Beacom School of Business, University of South Dakota, Vermillion, SD, USA
Accepted date: December 23, 2012; Published date: January 12, 2013
Visit for more related articles at Arts and Social Sciences Journal
The present paper attempts to reconcile two separate lines of research in organizational behavior. Specifically, it provides an examination of the dimensionality of organizational justice and emotional abuse in the workplace. It is proposed that the organizational justice facet known as interactional justice and the construct of emotional abuse actually constitute opposing ends of the same conceptual continuum. Further, it is argued that emotional abuse should be conceptually subsumed by broader measures of interactional justice. It is argued that research on the proposed overlap of these important constructs would do much to coalesce and bring clarity to both lines of research.
Emotional abuse; employee perceptions; interactional justice; organizational justice.
Organizational justice is a popular topic for organizational research. In the broadest terms, organizational justice is concerned with the perceptions of fair treatment by employees. In the past 35 years, the dimensionality of the justice construct has received much attention . These dimensions include perceptions of the favorability of outcomes (distributive justice), the fairness of procedures used to allocate rewards (procedural justice), and the quality of relationships with agents of the organization (interactional justice).
Related to the dimension of interpersonal treatment in the justice domain, emotional abuse in the workplace has emerged as a growing topic for organizational research . Emotional abuse involves harmful behaviors that undermine the target’s overall sense of self-worth. These types of behaviors have been shown to have outcomes that are detrimental to the organization such as decreased job satisfaction, higher levels of work stress, and a greater intent to turnover .
Emotional abuse can occur between members at any level within an organization. However, past research has generally focused on the supervisor-subordinate level . Although both the interactional justice and abuse constructs are thought to affect the quality of relationships between supervisors and subordinates, little research has examined the overlap of the two in terms of either theoretical conception or similarity of outcomes (see  for an exception). Therefore, this paper seeks to inspire contributions to the literature through the examination of the degree of conceptual congruence between the relational aspects of organizational justice (i.e., interactional justice) and emotional abuse. In particular, the convergent validity of the relational dimensions of justice and emotional abuse in the workplace would need to be explored in order to fully investigate their degree of conceptual overlap. Because of the interpersonal nature of emotional abuse, it is expected that these constructs would show a similar pattern of relationships with more proximal organizational outcomes, especially those that concern agents of the organization. The following sections review extant literature on these two broad areas of research, including the ways in which they have been defined and the manner in which they have been related to various organizationally-relevant outcomes.
Organizational justice in its most basic form concerns employees’ perceptions of fair treatment by an organization and its agents. While in its empirical infancy, organizational justice was thought to comprise two related, yet distinct, dimensions – distributive and procedural justice. Distributive justice refers to the favorability of a decision. Procedural justice on the other hand addresses the appropriateness of the procedures used to make a decision .
Although this two-factor model of organizational justice has traditionally been a popular conceptualization, it is not without challenge. For example, Cropanzano and Ambrose  suggested that organizational justice is best conceptualized as a single-factor model, based on high correlations obtained in research on the distributive and procedural dimensions. As an alternative to the ubiquitous two-dimensional model, however, this monistic perspective is not as prevalent as a three-factor model that incorporates the dimension of interactional justice. Interactional justice concerns the interpersonal treatment that one receives from another person [7-9].
The influence of interpersonal relationships in the justice literature is not a new notion. Many researchers have included interpersonal treatment under the dimension of procedural justice [10-12]. Indeed, the two dimensions are highly related. Using meta-analytic techniques, Cohen-Charash and Spector  estimated the correlation between the procedural and interactional facets of justice to be 0.54. While this relationship is high, it does not imply that the two dimensions are completely redundant. Moreover, it is possible to distinguish the two on a conceptual basis. If the interpersonal dimension is considered independently, procedural justice then addresses the policies and procedures set forth by an organization, and interactional justice characterizes one’s interactions with the agents of an organization (e.g., one’s supervisor). A logical extension of this argument, then, is that procedural justice should show stronger relationships with organizational-level variables and interactional justice should show stronger relationships with supervisor-related variables.
Research that has distinguished between procedural and interactional justice has supported the idea that the two have differential relationships with variables that are more organizational- or supervisor-based. For instance, Masterson et al.  found procedural justice to be more strongly related to organizational-level variables such as perceived organizational support (POS), organizational citizenship behaviors directed at the organization (OCBO), and organizational commitment. For interactional justice, however, the relationships were stronger with supervisor-related variables, such as leader-member exchange (LMX), and organizational citizenship behaviors directed at individuals (OCBI) such as one’s supervisor.
To further elucidate this issue, Cohen-Charash and Spector  conducted a meta-analysis on the three dimensional model of organizational justice. In many cases, there were not enough studies available to examine the outcomes of interactional justice. However, when enough studies were available, differential relationships were commonly observed between procedural and interactional justice dimensions and outcomes. For example, procedural justice showed stronger relationships with trust in the organization, work performance, and turnover intentions; while interactional justice was more strongly related to LMX.
