Abstract
Deviant behaviour in workplace has become an issue of concern in today’s corporate world. As a result of hard costs, the negative impact of deviant behaviour leaves a devastating effect on overall productivity and performance of the organization. We assumed that abusive supervision will be positively related to employee’s interpersonal and organizational deviance. Moreover, we also hypothesized that this relationship of abusive supervision with both interpersonal and organizational deviance will be moderated by power distance. We verified the formulated hypotheses using data collected from 256 mid-level managerial employees working in IT and software companies based in Delhi (India). Data analysis was done through confirmatory factor analysis (CFA) and moderated regression analysis. Results reveal significant and positive contribution of abusive supervision towards employee’s workplace deviance. Power distance orientation significantly moderates the relationship between abusive supervision and employee’s interpersonal and organizational deviance. The implications for the result obtained are discussed.
Workplace is a forum where a variety of behaviours are expressed. Many a time, some of the employees’ behaviours are so subtle that individuals do not find it hard to get affected. Sometimes, the impact of these behaviours leave positive effect on other’s well-being; often they do harm in a manner that lives of employees become surrounded by negativity and other harmful consequences. Each of this behaviour bears a consequence to the individual employee as well as the entire organization. Since its inception researchers have given these behaviours, many different titles and therefore synonyms used in the literature for these behaviour include workplace deviance (Bennett & Robinson, 2003), counterproductive behaviours (Mangione & Quinn, 1975), maladaptive (Perlow & Latham, 1993) and antisocial behaviours (Giacolone & Greenberg, 1997). Normally, employee’s behaviours/actions are considered as deviant when they violate the prescribed norms of the organizations, traditions and its guidelines. As a result, they endanger the entire organization and its members. These losses associated with organizations make researchers and practitioners to ponder about its detrimental outcomes. The sustainability of any firm becomes difficult with these sorts of corporal incidents. These kinds of discretionary behaviour cannot be always measured in monetary terms and therefore it is present in terms of impalpable damages such as frustration, sabotage, absenteeism, arriving late at work, burnout, and turnover intention and so on. Examining the relationship between workplace deviant behaviour and performance, Dunlop and Lee (2004) found that deviance was negatively associated with business unit performance. In addition to the monetary costs linked to property deviance, they also point to the fact that how notable amount of hidden costs suffered as a result of business units that are not operating at peak efficiency due to the presence of workplace deviant behaviour. Moreover, victims and wrongdoers of deviance have been found to suffer from stress and report decreased productivity. In recent years, employees have cleverly developed some means to involve in negative activity. To name a few, use of Internet for playing online games and download pornography, harassing and/or threatening someone through emails, and set up rouge websites to trash their companies to the world (Ackroyd & Thompson, 1999; Bennett & Robinson, 2003; Leonard, 1999). In this way, the traditional and deadly manifestations of workplace deviance continue to happen. A difficult endeavour in this regard would be to produce an accurate estimate of the cost associated with deviant behaviour in the workplace.
Workplace deviance has been defined as voluntary behaviour that violates significant organizational norms and in doing so threatens the well-being of an organization, its members or both (Robinson & Bennett, 1995). Employee deviance is said to be voluntary because employees either lack the motivation to conform to normative expectations of the social context or become motivated to violate those expectations. We talk of organizational norms or those underlying by formal and/or informal organizational policies, because the definition of deviance must be in terms of the standards of a specified social group rather than in reference to a system of absolute moral standards (Kaplan, 1975). In the mid-1990s, several researchers independently began to focus on the phenomenon of dysfunctional or harmful workplace behaviour in a more comprehensive fashion, attempting to conceptually integrate a range of dysfunctional behaviours into a meaningful whole. The grouping of counterproductive behaviours was first done by Holinger and Clark (1982). They grouped these behaviours into two broad categories: property deviance and production deviance. Property deviance was associated with acquiring or damaging property belonging to one’s employer such as theft, and production deviance was connected with violating organizational norms regarding the quantity and quality of worked performed such as tardiness and work delays. Later, Robinson and Bennett (1995) unified the various deviant workplace behaviours into a single framework in order to compile the scattered research available on the subject into one comprehensive chart. In this way, the researchers were able to assemble a number of deviant workplace behaviours into a single framework. They explain deviant behaviour along two dimensions. The first dimension of Robinson’s typology is the organizational-interpersonal dimension. The axis ranges from deviance directed towards individuals to deviance directed towards the organization. The second dimension depicts the severity of workplace deviance ranging from minor to serious. This typology arranges deviant workplace behaviour into four quadrants, namely production deviance, property deviance, political deviance and personal aggression (Robinson & Bennett, 1995). The present work is basically based upon the classification of workplace deviance advocated by Robinson and Bennett (1995).
