single point in time.Self-report questionnaires were utilised for data collection.

Research participants
The participants consisted of a convenience sample of 200 white-collar workers employed at a large South African construction company.Of the prospective sample of 200 within head office, 100 responses were received, thus indicating a 50% response rate.Of the respondents, 53 were males and 47 were females.Ages of the participants varied between 22 and 62 years (M = 40 years).Of the participants, 41% were white, 38% were black, 14% were Indian and 7% were mixed-race participants.Of the participants, 41% reported English as their home language, 21% reported Afrikaans as their home language and 36% reported an African language as their home language.The two highest levels of education indicated were matric (37%) and a diploma or certificate (29%).The majority of the sample (65%) indicated that they were married and the number of http://www.sajhrm.co.zaOpen Access years employed in the current organisation ranged from 3 months to 40 years (M = 8 years) (see Table 1a and Table 1b).
Determination of the type of coping strategy used in bullying situations was made using the Coping with Bullying Scale developed by Olafsson and Johannsdottir (2004).The measure consisted of a single question: 'How would you react if you were subjected to bullying in your workplace?' Participants were given options of four coping strategies, each scored on a four-point Likert scale (1 = I would do it; 2 = I would probably do it; 3 = I would probably not do it and 4 = I would never do it).Examples of seeking help are 'tell my boss' and 'tell the HR director about it'.Examples of assertiveness are 'stand my ground and answer back' and 'publicly confront the bully'.Examples of avoidance are 'take sick leave' and 'look for a transfer within the company'.Examples of doing nothing are 'wait it out' and 'hope it stops'.From these items, the four types of coping strategies can be derived: seeking help, assertiveness, avoidance and doing nothing.The reliability of the coping strategies for each subscale in the study was, respectively, 0.70, 0.67, 0.73 and 0.66.
Psychological well-being was measured using the General Health Questionnaire (GHQ) (Goldberg, 1972).This study used the 12-item format.Responses were recorded on a four-point Likert scale that ranges from (1) 'less than usual' to (4) 'much more than usual' (Goldberg, 1972), with examples of items being 'Have you recently lost much sleep over worry?' and 'Have you recently felt you could not overcome your difficulties?'.In this study, an internal reliability of 0.87 was obtained.A high(er) score on this scale indicates poor wellbeing, while a low(er) score indicates good well-being.
Self-esteem was measured using the Self-Esteem at Work Scale (Quinn & Shepard, 1974).The four-item scale comprises items that refer to self-esteem within a job-related context and are bipolar adjectival descriptors, for example, I feel 'not successful'/'successful' separated on a seven-point Likert scale.A high score on the scale represents a high sense of selfesteem, while a low score represents a low sense of self-esteem.
In this study, an internal reliability of 0.76 was obtained.
Job satisfaction was measured using the Overall Job Satisfaction Scale (Warr, Cook & Wall, 1979) Intention to leave one's job was assessed using the Intention to Leave Scale (Lyons, 1972).The scale actually assesses the intention of respondents to stay with their organisation.The three items refer to (1) how long respondents would like to continue working in their present place of employment, (2) whether they would continue to work in their present place of employment if they were given the freedom to choose and (3) whether they would return to their present place of employment if, for some reason, such as ill-health and pregnancy, they had to leave for a period of time.A high or high(er) score on this scale indicates low intention to leave, while a low(er) score indicates increased intention to leave.In this study, an internal reliability of 0.70 was obtained.

Research procedure
Data were collected by means of a hard copy self-administered self-report questionnaire handed out to prospective participants.Permission was granted by the HR director of the construction company.The questionnaire contained a covering letter, biographical information sheet and the measuring instruments outlined above.Participants were informed within the covering letter about the purpose of the study and that participation was confidential and anonymous.All potential participants were informed that their participation was voluntary and that they would not be advantaged or disadvantaged in anyway by choosing to participate or not to participate.Potential participants were informed that by completing and submitting the survey into a sealed box (placed in a conveniently private position within the organisation) they were deemed to be providing consent to participate.However, they were also informed that prior to submitting the survey they could withdraw at any time by not completing or not submitting the survey.Potential participants were also informed that they were not required to provide any identifying information, that no one other than the researcher would have access to their responses and that no single person's response could be identified as the results of the survey would be reported as a summary of general trends.Thus, ethical considerations were taken into account.On completion, respondents were instructed to seal their responses in the unmarked envelope provided and return it to the sealed box placed in a conveniently private location within the organisation's offices.

