Insecure Attachment Styles and Bullying Perpetrators in the Workplace: An Exploratory Study

Purpose: It has been widely debated over the past two decades that workplace bullying harms both victims and organizations equall y. However, the perpetrator’s perspective on bullying is an inadequately researched area that has been highlighted by many researchers. This study aims to explore the predictive effect of an insecure attachment (avoidant attachment style and anxious attachment style) on the bullying behavior of perpetrators in the workplace. Methodology: Data is collected from different organizations in Pakistan for this cross-sectional study. The sample size is 203. For the analysis of hypotheses, the SEM model of SmartPLS is used. Findings : The results indicate that an avoidant attachment style positively affects bullying behavior in the workplace. Unlike the anxiety attachment style which has no significant relationship with workplace bullying behavior. Practical implications: These findings highlight the importance of secure attachment for healthy interpersonal relationships and the role of insecure attachment in emerging bullying behaviors in the workplace.


INTRODUCTION
Over the last few decades, bullying at work has been acknowledged as a source of interpersonal stress in the work setting (Samnani & Singh, 2016;Thomas, Connor, & Scott, 2018). "The Harassed Worker," a significant book by Brodsky (1976) highlighted this issue, later, in 1990, Leymann published a journal paper on this issue. Alternative names have been used for it in literature such as social deterioration (Duffy, Ganster, & Pagon, 2002), emotional abuse (Keashly, 1998) ostracism (Williams, 2007), harassment (Nielsen, Glasø, & Einarsen, 2017), incivility (Cortina, Magley, Williams, & Langhout, 2001) deviation in interpersonal relationships (Berry, Ones, & Sackett, 2007) supervisory abuse (Tepper, 2007) and bullying (Einarsen, Hoel, Zapf, & Cooper, 2011). Workplace bullying is a particular and distinct form of aggression at the workplace. It is primarily repeated, hostility and power inequity Hoprekstad, Hetland, & Einarsen, 2021). This includes insolence, fright, isolation and negatively influencing someone's work activities frequently and over time (Einarsen, Hoel, Zapf, & Cooper, 2003;Mikkelsen, Hansen, Persson, Byrgesen, & Hogh, 2020). A meta-analysis revealed that workplace bullying is widespread worldwide, with a prevalence rate of about 15%. However, this figure varies geographically and demographically (Nielsen, Matthiesen, & Einarsen, 2010;Nielsen. & Einarsen, 2018). The social and management sciences have thoroughly studied this phenomenon. The social scientists focused on bullying's victims and their findings indicated that workplace bullying has a negative impact on the psychological and physical health of targets, including self-destructive thoughts (Nielsen, Hoel, Zapf, & Einarsen, 2016), sleep disturbance (Hansen, Hogh, Garde, & Persson, 2014) neck stiffness (Tynes, Johannessen, & Sterud, 2013), chronic daily headache (Kääriä, Laaksonen, Rahkonen, Lahelma, & Leino-Arjas, 2012) and fibromyalgia (Kivimaki et al., 2004). On the other hand, the emphasis of management sciences studies remained mainly on the financial success and losses of organizations in relation to bullying behavior in the workplace (Bano & Malik, 2013;Einarsen, Hoel, Zapf, & Cooper, 2020;Nielsen & Knardahl, 2015). The previous literature highlights and thoroughly validates the occurrence of workplace bullying and the damage due to which the victims and the organizations both suffer. It further, illuminates the critical knowledge gap from the perspective of the perpetrator of bullying. This gap must be filled in order to gain a better understanding of the phenomenon and develop effective individual and organizational intervention strategies. This gap is termed a "black hole" by Rayner and Cooper (2003) because perpetrators have received less direct attention. It was deemed by Aquino and Thau (2009) that the least quantity of research was available on this phenomenon. Nielsen and Einarsen (2018) point out two gaps in this arena, firstly, the lack of research on workplace bullies, and secondly, the overreliance on single-source data. The current studies examine the perpetrators' traits to comprehend why some people behave in a hostile manner (Baillien, Rodriguez-Muñoz, Van den Broeck, & De Witte, 2011;Einarsen et al., 2020;.
In Pakistan, a limited number of studies are conducted to investigate the occurrence of workplace bullying in various sectors. However, the perspective of perpetrators of bullying is either unnoticed or rarely studied. Keeping the scenario in mind, this study aimed to raise awareness of the issue by focusing on the insecure attachment styles (avoidant and anxious attachment styles) of bullying perpetrators. This research may help in comprehending the nature of workplace perpetrators' behavior by examining their attachment preferences. The following objectives are examined in the present study: 1. To investigate the relationship between avoidant attachment style and workplace bullying. 2. To examine the link between anxious attachment style and workplace bullying. The individual-dispositions hypothesis suggests that personal characteristics like attachment styles and personality traits are possible precursors of bullying (Einarsen et al., 2020;.