Abusive Supervision and Workplace Deviance
There may be numerous reasons why individuals engage in deviant behaviour in the workplace; sometimes it is the characteristic or tendency of the employees to engage in deviant behaviours which harm the organization to a large extent. Often it is the organization in which they work supports or encourages such behaviour. Deviant behaviour can be one of the outcomes of abusive supervision, which may be defined as the subordinate’s perception of the extent to which his/her supervisor engage in the sustained display of hostile verbal and non-verbal behaviours (Tepper, 2000). Mitchell & Ambrose (2007) states that workplace deviance occurs as a result of the shift of employee’s frustration with their supervisor; it becomes directed towards co-workers and the organization.
Lind and Tyler (1988) assert that people are influenced by the kind of interpersonal treatment they receive from organizational authorities. When treated with fairness and dignity, employees feel valued (Lind & Tyler, 1988; Sia & Bhardwaj, 2009; Tyler & Lind, 1992) and they hold a secure position in the organization (Van den Bos, 2001). According to fairness heuristic theory given by Tyler and Lind (1992), once perception of fair judgements is formed, people use these in deciding how to act and behave. If supervisors in the organization are perceived as fair, employees will conduct themselves in a favourable manner and may comply to the demands or requests. The group engagement model proposed by Tyler and Blader (2003) substantiates it further and mentions that this is specifically the case for discretionary behaviours. Employees who feel valued by their superiors exhibit high level of commitment and motivation for the work group and assist the group. Another theory, that is, social exchange theory, also makes a similar kind of proposition. It views that individuals repay the positive regard received by them from the organization. In return, they maintain organizational citizenship behaviour and high organizational commitment (Masterson et al., 2000).
Based on the approach of social exchange (Blau, 1964), findings report that mistreatment by supervisor leads to retaliatory behaviour (Mitchell & Ambrose, 2007; Scarlicki & Folger, 1997). Being targeted and ill-treated by their supervisors, employees seek to take revenge or make their supervisor pay in one way or other (Scarlicki & Folger, 2004). Previous studies also support these arguments. Victims of supervisory abuse develop a negative perception of interpersonal mistreatment by their superiors and resist their supervisor’s influence strategies (Tepper et al., 2001), withhold positive work attitude like organizational citizenship behaviour (Zellars et al., 2002), and can indulge in deviant behaviour directed towards the supervisor (Baron et al., 1999; Inness et al., 2005) and towards the organization (Aquino et al., 1999; Hussain et al., 2014; Detert, Trevino, Burris & Andiappam, 2007).
Tepper (2000) studied the consequences of abusive supervision. His findings revealed subordinates quitting their jobs, lower job and life satisfaction, lower normative and affective commitment, conflict between work and family and psychological distress. Burton and Hoobler (2006) studied about the relationship between nature of supervision and subordinate’s self-esteem. They found negative relationship between nature of supervision and subordinate’s state self-esteem moderated by influence of gender.
In another study, abusive supervision was found to be related to supervisor-directed deviance, organizational deviance and interpersonal deviance along with the moderating effects of negative reciprocity beliefs. The relationship between abusive supervision and supervisor-directed deviance was found to be stronger when individuals hold higher negative reciprocity beliefs (Mitchell & Ambrose, 2007).
In one study, positive relationship was found between abusive supervision and organizational deviance supervisor-directed and organizational deviance (Thau, Bennett, Mitchell & Marrs, 2008).