Statistical analysis
Statistical analysis was conducted by means of the SAS Enterprise Guide 4 program.Cronbach's alpha was used to assess the reliability of the measuring instruments.Descriptive statistics were used to describe the data and Pearson's product-moment correlation coefficients were used to assess the relationships between the variables.Moderated multiple regression (MMR) was used in order to assess the contribution of the independent variable and the moderator variable to the dependent variables.The 0.05 level of significance was selected to determine the presence of significant effects.
Table 2 gives the descriptive statistics, reliabilities and correlations between the variables under study.Pearson's correlations between workplace bullying and the dependent variables were all significant and in the expected direction.
As high scores on the GHQ and on the self-esteem scale indicated poor psychological well-being and poor selfesteem, positive correlations with bullying were expected.High scores on the job satisfaction scale and intention to leave scale indicated high levels of job satisfaction and a low intention to leave, and thus inverse correlations were expected.
Table 3 gives an overview of the findings with regard to bullying or negative acts occurring within the organisation.
In Table 3, 28% of the sample reported experiencing bullying, which ranged in terms of the extent of occurrence from 21% (now and then) to as many as 4% experiencing bullying or negative acts daily or weekly.

Main effects of workplace bullying on the dependent variables
Direct relationships were found on all four of the dependent variables in that the greater the perceived bullying the poorer the psychological well-being, self-esteem and job satisfaction and the greater the intention to leave.As mentioned, the relationships were positive between psychological wellbeing and self-esteem as high scores on these instruments indicate poor psychological well-being and low self-esteem.
Inverse relationships were indicated between job satisfaction and intention to leave as low scores on these instruments indicate low job satisfaction and high intention to leave.
Tables 4 and 5 show the results of the MMR analysis, with workplace bullying (as measured by the NAQ-R) as the independent variable, and seeking help, assertiveness, avoidance and doing nothing (as measured by the Coping with Bullying Scale) as the moderator variables.Psychological well-being, self-esteem, job satisfaction and intention to leave were the dependent variables.
Seeking help moderated the relationship between bullying and psychological well-being and, independent of the level of bullying experienced, had a direct impact on psychological well-being.
The overall r-squared obtained was 0.1934, indicating that 19.34% of the variance in psychological well-being was explained by bullying, seeking help and the interaction between bullying and seeking help.
To determine the contribution of each of these independent variables to the variance in the dependent variables, further regression analyses were conducted for all regressions in which each variable and the interaction term were taken out of the regression in order to determine their individual contribution to the variance.
Results of this procedure indicated that for seeking help bullying on its own added 11.50% to the variance explained, while seeking help increased this variance to 14.50% and the interaction term increased the total variance explained to 19.34%.
No main effect for assertiveness coping on the dependent variables was found; however, assertiveness coping moderated the relationship between bullying and psychological well-being.The overall r-squared obtained was 0.2074, indicating that 20.74% of the variance in psychological well-being was explained by bullying, assertiveness and the interaction between bullying and assertiveness.Bullying on its own added 11.10% to the variance explained, and while assertiveness did not contribute to any variance on its own, the interaction term increased the total variance explained to 20.74%, therefore explaining 9.64% of the variance.
Avoidance had had only one main effect on the dependent variables, namely, on job satisfaction.The overall r-squared obtained was 0.2654, indicating that 26.54% of the variance job satisfaction was explained by bullying, avoidance and the interaction between bullying and avoidance.Bullying on its own added 22.22% to the variance explained, and while the interaction terms did not contribute to any variance, avoidance had a main effect, increasing the total variance explained to 26.54%, thereby contributing 4.32% to the variance.
Furthermore, avoidance moderated the relationship between bullying and psychological well-being and bullying and  self-esteem.For psychological well-being, the overall r-squared obtained was 0.1890, indicating that 18.90% of the variance in psychological well-being was explained by bullying, avoidance and the interaction between bullying and avoidance.Bullying on its own added 15.99% to the variance explained, and while avoidance did not contribute to any variance, the interaction term increased the total variance explained to 18.90%, thereby contributing 2.91% of the variance.
For self-esteem, the overall r-squared obtained was 0.2440, indicating that 24.40% of the variance in psychological wellbeing was explained by bullying, avoidance and the interaction between bullying and avoidance.Bullying on its own added 18.45% to the variance explained and while avoidance did not contribute to any variance, the interaction term increased the total variance explained to 24.40%, thereby adding 5.95% to the variance.
Yet these moderating effects were counterintuitive in that when individuals experienced bullying, avoidance coping worsened psychological well-being and self-esteem.These findings suggest that avoidance coping does not act as a moderator in the expected way in that it does not ameliorate the relationship between bullying, psychological well-being and self esteem.Rather when bullying is experienced, it tends to exacerbate the impact on the well-being and selfesteem.
Doing nothing demonstrated no main effects on the dependent variables; however, in terms of moderating effects, doing nothing as a coping strategy was shown to moderate the relationship between bullying and psychological well-being, again not in the expected direction.As with avoidance, when people experience bullying and utilise the coping strategy of doing nothing, this was related to poorer psychological wellbeing.That is, if one tended to do nothing about the situation, psychological well-being worsened.For psychological wellbeing, the overall r-squared obtained was 0.1801, indicating that 18.01% of the variance in psychological well-being was explained by bullying, doing nothing and the interaction between bullying and doing nothing.Bullying on its own added 15.28% to the variance explained, and while doing did not contribute to any variance, the interaction term increased the total variance explained to 18.01%, thereby contributing 2.73% of the variance.