In a similar fashion, research by Colquitt  has also questioned the dimensionality of the justice construct. However, Colquitt has conceptualized justice using a four-factor model. This model includes the traditional distributive and procedural dimensions, but further divides the interactional dimension into two components: interpersonal and informational justice. Informational justice concerns perceptions of being informed and receiving adequate explanations of procedures. This facet is important because explanations are often necessary for workers to accurately evaluate procedures. Interpersonal justice, on the other hand, encompasses the relational aspects of interactional justice, including respect and sensitivity .
Colquitt  examined each of these four dimensions of justice in both lab and field studies and found support for the construct validity of the four-factor model. Additionally, the relationships observed were consistent with the notion that procedural justice shows stronger relationships to more distal, organizational factors, and interactional justice (both interpersonal and informational) shows stronger relationships to more proximal, relational factors. In the structural model reported by Colquitt, procedural justice showed direct effects on rule compliance and group commitment, whereas interpersonal justice was most strongly related to helping behaviors and leader evaluation (although the model did fit better when procedural justice was allowed to crossload on the leader evaluation factor). Finally, informational justice was most strongly related to collective esteem, which reflected the feelings of self-worth within the group. The theoretical explanation for this relationship was that more information about group members would lead to more mutual liking among group members. Therefore, this finding suggests that informational justice would also be related to more proximal, relational factors. As few researchers have differentiated between interpersonal and informational justice (i.e., the four-dimensional model), however, these two facets are considered together as interactional justice for the remainder of this paper.
On the low end of interactional justice, the quality of supervisor-subordinate interactions is likely to be severely compromised. Low interactional justice (i.e., interactional injustice) may include perceptions of being treated with a lack of trust and respect. These injustices may result in an abusive relationship characterized by betrayal of confidence, derogatory statements, and/or unjustified accusations . Clearly the potential for such negative experiences highlights the importance of interactional justice in supervisor-subordinate relationships.
Researchers of spousal and child abuse have long recognized the insidious and long-lasting effects of emotional abuse in domestic situations. However, it was not until about 14 years ago that psychologists first began to examine the possibility of similar forms of abuse in the workplace. Keashly and her colleagues  were the first to apply the term “emotional abuse” to forms of interpersonal treatment in organizational settings. In Keashly’s  conceptualization, emotional abuse is “the hostile verbal and nonverbal behaviors that are not explicitly tied to sexual or racial content yet are directed at gaining compliance from others” (p. 85). Examples of emotionally abusive behavior include public humiliation, yelling, use of derogatory terms, intense fits of anger, withholding resources, spreading rumors, wrongfully blaming, and giving the “silent treatment.”
A key distinction implicit in this conceptualization is that emotionally abusive behaviors are beyond the realm of what is normally acceptable within a given work environment. Hence, what might be considered abusive on the part of a manager of a high-pressure sales environment is likely to be far more extreme than what might be considered abusive on the part of a manager of a floral shop, for example. Conceptually, the sphere of what is considered an emotionally abusive behavior is taken within the context of a typical, or average work setting (i.e., somewhere in between the two contextual examples given). Ultimately, however, it is the phenomenological experience of being emotionally abused on the part of the target that determines what actually is or is not abusive.
Leading the research on emotional abuse, Keashly [2, 18] has identified six interpersonal features found in other studies of organizational aggression that provide a comprehensive framework of what constitutes emotional abuse. First, and perhaps most importantly, there must be an identifiable pattern of behavior. A single isolated incident involving an aggressive behavior does not necessarily meet the criteria for emotional abuse in Keashly’s typology. Rather, emotional abuse is characterized as a reoccurring pattern of behaviors. It is precisely the repetitive nature of these behaviors that is believed to be so psychologically harmful to the target in question.
Second, the behavior must be unwelcome and unsolicited by the target. For example, a raised tone of voice is only abusive to the extent that it is considered undesirable by the other party. Moreover, what constitutes an unwanted behavior may differ from person to person. As with the experience of other potential stressors, emotional abuse is a subjective phenomenon.
Third, the behavior must be a violation of a cultural norm or an individual’s rights. This includes behaviors that might be considered unethical, inappropriate, or unfair. For example, an open display of favoritism may be regarded as emotionally abusive, provided that the other criteria are met.
Fourth, the behavior must cause harm to the target. Keashly et al.  described this harm as occurring to one’s sense of competence, self-esteem, or both. For some individuals, this may occur relatively easily, whereas other, hardier individuals may be able to employ coping strategies that allow them to handle greater levels of abuse before injury occurs. Therefore, use of coping strategies may serve to moderate the relationship between abusive behaviors and harmful consequences . In any event, it is the attack on self-related constructs that makes emotional abuse a particularly virulent and destructive type of work stressor.
The fifth interpersonal feature involves the question of intent or controllability by the actor. Keashly  reports that for some researchers, the “reasonable person” standard applies to their definitions of abuse. Specifically, for a behavior to be considered abusive, a reasonable person would have to have known that the behavior would be experienced as such. Keashly, however, has argued that the critical factor is not intent. Rather, it is the target’s experience of abuse that is key, and hence, intent on the part of the actor is less important .