Tepper et al. (2008) investigated the role of relationship abusive supervision on subordinate’s organization deviance. They found it indirectly related to organization deviance when subordinates perceived that their co-workers were more approving of organization deviance and when their co-workers performed more acts of organization deviance. Based upon the above discussion, we make the following hypotheses:
H1: Abusive supervision will be positively related to employee’s interpersonal deviance. H2: Abusive supervision will be positively related to employee’s organizational deviance.
The Moderating Role of Power Distance Orientation
The present investigation focuses on a cultural value, namely, power distance. It refers to the degree to which one accepts the legitimacy of unequally distributed power in institutions and organizations (Hofstede, 1980). A society where high degree of power distance is exhibited, people accept the hierarchies where every individual has a place to exercise his/her potentials. Derived from the work of Hofstede (1980) individual’s power distance has been frequently examined in cross-cultural and organizational research. In the present study, we consider power distance orientation as the moderator between abusive supervision and interpersonal as well as organizational deviance. Power distance reflects an individual’s regard for authority and power, which in turn moulds the superior–subordinate relationship. Subordinates who possess higher level of power distance orientation acknowledge and accept the prevalence of hierarchy and remain obedient to authority figures (Farh et al., 2007). These employees feel that their position in the organization is below their supervisors and as a matter of fact they are inferior to their supervisors in status. They accept the disparity in power (Tyler et al., 2000). Therefore, they do not bother much about how they are treated by their supervisors and therefore may consider supervisory mistreatment as a normal thing. Hence, they may demonstrate less deviant behaviour. Contrary to this, subordinates with low power distance orientation have a firm belief of equal status with their supervisors. They may have disagreement and criticism with authorities and find themselves more accomplished to communicate to their supervisors regarding the rules and regulations governing the employees in the organization (Farh et al., 2007; Tyler et al., 2000).
Study by Lian, Ferris and Brown (2012) shows that the relation between abusive supervision and subordinate’s interpersonal deviance is moderated by power distance orientation. Their research further indicates the moderating effect of power distance orientation on the relation between abusive supervision and likelihood of rewards and self-regulation impairment. In another study, Wang et al. (2012) found that abusive supervision had significant positive influence on employee’s organizational deviance, interpersonal deviance and supervisor-directed deviance. They also found that power distance moderated the relationship of abusive supervision with perceived interactional justice. Recently, in a study, Lin et al. (2013) found the moderating role of employee’s power distance orientation in the relationship between abusive supervision and employee’s job satisfaction.
Going through these literatures, we hypothesize:
H3: Power distance orientation will have moderating effect on the relationship between abusive supervision and perceived interpersonal deviance in a manner that the relationship will be stronger for those having low power distance compared to those with high power distance. H4: Power distance orientation will have moderating effect on the relationship between abusive supervision and perceived organizational deviance in a manner that the relationship will be stronger for those having low power distance compared to those with high power distance.
Method
Sample
The study was carried out in four different IT and software organizations, namely HCL, WIPRO, Accenture and Hewlett Packard situated at Delhi, India. Permission to collect data was pursued from HR managers of the concerned organizations. Questionnaires were administered to 320 mid-level managerial employees including both male and female employees. In HCL, 90 questionnaires were administered and 77 (male: 40; 44.4 per cent and female: 37; 41.1 per cent) were returned. In WIPRO, 80 questionnaires were administered and a total of 78 (male: 52; 65 per cent and female: 23; 32.5 per cent) were returned. In Accenture, 80 questionnaires were administered and 70 were returned (male: 47; 58.7 per cent and female: 23; 28.7 per cent). In HP, a total of 70 questionnaires were administered and only 31 were returned (male: 20; 28.5 per cent and female: 11; 15.7 per cent). The respondents were assured of complete anonymity of responses and were requested to return filled questionnaires within 2 days. The responses in the filled questionnaires were collected personally by the first author. The researchers obtained a total of 256 filled questionnaires, resulting with a response rate of 85.3 per cent. Participants’ different demographic backgrounds, such as age, gender, marital status, language, and years of working and educational qualification, were recorded through demographic data sheet.