Discussion
The aim of this study was to investigate whether different coping styles moderated the relationship between perceived bullying and individual and organisational health and wellbeing (i.e. the dependent variables of psychological wellbeing, self-esteem, job satisfaction and intention to leave) and whether perceived bullying had a direct effect on psychological well-being, self-esteem, job satisfaction and intention to leave.Results indicated that the hypotheses were largely supported in that bullying had a deleterious impact on all the outcome variables.However, coping did not always moderate the relationships examined.In addition, when moderation was observed, these effects were not always in the expected direction.
The main effects of the workplace bullying on the dependent variables were found to be consistent with results from previous research which has demonstrated that bullying impairs psychological well-being, erodes self-esteem and has deleterious organisational outcomes.For example, Nielsen and Einersen (2012) in a meta-analyses of 66 independent studies (N= 77 721) indicated that workplace bullying manifested in outcomes of poor psychological well-being, that is, anxiety, depression, post-traumatic stress, burnout, negative core self-evaluations and physical health problems.Furthermore, they indicated that across these studies there was reported increased intention to leave and absenteeism, lowered job satisfaction and organisational commitment, and lowered job performance and productivity.
Similarly, the main effect of seeking help on psychological well-being was found to be consistent with results from the literature, more particularly the literature on social support that indicates that people who have access to help and resources tend to have improved well-being, irrespective of whether stress is experienced (Lakey & Orehek, 2011;Schirey, 2004;Schroevers, Helgeson, Sanderman & Ranchor, 2009;Schwarzer & Guiererez-Dona, 2005;Taylor, 2007;Taylor, Sherman, Jarcho, Takagi & Dunagan, 2004;Zachariah, 2009).
Seeking help also aligns with the Conservation of Resources Model proposed by Hobfoll, Freedy, Lane and Geller (1990) within which targets may find their personal resources deplete when they are compelled to deal with the unwanted aggression of bullies (Lee & Brotheridge, 2006).In such an instance, targets may seek to bolster their personal resources by drawing on available social resources, that is, by seeking help and social support from others.Within the context of the bullying literature, seeking help is also noted as an effective strategy to deal with bullying.For example, Murray-Harvey et al. (2012) explored ratings by professionals of ineffective and effective coping strategies and the degree to which students utilise these strategies.Seeking help was deemed to be the most productive 'other-focused strategy' by professionals, that is, by researchers, teachers and counsellors who had worked with bullied adolescents.
Assertiveness as a moderator between bullying and wellbeing is also supported within the bullying literature.Murray-Harvey et al. (2012) note that assertiveness is an effective strategy in that it enables individuals to 'take control of the situation' and facilitates the development of resilience within the individual if utilised successfully.Moreno-Jimenez et al. (2007) report similar moderating effects for assertiveness in their study of Latin-American immigrants working within Spanish organisations.
With regard to avoidance and doing nothing, while avoidance moderated between bullying and well-being and bullying and self-esteem, and doing nothing moderated between bullying and well-being; the direction of these effects was counterintuitive in that avoidance and doing nothing had a deleterious impact on self-esteem and wellbeing.Thus, while seeking help and assertiveness were constructive coping strategies, avoidance and doing nothing were not so.These counterintuitive findings for avoidance and doing nothing are also noted by Dehue et al. (2012).In their study of 361 Dutch employees, they indicated that bullied individuals who engaged in avoidant coping strategies, denial and doing nothing suffered from a high degree of health-related problems and poor psychological well-being.More specifically, they noted that targets using such strategies suffered from poor psychological well-being (as measured by the GHQ), increased depressive symptoms (as measured by Beck's Depression Inventory), increased physical health complaints (as measured by the VOEG, that is, the Dutch Physical Health Questionnaire) and engaged in the highest degree of absenteeism.Thus, they similarly reported that these coping strategies were not efficacious in that although they moderated the bullying-wellbeing relationship, they were detrimental, exacerbating the impact of bullying on well-being outcomes.Similar findings for the negative effects of lack of assertiveness and avoidance were reported by Zapf (1999) andJiminez et al. (2007).Both these studies noted that evasive and negatory coping styles of avoidance and doing nothing in the face of bullying were nonproductive strategies that could exacerbate the impact of bullying on well-being.Salin, Tenhiala, Roberge and Berdahl (2014) and Dijkstra, De Dreu, Evers and Van Dierendonck (2009) also note that those who engage in passive strategies of avoidance, denial and doing nothing, enduring their bullying in silence, suffer the greatest amount of psychological strain and impairment to self-esteem.