Attachment styles and Bullying
A careful observation of the literature review shows that few researchers have investigated the process by which people become bullies in work place. On the other hand, the literature supports the idea that early attachment, internal working models of the individual and social interactions in general, may set the tone for the aggressive behavior of an individual. Internal working models are self and other cognitive schemas that make beliefs about other's people expectations and suitable behavior based on those expectations (Fraley, 2019;Shaver, Collins, & Clark, 1996). They function consciously or unconsciously, assisting in the organization of previous and future social interactions in a non-observable fashion (Bretherton & Munholland, 2008;Keizer, Helmerhorst, & van Rijn-van Gelderen, 2019). As emotions are fully loaded, so, working models consist of more than just cool beliefs (Bretherton & Munholland, 2008). Bowlby (1982) argued that the ability to create relationships is innate, and necessary for existence. The innate behavioral system of the individual constantly evaluates demands for attachment. When the child's needs are addressed in an empathetic, loving and attentive way, the infant develops a consistent and reliable relationship and believes that they are beloved and special. Such an individual expresses feelings and behaviors that indicate their relationship is satisfied and safe, they trust others, low in anger, receptive to social contacts and are assured of the caregiver's comfort and protection. On the other hand, if needs are not satisfied and the caregiver is careless, inaccessible, unresponsive, unpredictable or intimidating, the infant exhibits fear and anxiety, a lack of confidence and trust and a desire to avoid the relationship (Bretherton & Munholland, 2008;Hazan & Shaver, 2017). These relationships create emotional, cognitive and behavioral patterns resulting in long-term individual alterations in attachment styles. Interaction in important relationships is influenced by internal working models of self and others (Hazan & Shaver, 2017;Sroufe, 1985;Sukhawatthanakun, 2022). Charalampous et al. (2018) studied the association between attachment and bullying behaviors and found evidence that attachment influences bullying behaviors. In addition, Marini, Dane, Bosacki, and CURA (2006) found that bullies had poor maternal attachment. Studies on young children suggest that bullies and victims are almost certain to have been diagnosed as having an insecure attachment type (Murphy, Laible, & Augustine, 2017;Troy & Sroufe, 1987). Ledley and Heimberg (2006) investigated the long-term impacts of bullying on adults who were bullied as children (Powell & Ladd, 2010). The conceptual framework shown in Figure 1 is supported by the earlier discussion.

Participants
Participants in the study comprised 203 (men = 125, women = 78) ranging in age from 23-56 years (M = 38.15, SD = 8.42). A purposive convenient sampling method was used. Initially, 230 participants were approached, 27 participants did not complete their questionnaire and left the booklet of questionnaires incomplete so, their data was discarded. The final sample size was comprised of 203 participants with an attrition rate of almost 11.7 %. The current study's inclusion criteria included participants exhibiting negative behaviors for at least the last 6 months, this aligns with Einarsen and Rakne's perspective (Einarsen & Raknes, 1997). Secondly, the participant had worked at the same place of at least two years. The participants were further requested to allow the researcher to take the ratings of their associates on the behavior of participants. Hence, only those participants were included in researches that do not have any objection on it.

Procedure
The sample was approached and briefed about the nature of the research. The prior discussion was extremely beneficial. Following all ethical requirements, informed consent was requested before collecting any data from the participants. Moreover, to maintain anonymity and confidentiality no specific information was asked in the booklet. The limitation of one source of data was also addressed so associates and colleagues of the participants were requested to rate the behavior of participants on the behavior checklist. This checklist was designed using all the items of the Negative Act Questionnaire-R (NAQ-R).