Measures
Three measures were used to operationalize the study variables. The convergent and discriminant validity of the constructs were tested by confirmatory factor analysis (CFA). Outputs from the CFA reflect significant loadings of each indicator on the intended factor (p < 0.05). This supports the convergent validity (Anderson & Gerbing, 1988). The IBM AMOS version 22 software package was used to analyse the responses. We chose to eliminate items with factor loading less than 0.5. The identification and elimination of poorly fit items served our purpose in this stage of analysis. Few items were deleted under each construct. Various fit measures of comparative fit index (CFI), goodness-of-fit index (GFI), normed fit index (NFI) and root mean square error of approximation (RMSEA) for the scales were obtained (Table 1). Along with validity of the items, the inter-item consistency of each construct was reported (Cronbach alpha > 0.60). The measures are mentioned Table 1.
Abusive supervision
Tepper’s (2000) abusive supervision scale was adapted to measure respondent’s perception of supervisory treatment. This was a one-dimensional scale originally consisting of 15 items. Response description against each item was obtained on five-point Likert-type scale—never (1), rarely (2), sometimes (3), most of the time (4) and always (5). Some of the items include: ‘My supervisor ridicules me’ and ‘My supervisor makes negative comment about me to others’. Confirmatory factor analysis of this scale with 14 items depicts better fit model as compared to original scale. Values of various fit measures are found as: GFI = 0.92, CFI = 0.92, NFI = 0.96, RMSEA = 0.05. The Cronbach’s α is 0.92 which signifies the high reliability of the scale (Table 1).
Reliability and Validity of the Measures Used
Power distance
Power distance was assessed using power distance orientation scale by Kirkman et al. (2009). The original scale comprises of eight items. Response description against each item was obtained on five-point Likert-type scale. Sample items include ‘There is nothing wrong in manager/authority making decision without consulting subordinates’ and ‘If employees often question authority, it will make the authority ineffective.’ Model fit is obtained through CFA. Various fit measures of the scale with five items retained from original eight items are as follows: GFI = 0.98, CFI = 0.97, NFI = 0.99, RMSEA = 0.04. Cronbach’s α is found to be 0.77 (Table 1).
Workplace deviance
Workplace deviance was evaluated using the scale developed by Bennett and Robinson (2000). This measure consisting of 19 items assessed two dimensions: (i) interpersonal deviance (having seven items) and (ii) organizational deviance (having 12 items). The response description against each item was obtained on five-point Likert-type scale ranging from never (1) to always (5). Sample items include, ‘Publicly embarrassed someone at work’ and ‘Discussed confidential company information with an unauthorized person.’ Higher additive scores on a dimension included more favourable evaluation of that dimension. Model fit is obtained through CFA. The values of various fit measures include GFI, CFI, NFI and RMSEA. Corresponding values of this fit measure includes 0.91, 0.93, 0.97 and 0.05, respectively. Cronbach’s α of this scale is 0.90 (Table 1).
Results
The obtained data on abusive supervision and power distance for each of the four organizations were analysed to see whether these organizations have any significant difference on the two variables. The result depicted an insignificant difference among the four organizations. Therefore, the researchers clubbed the data and proceeded for further analysis. Pearson correlations among the variable were estimated to indicate the relationships (Table 2). Correlation table shows positive and significant relationship of abusive supervision with interpersonal as well as organizational deviance (r = 0.43, r = 0.44), non-significant but negative relationship of power distance with interpersonal and organizational deviance (r = −0.11, r = −0.08). Therefore, we interpret that greater the perception of supervisory mistreatment higher will be deviance. Similarly, greater the power distance orientation lower is the incidence of deviant behaviour.
Mean, SD and Correlation Coefficients of Study Variables
Moderated regression analysis was carried out to test the hypotheses. Following Zedeck (1971) and Cohen and Cohen (1975), moderated regression analyses were conducted by regressing workplace deviance, on a linear combination of predictors, moderator and predictors–moderator interactions. Moderation occurs when the relationship between two variables depends on a third variable. The third variable is referred to as moderator variable or simply the moderator. In the model, the effect of a moderating variable is characterized statistically as an interaction.
In the present analysis, we carried out two separate moderated regression analyses to verify our hypotheses. Each moderation analysis was accomplished in three steps. The two demographic variables, namely age and tenure were entered to examine their direct contribution. In the second step, abusive supervision and power distance were entered in the second step. In the last step, the interaction of abusive supervision and power distance was entered in order to verify the hypotheses regarding moderating effect.