However, while seeking help and assertiveness moderated the relationship between bullying and psychological wellbeing in a beneficial way, these strategies only contributed a small proportion to the variance in well-being.This may be because of a number of factors, namely, organisational structure and power differentials and HR responsiveness and HR policies and procedures.
With regard to organisational structure and power differentials, these factors could have limited the extent to which individuals felt capacitated to seek help or be assertive (D'Cruz & Noronho, 2010;Olafsson & Johannsdottir, 2004).Because of power differentials, assertiveness and seeking help may not always be possible and may compel and account for individuals engaging in avoidance and doing nothing (Salin et al., 2014).Power differentials are of special importance in that how the target chooses to react will depend on whether he or she is being bullied vertically or horizontally.While vertical bullying describes situations in which an individual may be bullied by subordinates (vertical-upward), the vast majority of vertical bullying occurs when a subordinate is bullied by a superior (Cunniff & Mostert, 2012).According to Namie (2000), almost 80% of the time bullies are in superior positions to the victim and, therefore, the bullying is vertical-downward.Horizontal bullying refers to bullying by colleagues.Within South African research, Cunniff and Mostert (2012) note that while it is twice as likely individuals will be bullied by superiors as opposed to colleagues, the prevalence of horizontal bullying is still high.In instances of vertical-downward bullying, the power differential is often considered to be too high and as such this is likely to curtail the individuals' choice of coping.Thus, when power differentials are high, assertiveness and seeking help may be less likely, while avoidance or simply doing nothing are far more probable.Salin et al. (2014) refer to the capacity to exercise 'voice' against bullying, noting that those who are in positions of greater power and status are more likely to exercise strategies of voice, that is, assertion or advocacy, as opposed to those with less power who are more likely to self-silence and engage in avoidance, denial or doing nothing.Furthermore, they note that power differentials relate to fears of retaliation with regard to choosing to use active voice strategies as opposed to passive strategies.Those who have greater power have less concern about retaliation and feel more assured that seeking help and advocacy will elicit the required degree of support for their cause.Those who have less power may be afraid that using voice as 'going up against' the bully may be too risky in that it could lead to increased aggression on the part of the bully or censure from organisational members and superiors if they 'side' with the more powerful aggressor.In the construction company under study, there was a strict line of authority and control which may have encouraged perceived high power differentials.In such organisations with their emphasis on hierarchical control of employees and downward pressures to gain compliance and increased productivity, the 'perfect storm' is created in which there is a climate of fear and thereby within which bullying can flourish (Beale & Hoel, 2011).
With horizontal bullying, when one is bullied by a peer, power differentials may be reduced, and more active coping strategies may be utilised.However, if the bully is supported by other members within the bullied individual's cohort, such active strategies may be curtailed.To the extent that an individual is 'ganged up on' by a group of his or her peers, so will the power differentials be increased and so may it be more likely that the target will rather avoid or do nothing.However, if the bullied individual can gather around him or her a cohort of supportive individuals against the bully, so will power differentials be reduced.By seeking help in this manner, the targeted individual has not only utilised the seeking help strategy but may be emboldened to be more assertive in actual interactions with the bully.Turning to sympathetic co-workers for support in order to enhance one's power base through a 'collective voice of resistance' may thus serve to reduce the power differential and enables the individual to engage in a more assertive coping strategy (Branch, Ramsay & Barker, 2013, p. 288).