Instruments 2.3.1. Negative Act Questionnaire-R (NAQ-R)
The Negative Act Questionnaire-R (NAQ-R) was developed by Einarsen, Hoel, and Notelaers (2009) and comprised twenty-two items. To measure the workplace bullying behavior of perpetrators, the phrase "I have" is added at the beginning of each item. The participants were instructed to rate themselves based on the previous six months on a scale ranging from never (0) to daily (4). The high score indicates a high level of workplace bullying. It's a reliable measure with an alpha ranging from 0.87 to 0.93 (Einarsen et al., 2009). The associates of the participants were also asked to rate the participant using the checklist which was designed in accordance with the NAQ-R scale.

Revised Adult Attachment Scale (RAAS)
This scale is comprised of 18-items (Collins, 1996). This scale allows for the measurement of two attachment dimensions: attachment anxiety and attachment avoidance on a 5-point rating scale with response options ranging from 1 (not at all) to 5 (very much). A higher score on each scale indicates a specific style of attachment.

STATISTICAL ANALYSIS AND RESULT
For analysis, a total of 203 valid questionnaires were used. The sample consisted of 125 men (61.57%) and 78 women (38.42%) ranging in age from 23 to 56 (M = 38.15, SD = 8.42).  Table 1 shows that composite reliability is greater than 0.89-0.95, Cronbach's alpha from 0.89-.96 and rho a are ranging from .090-0.96. The extracted average of variance was clearly above 0.5 threshold indicating good convergent reliability.  Table 2 shows the discriminant validity of scales by calculated heterotrait-monotrait ratios (HTMT) as per the recommendation of Henseler, Ringle, and Sarstedt (2015). The values of HTMT range from 0.70 to 0.77 which are not close to 1 indicating good discriminant validity. For the testing of hypotheses, Smart PLS 3 version 3.3 is used. Table 3 shows the direct effect of the dependent variables, avoidant attachment style and anxious attachment style on the dependent variable, workplace bullying behavior. The avoidant attachment style had a positive and significant impact on workplace bullying (B = 0.15, t = 2.11, p < 0.05). Thus, hypothesis 1 was accepted. Anxious attachment style had a positive but non-significant effect on workplace bullying behavior (B= 0.11, t = 1.82, P > 0.05). Hence, hypothesis 2 was not rejected. The findings indicated that the avoidant attachment style is positively influencing workplace bullying behavior, but, the impact of the anxious attachment style is non-significant.