Table 3 indicates three-step moderated regression analysis for two criterion variables, that is, interpersonal deviance and organizational deviance. Controlled variables were entered in step 1. These variables show no significant effect towards the first criterion variable, interpersonal deviance. For organizational deviance, only tenure shows negative and significant contribution. In step 2, two predictor variables, namely abusive supervision and power distance, were entered. Result shows that abusive supervision has direct positive contribution towards both interpersonal deviance (β = 0.40, p < 0.01) and organizational deviance ( β = 0.41, p < 0.01) meaning that supervisory mistreatment has significant independent contribution towards interpersonal and organizational deviance. Total variance explained by abusive supervision for both the criterion variables are 40 per cent and 41 per cent, respectively. Therefore, H1 and H2 are supported. However, power distance does not have significant independent contribution towards interpersonal deviance ( β = −0.05) and organizational deviance ( β = 0.03).
In step 3, a casual inspection of interaction effect depicts that power distance significantly moderates the relationship of abusive supervision with both interpersonal deviance and organizational deviance in the predicted direction. The interaction of power distance and abusive supervision has a significant influence on employee’s interpersonal deviance ( β = −0.14, p < 0.01), which indicates that the positive relationship between abusive supervision and employee’s interpersonal deviance is stronger when employees are low rather than high in power distance (refer Figure 1). The value of ∆R (0.02, p < 0.01) and F (6.8, p < 0.01) are also significant. The slope is larger for the employees with low power distance scores and smaller for those having high power distance scores.
Results of Moderated Regression Analysis with Interpersonal Deviance and Organizational Deviance as Dependent Variables

For the criterion variable organizational deviance, the interaction effect depicts similar result. Power distance acts as a moderator in the relationship between abusive supervision and organizational deviance in the expected direction, that is, the interaction of power distance and abusive supervision has significant influence on employee’s organizational deviance ( β = −0.17, p < 0.01) and the interaction graph indicates that the relationship between abusive supervision and organizational deviance is stronger for the employees who are low than high in power distance (refer Figure 2). The value of ∆R(0.02, p < 0.01) and F (9.4, p < 0.01) are also significant. The slope is larger for respondents with low power distance than respondents with high power distance. These findings substantiate hypotheses 3 and 4.
Discussion
The objective and purpose of the present study was to investigate and enlarge the previous literature on workplace deviance by investigating the extent to which an individual cultural value, that is, power distance orientation, moderates the relationship between abusive supervision and employee’s interpersonal and organizational deviance. The findings aptly showed the role of moderation for power distance orientation.
Employees in IT and software firms are engaged in innovations and technologies. They also experience much work pressure since nature of the job is demanding (Ahuja et al., 2007; Appu et al., 2015; Moore, 2000; Sia & Appu, 2015). In this circumstance, behaviour of the supervisor can have far-reaching consequence for the subordinates. From this study, it is clear that employees deviate in the workplace as a result of supervisory mistreatment. When they feel abused by their superiors they attempt to engage in deviant behaviour. Interpersonal mistreatment in organization is a major component of abusive supervision which acts as a driving force for employee’s deviant behaviour (Robinson & Greenberg, 1998).

Research evidences suggest that employees reply back in a negative way to supervisory ill treatment by indulging in behaviours which can be detrimental to the organization as well as its members (Aquino et al., 2006; Mitchell & Ambrose, 2007) Our assumption as well as findings for IT and software employees is substantiated by these studies. Bies and Tripp (1998) in their study observed that victimized employees hinder the abusive leaders openly as well as in private. The findings of our study are also supported by researchers done in the past. Mitchell and Ambrose (2007) found positive relation between abusive supervision and supervisor-directed deviance, organization-directed deviance and interpersonal deviance. Schaubhut et al. (2004) found abusive supervision to be related to subordinate’s interpersonal and organizational deviance. They further explained the strength of this relationship as a cause of subordinate’s self-esteem. They also found out that subordinate whose self-esteem was low, abusive supervision was unrelated to their deviant behaviour. However, subordinates who had high self-esteem, abusive supervision was positively related to workplace deviant behaviour.