Sixth, the behavior must involve power differences. By virtue of their status, managers are in a stronger position to inflict harm on their subordinates than the other way around. This is a particularly common theme in research on workplace aggression. For example, inappropriate or excessive use of power is central to the form of emotional abuse known as “petty tyranny” . Also, the extensive body of research examining the interpersonal dynamics between doctors and nurses clearly relates to this theme (e.g., ). Finally, it is interesting to note that no gender differences have been found for frequency of abuse or types of behaviors engaged in by the actor .
Although Keashly and her colleagues  were the first to apply theories of emotional abuse to the workplace, other researchers have explored the nature and consequences of a similar range of behaviors using a variety of different conceptualizations and labels. Some of these theories closely reflect Keashly’s  conception of emotional abuse, differing only in minor details, while others represent a narrower subset of the behaviors typically regarded as emotionally abusive. For example, Ashforth  has investigated a somewhat more narrow type of abusive behavior among managers he calls “petty tyranny.” Petty tyranny is the tendency to lord one’s power over others and is characterized by arbitrary decision-making, belittling, lack of consideration, domineering style of conflict resolution, emphasis on power differentials, noncontingent punishment, and intolerance of subordinates’ initiative and dissent. Petty tyrants would be expected to micro-manage employees, treating them in an arbitrary and callous manner. Clearly, these types of behavior carry the potential to be perceived as emotionally abusive by the target of the tyranny.
A closer approximation to emotional abuse is what is known in Europe and Australia as “bullying” [22-24]. Bullying may be defined as non-sexual harassment that is repetitive and persistent. Here again we see a reference to a pattern of behaviors, which is consistent with the first interpersonal feature in Keashly’s taxonomy . Further, bullying serves to torment or wear down the target, often leading to feelings of humiliation, offence, and distress . Forms of behavior frequently considered bullying in nature include intimidation, belittling remarks, and persistent criticism .
Regardless of label, emotional abuse involves hostile treatment of subordinates by supervisors. Emotional abuse is a relatively new construct, and thus, its consequences have not been as thoroughly explored as those of the justice dimensions. However, several studies have highlighted the adverse effects of emotional abuse on the target. For example, researchers have linked emotional abuse to lower levels of job satisfaction [3, 17, 25], higher levels of workplace stress [22, 26, 27], and greater intentions to turnover [19, 21]. A logical extension of the aforementioned research on emotional abuse is that, like interactional justice, emotional abuse should also show stronger relationships with supervisor-related variables than with organizational-level variables.
In light of the theoretical overlap between emotional abuse and interactional justice (or more specifically, injustice), it is reasonable to suspect that those researching emotional abuse in the workplace are simply exploring a rather extreme form of interactional injustice. Although couched in the rubric of organizational justice, interactional justice deals with the quality of interpersonal treatment that may even be independent of procedures and outcomes . Indeed, Bies  proclaimed, “The intense and personal pain associated with interactional injustice is experienced as a profound harm to one’s psyche and identity – that is, one’s sense of self” (p. 90). This notion of interpersonal treatment bringing harm to self-related constructs is strikingly similar to Keashly’s conception of emotional abuse .
In addition to the conceptual similarities between emotional abuse and interactional injustice, it is particularly enlightening to examine the redundancy in the types of behaviors that typify both constructs. Examples of coinciding behaviors include: public criticism or ridicule, disregard for the feelings and needs of others, rudeness, and angry outbursts [2, 8].
Although the constructs of emotional abuse and interactional injustice appear to demonstrate a remarkable level of overlap, research has never sought to empirically test the degree of their potential redundancy. Therefore, the purpose of this paper is to call for research to elucidate the congruence, both in measurement and in predictive validity, of emotional abuse and interactional justice. Specifically, it is proposed that emotional abuse represents an extreme form of interactional injustice. In addition, because interactional justice represents the quality of interpersonal treatment beyond what might be considered abusive , a more parsimonious model of these constructs would have emotional abuse subsumed by interactional justice. This is simply because the body of knowledge on the nature of interactional justice is far greater than that for emotional abuse at work.
It is likely that both stems of research would benefit from this amalgamation. For interactional justice, the addition of emotional abuse items to scales that assess interactional justice would serve to broaden and enrich the range of interactions that potentially impact on the quality of interpersonal treatment. With the inclusion of emotional abuse items, interactional justice scales would be expected to account for greater variance in important outcomes such as OCBI, for example. For emotional abuse, inclusion under the heading of interactional justice would lend it position within the broader framework of organizational justice. To be couched in the framework of organizational justice provides a clearer understanding of the role of emotional abuse as an antecedent to a variety of outcomes relevant to organizational research, particularly those that are supervisor-related.
Through a review of the literature on interactional justice and emotional abuse in the workplace, the present paper has identified considerable overlap between these two constructs. In light of the wealth of research on organizational justice in comparison to that conducted on emotional abuse, it also advocates for a conceptual broadening of interactional justice to include elements of emotional abuse. Researchers are encouraged to test the overlap of these parallel constructs to determine the veracity of the proposed relationships. It is believed that such a concerted, theory-driven effort would do much to coalesce and bring clarity to what are currently two separate lines of research. The end result is a valuable adjustment to the organizational justice literature that has so richly benefited our understanding of employee attitudes and behaviors .