DISCUSSION AND CONCLUSION
The present study aimed to examine the relationship between an insecure attachment style and the workplace bullying behavior of perpetrators. An individual develops an internal working model of oneself and of their caregivers, this is a key concept in attachment theory. The internal working model is basically cognitive representations that guide the individual about the expectations of others and further educate them on how to respond appropriately to those expectations (Bowlby, 1982;Shaver & Mikulincer, 2014). This internal working model operates both consciously and unconsciously and monitors previous and future social interactions in a non-visible fashion (Stern, 2018) . So the way of communicating and interacting with others is primarily shaped by the internal working models of the self and others (Hazan & Shaver, 1994). Literature on bullying has recognized and emphasized extensively that bullying interactions are best understood by reviewing family, group, school, and community factors (Christie-Mizell, 2003;Espelage & Swearer, 2003;Zahn-Waxler & Radke-Yarrow, 1990). Therefore, it is important to study the impact of an individual's attachment style and intrapersonal relationships in a work setting. It is also believed that the internal working models of an individual's attachment profile influence bullying behaviors (Renken, Egeland, Marvinney, Mangelsdorf, & Sroufe, 1989). On the basis of literature, it was hypothesized that an insecure attachment style would evoke adverse feelings that directly encourage workplace bullying behavior.
The link between workplace bullying behavior and attachment styles has not been well documented. However, it is observed that early attachment representations and internal working models, in general, are thought to have paved the way for people's aggressive social behavior. The key principles of attachment theory are interconnected with anger revealing that interpersonal relationship of individuals is influenced by their attachment styles, including interaction with caregivers, other family members, peers and even strangers. As a result, attachment style influences how people interact with their surroundings (Allen, 2008;Feeney, 2008;Magai, 2008). The first hypothesis stated that an avoidant attachment style is positively associated with workplace bullying behavior. The findings of the study (Table 3) affirmed one assumption and found that an avoidant attachment style had a significant positive impact on the workplace bullying behavior of perpetrators. These results are quite congruent with the literature. Many researchers have explored the effects of adult attachment on interpersonal problems and personality disorders (Aricak & Ozbay, 2016;Cummings-Robeau, Lopez, & Rice, 2009). Bartholomew and Horowitz (1991) suggested that attachment avoidance is associated with interpersonal problems such as increased sensitivity to criticism and aggression. Aggression, antagonism and the projection of undesirable self-traits onto others are associated with an avoidant attachment style (Mario Mikulincer, 1998;Mikulincer & Horesh, 1999). As a result of projecting their anger and hatred onto others, those with avoidant attachments are more likely to engage in active bullying. Thus, it is concluded that an avoidant attachment style has poor interpersonal relationships which leads to bullying behavior in the work setting. The second assumption of the study is that the anxious attachment style is positively associated with the workplace bullying behaviors of perpetrators. This premise was not supported by the findings of the study. There is evidence that high levels of attachment anxiety are correlated with emotional reactive behaviors (Wei, Shaffer, Young, & Zakalik, 2005), excessive reassurance-seeking (Berry, Wearden, Barrowclough, & Liversidge, 2006) hypersensitivity and frustration (Berry et al., 2006;Weems, Hayward, Killen, & Taylor, 2002). Consequently, passive-aggressive bullying behavior is more likely to increase insecure anxious attachment. Mikulincer, Shaver, and Pereg (2003) found that an individual with anxiety attachment has a fear that in tough times people around them may be unreachable and unresponsive. Such people have a negative working model of themselves and a positive working model of others (Vrtička & Vuilleumier, 2012). Therefore, they are extremely sensitive to rejection and alienation indicators (Fraley & Shaver, 2000;Shaver & Mikulincer, 2014). Shaver and Mikulincer (2014) stated that according to theoretical conceptualizations of adult attachment, the attachment system is hyper-activated in the case of anxiety attachment style in which individuals are more likely to engage in activities designed to evoke support and minimize the threat, including clinging or coercive actions and persistently seeking reassurance. In the workplace, such individuals are more likely to be targeted than to perpetrate bullying against a colleague. Bullying at work is widely recognized as a serious problem that has an adverse impact on all entities involved in this situation. Nevertheless, identifying strategies to reduce certain undesirable practices is inevitable. Exploring the causes of bullying in the workplace, especially with regard to the perpetrator is a valuable aspect of this study for organizations. In order to improve the recruitment process and foster a positive work atmosphere, managers would benefit from addressing the attachment patterns of bullies. Future research can be done to develop a tool for assessing potential offenders in the workplace. Besides that, it would enable professionals in mental health to design intervention strategies to facilitate perpetrators in coping with stress and frustration. The research design for the current study was an effort to minimize the limitations observed by some former researchers, such as the single informant approach and examining one industry solely (Hershcovis & Reich, 2013;Tuckey, Dollard, Hosking, & Winefield, 2009). In a current study, a multi-informant approach is used for an in-depth investigation of the phenomenon of bullying in the workplace from the perspective of perpetrators. This method was also helpful in minimizing respondent biases which might have hindered the generalizability of the study's findings. The data collection procedure also includes preventative measures to protect the identity of the study participants and reduce the risk of bias. Despite its potential implications, the study has a few limitations as well. There are numerous individual and environmental factors that contribute to bullying behavior and really need to be studied, while this study focuses on only insecure attachment styles. It is recommended that future studies include those factors for a detailed comprehension of this phenomenon. It is necessary to look at protective factors such as workplace bullying as it is equally dangerous for the victim, organization and offenders. The current study used a cross-sectional and quantitative design. It is suggested that qualitative and mixedmethod study designs be used in the future for a rigorous and broad investigation of bullying behavior. As perpetrators, they are vulnerable groups that have experienced distress and negative emotions in their lives. Moreover, a longitudinal research design could also help in understanding the dynamics of negative behaviors