The finding of the present study is also supported by social exchange theory (Cropanzano & Mitchell, 2005; Gouldner, 1960), which emphasizes on the concept of ‘reciprocity’ as one of its dogma. Reciprocity, though usually thought of as positive, can sometimes be negative too. The perspective of social exchange views that maltreated employees may repay the same to their supervisors by engaging in deviant behaviour. They engage in destructive behaviour and target those whom they perceive fearsome or menacing (Bies & Tripp, 1996). Moreover deviant behaviours of employees may also be targeted towards the organization or co-workers. If targeting the abusive supervisors does not bring benefits for the employees, they shift to other targets. Thus, they displace their anger and aggression to organization and individuals other than the supervisor.
Few reasons may explain why these displacements and anger towards the organization take place. Employee may not always get the opportunity to exhibit retaliatory behaviour towards the particular supervisor. Also the employee might be fearful to retaliate directly towards the individual whom he or she perceives to be at fault. Often experiences that employees gain in the workplace become one of the causes for worker to act out or deviate.
According to Appelbaum, Deguire & Lay (2005) another factor that works behind an employee’s deviance is deviant role models. Individuals or any group in an organization can influence others to commit the acts of deviance. They emphasize on the influence of groups in the workplace. Their research suggests that aggressors and deviant role models have long-lasting effect on victims of deviant behaviours and hence influence others in the group significantly.
Research works have been conducted to see how the perception of being disrespected is one of the main causes for workplace deviance. Disrespect causes dissatisfaction which in turn lead to deviant behaviour. Perhaps this displacement of employees’ frustration to abusive supervision has earlier been noticed by Mitchell & Ambrose (2007). According to Bolin and Heatherly (2001), dissatisfaction results in higher incidence of minor offenses but does not necessarily lead to severe offense. A less satisfied employee with his/her work may become less productive if the needs are not properly fulfilled. In workplace, frustration, injustices and threats to self are primary antecedents to employee deviance.
As for power distance orientation our findings also substantiate the hypotheses. When employees are high power distance oriented, perception of supervisory mistreatment is less because holding and accepting the difference in hierarchy and of their strong deference to authority figures in the organization makes employees to perceive mistreatment as less severe compared to employees who hold low power distance. Our findings are consistent with research done by Lian, Ferris and Brown (2012), Wang et al. (2012) and Lin et al. (2013). It can be inferred that high power distance individuals engage in less interpersonal deviance even if their supervisors exhibit abusive supervision. Thus, the findings corroborate the moderating role of power distance orientation. Moreover, employees having high power distance orientation may have low self-esteem. This can influence the relation between abusive supervision and deviant behaviour (Schaubhut et al., 2004).
Implications
Since deviance in the workplace is a serious threat and is a cost burden, proper and healthy treatment of employees matters a lot. Employees are the vital assets of any organization; they need to be recognized rather than being abused. Their performance is closely related to organization’s success. Therefore, diminishing the negative effects of deviant organizational behaviour becomes essential and draws attention. Our research depicted the buffering outcome of power distance and its impact between abusive supervision and employee’s deviant behaviour. Hence, we put forth two practical implications of the finding. Because abusive supervision leads to the incidents of deviant acts, authority should make all possible attempts to reduce mistreatment by superiors in workplace. An organization can try its level best to understand the influence of abusive supervision on workplace deviance and emphasize upon the solutions.
Thau and Mitchell (2010) suggested a very good solution to reduce the phenomenon. According to them, abusive supervisors should be identified in first step to attenuate the ruinous effect of abusive supervision. Also individuals who bear negative affectivity and abusive orientations should not be placed in such posts. Interventions should also be arranged for the superiors to acquire skills for appropriate treatment and communication with the subordinates.
The second implication is that of proper monitoring of employees’ stressful situations. An employee under potential stress cannot carry out his duties skilfully and suitably. His stress could be detrimental to other employees and his employer as well. His deterioration in health may affect the organizational working culture. Authorities can go for surveys where subordinate’s prolonged abused state can be traced. By doing this, authority can reduce the impairment in their performance. They can obtain the information directly from the subordinates concerning their problems.