It can thus be seen that power differentials can affect the nature of the relationship between the bully and the bullied and can, in turn, influence what strategy is chosen.As both the source of bullying and the numbers of bullies by which the individual was targeted were not established in this research, if high power differentials were indeed evident, this could have accounted for the (smaller) contribution to the variance in outcomes of the moderating effects of assertiveness and seeking help.
With regard to HR policies and procedures and HR responsiveness, the degree to which HR was perceived to be responsive to and effective in dealing with bullying grievances could also have determined the extent to which targets would seek help from this resource.If individuals perceived that HR would not be responsive to dealing with complaints, so would it have been unlikely that individuals would seek help from this 'resource'.Although they may still have sought out social support from significant others, in the event of being bullied, reporting the matter to the HR department may not have been perceived of as a viable option.In this regard, D 'Cruz et al. (2010) distinguish between organisations that adopt 'soft' or 'hard' HRM practices.Within hard HRM practice, 'utilitarian instrumentalism' is stressed with employees being seen as mere headcount whose purpose is to support business objectives and who are managed in an impersonal and rational, if not exploitative, manner.Soft HRM is aligned more with 'developmental humanism', which while still associated with business objectives sees employees more as valued assets.Although D'Cruz et al. (2010) note that often soft HRM is 'spoken' but not necessarily 'practiced', it is likely that organisations that adopt a soft HRM approach might demonstrate more concern in assisting bullied employees as compared to those that are more aligned to hard HRM practices.Furthermore, it is likely that in organisations that adopt hard HRM, the structure and leadership of the organisation along with culture and climate would be more authoritarian and task-centred as opposed to those that adopt soft HRM who may be more likely to lean towards leadership and practices that are more employeecentred.As the policies and procedures of HRM in terms of hard or soft practices were not assessed within the present research, it could not be ascertained whether the manner of dealing with matters of bullying would be more responsive and therefore whether targets felt able and willing to utilise this resource for help.
Consequently, the possible inhibiting effect of power differentials and non-responsive HR policies and procedures within the organisation with regard to dealing with bullying may have left targeted individuals with no other options but to utilise strategies of avoiding or doing nothing, both of which were shown to exacerbate the bullying-well-being relationship.Counterintuitive findings for avoidance and doing nothing may also be explained by the masculine culture and climate of the organisation.Loosemore and Waters (2004) note that the organisational culture and climate in the construction industry is traditionally male-dominated, demonstrating a culture of competitive, confrontational practices and high levels of conflict.Similarly, Lingard, Brown, Bradley, Bailey and Townsend (2007) note that the construction industry is regarded to be a tough, masculine work environment characterised by long work hours and following the traditional work patterns of gender assumptions, that is, construction is 'men's work'.Consequently, the coping styles of avoidance and doing nothing are styles that are likely to be less efficacious within this culture and climate.Not fighting back or 'running away', particularly in an environment where one is expected to 'man-up', may have far more deleterious consequences in that such an approach may be perceived of as weak and cowardly.Thus, the inverse effects of such coping styles on selected outcomes within the present sample were not entirely surprising.Engaging in behaviours that may be perceived of as 'unmanly', particularly for males, could negatively affect one's sense of self-esteem and overall psychological well-being (Schippers, 2007).Thus, those who engaged in the coping response of avoidance and doing nothing may have found that their sense of well-being and self-esteem eroded in the face of a humiliating inability to fight back.Such negative self-evaluations may also have, in turn, been more likely to leave provoke their desire to leave their employment (Agervold & Mikkelson, 2004).

Limitations of the study
Although the focus of this study was on the moderating role of coping styles in the bullying-well-being relationship, it had a number of limitations.Most pertinent of these limitations was the sample size.Although the number of participants in the study was adequate for the analysis conducted, the sample size was small.In addition, the sample examined was obtained from within a construction organisation and, as such, the culture within this organisation may have influenced the type of coping strategy used.Furthermore, the cross-sectional nature of the research would have precluded a determination of where, in the lifecycle of bullying, participants were situated.Another limitation is the exclusive use of self-report questionnaires.Research recommends the use of both questionnaires and personal interviews (Pietersen, 2007) in order to gain more in-depth accounts of bullying experiences from respondents.A further limitation may pertain to common method variance.As the data were collected utilising self-report measures, the relationship between variables may have been biased by common method variance.Although, it has been noted that common method variance may be more likely to deflate interaction terms, therefore some confidence can be placed in the few interactions found (Siemsen, Roth & Oliveria, 2010).

Recommendations
From a research perspective, this research has discussed and highlighted a number of factors that still remain to be studied in order to better understand how they contribute to bullying and the life cycle of the bullying process.These pertain to organisational climate and culture, organisational structure, power differentials, HR policies and procedures and HR responsiveness.Research has also indicated that demographic factors such as race, gender and education along with personality predispositions may play a role in coping strategies utilised (Branch et al., 2013;Cunniff & Mostert, 2013;Murray-Harvey et al., 2012;Nielsen & Knardahl, 2015;Salin et al., 2014).As the main focus of this study was to assess the efficacy of four coping styles on the bullying-well-being relationship at particular point in time, attention was not focused on addressing these factors and, consequently, their role in affecting the coping strategy that an individual will feel capacitated to or more likely to utilise still needs to be examined.Furthermore, it has been recommended that longitudinal research should be conducted as such research may be able to examine the unfolding of the bullying process over time.Thomas et al., 2013).Primary interventions would entail introducing legislation, policies and procedures that will assist management in reducing the occurrence and recurrence of bullying incidents in the organisation, as well as overtly expressing that bullying in the workplace will not be tolerated (Djurkovic, McCormack & Casimir, 2006;Fox & Stallworth, 2005).Secondary interventions would entail training of all stakeholders: management, employees and trade union representatives, sensitising them to the values and behaviour that are expected as well as the company culture that the organisation endeavours to uphold in order to limit bullying behaviour in the workplace (Pearson, Andersson & Porath, 2000;Pietersen, 2005).
In addition, certain personality traits, for example, being high on aggression and being overly sensitive in terms of the need to protect and conserve one's own self-esteem and position from any perceived threat, are regarded as perpetrator traits that can lead to bullying behaviour.Thus, ensuring that individuals who have the potential to clash in terms of personality traits are not expected to work too closely with one another may be another avenue to limit bullying.However, this may not always be possible.Therefore, training should try to ensure that potentially aggressive individuals are better able to manage this personality trait within the context of their interactions with others within the organisation.By creating awareness that aggression will not be tolerated and by enhancing the communication styles of such individuals, the potential for their bullying behaviour in the future may be curtailed.
In the event of bullying occurring, in spite of these interventions it is advocated that the organisation goes beyond mediation, adopting a punitive approach towards the bully in terms of applied sanctions underpinned by the organisation's legislative framework.A legislative framework, which outlines a formal grievance procedure and the sanctions to be imposed when bullying occurs, facilitates an organisational culture in which bullying is not tolerated and in which values of civility, respect and engagement as opposed to aggression and excessive competitiveness are emphasised if not enforced (Dao et al., 2006;Lingard et al., 2007;Richards & Daley, 2003).Employee assistance programmes that provide coaching and counselling for both targets and bullies could also be utilised (Branch et al., 2013).These programmes could be used to facilitate the development of behavioural and emotional skills to better manage interpersonal interactions.For the bully, reduction of aggressive interactions that are lacking in civility and respect would be targeted, whilst for the victim skills would entail how to manage bullying interactions both currently and in the future if they occur.In addition, for victims, support may need to extend long after the bullying has actually ceased.Lutgen-Sandvik (2008) notes that even when there is intervention and there is a cessation of bullying, the aftermath of the bullying can take months and even years for individuals to overcome.He notes that the post-bullying phase is one of intense grieving in which individuals have to struggle to rebuild themselves and their sense of loss, the latter revolving around their loss of reputation, loss of self-confidence and feelings of lack of justice and fairness in terms of how they were treated up until the time the intervention to stop the bullying was enacted.Investment in and the adoption of such anti-bullying initiatives could go a long way in fostering a culture in which there is a high degree of PSC and POS (Cooper-Thomas et al., 2013).In this manner, a proactive health and safety approach to bullying is adopted as opposed to a reactive complaints-based approach only (Branch et al., 2013).Within this context, if and when individuals are bullied, they will be far more inclined to engage in coping mechanisms of seeking help.In addition, within the context of a rights-based organisation in which acceptable behaviours are clearly outlined and reinforced and unacceptable behaviours are clearly outlined and sanctioned, victims may be far more comfortable in asserting themselves.Thus, more efficacious forms of coping may be fostered as opposed to maladaptive forms such as avoidance and doing nothing, which individuals engage in when they feel they have no other options.Research also needs to be directed towards examining the efficacy of interventions (Branch et al., 2013).In this regard, Branch et al. (2013) note that very little empirical research has been conducted evaluating bullying interventions and this remains a critical area in which future research must be undertaken.

Conclusion
In conclusion, this research has added to the body of literature on workplace bullying within the South African context by noting that certain types of coping may be efficacious whilst other types may have a maladaptive effect, thereby promoting the vicious cycle of bullying and its aftermath even further.
The results presented here are based on an empirical study of a sample of employees within the construction industry.However, as mentioned before, there is a need for further research to be conducted across a broader range of industries, including the examination of additional variables within the bullying process in order to enhance the understanding of bullying and its implications for both individuals and organisations.

TABLE 1a :
Demographic characteristics of participants.

TABLE 1b :
Demographic characteristics of participants.

TABLE 3 :
Reported experience of negative acts in the workplace compared to reported experience of being bullied in the workplace.

TABLE 4 :
Regression of psychological well-being and self-esteem onto bullying, seeking help, avoidance, assertiveness and doing nothing.

TABLE 5 :
Regression job satisfaction and intention to leave onto bullying, seeking help, avoidance, assertiveness and doing nothing.