Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Why managers don’t always do the right thing when delivering bad news : the effect of empathy, self-esteem,… Patient, David Leonard 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2006-200650.pdf [ 9.54MB ]
Metadata
JSON: 831-1.0092981.json
JSON-LD: 831-1.0092981-ld.json
RDF/XML (Pretty): 831-1.0092981-rdf.xml
RDF/JSON: 831-1.0092981-rdf.json
Turtle: 831-1.0092981-turtle.txt
N-Triples: 831-1.0092981-rdf-ntriples.txt
Original Record: 831-1.0092981-source.json
Full Text
831-1.0092981-fulltext.txt
Citation
831-1.0092981.ris

Full Text

W H Y MANAGERS DON'T A L W A Y S DO T H E RIGHT THING WHEN DELIVERING B A D NEWS: THE EFFECT OF EMPATHY, SELF-ESTEEM, EMOTIONAL INTELLIGENCE, M O R A L REASONING, AND M O R A L IDENTITY by DAVID LEONARD PATIENT MBA, University of British Columbia, 1989 LL.B . (Hons.), University of London, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Business Administration) THE UNIVERSITY OF BRITISH COLUMBIA August 2006 © David Leonard Patient, 2006 A B S T R A C T Previous research shows that some managers do not deliver bad news in ways deemed interactionally fair (with dignity, respect, and adequate explanations). In this dissertation I explore whether specific individual characteristics predict the tendency to deliver bad news in ways regarded as interactionally (un)fair: the communicator's empathy, self-esteem, moral development, emotional intelligence, and moral identity. In Study 1, 173 practicing managers responded to a scenario task in which a layoff was to be communicated and their written messages were coded for interactional justice. Results showed that empathic concern, moral development, and strategic emotional intelligence each individually predicted interactional justice. However, these relationships were not significant when all of the predictors and proposed interactions were included. Moral development moderated the relationship between empathic concern and interactional justice. In Study 2, 81 students provided face-to-face feedback containing negative news to a confederate, and the videotaped feedback was coded for interactional justice. In Study 2, empathic induction was manipulated and moral identity was primed. Results showed that empathic induction increased interactional justice. Further, a significant three-way interaction showed that when moral identity was high, moral development moderated the effect of the empathic induction on interactional justice. Specifically, the interactional fairness of high (versus low) moral development communicators was lower in the control condition and increased by the empathic induction, whereas the interactional fairness of low moral development communicators was not affected by the empathic induction. TABLE OF CONTENTS Abstract... ii Table of Contents in List of Tables vi List of Figures viii Chapter 1: Introduction 1 Chapter 2: Study 1 6 The Link Between Interactional Fairness and Communicating Bad News 7 Obstacles to Interactionally Fair Communication 10 Theory and Hypotheses for Study 1 13 Empathy and Communication of Negative News : 13 Affective Dimensions of Empathy 15 Empathic Concern and Interactionally Fair Communication 16 Personal Distress and Interactionally Fair Communication .-. 19 Self-Esteem and Communication of Negative News 21 Self-esteem and Expectations of Social Rejection 24 Self-esteem and Motivation 25 Self-esteem and Interactionally Fair Communication 26 Self-Esteem as a Moderator between Empathic Concern and Interactionally Fair Communication 28 Moral Development and Communication of Negative News 29 Moral Development and Interactionally Fair Communication 31 Moral Development as a Moderator between Empathic Concern and Interactionally Fair Communication 32 Emotional Intelligence and Communication of Negative News 33 The Ability Model of Emotional Intelligence '. 35 Emotional Intelligence and Interactionally Fair Communication 37 Summary of Study 1 Hypotheses 39 Study 1 Methods 40 Participants • 40 Procedure 40 Measures 42 Predictor variables 42 Trait Empathic Concern 42 Trait Personal Distress • 42 State Self-esteem 42 Moral development 43 Strategic Emotional Intelligence .'. 44 Criterion variable 44 Interactional justice ; 44 Control variables 45 Demographic variables 45 Study 1 Results 45 Study 1 Discussion 48 Study 1 Theoretical Implications 51 Study 1 Strengths and Limitations 52 Chapter 3: Study 2 57 Theory and Hypotheses for Study 2 . 57 Empathic Induction and Interactionally Fair Communication 58 Moral Identity and Interactionally Fair Communication 59 Moral Identity Priming and Communicating Negative News 60 The Internalization Dimension of Moral Identity and Communicating Negative News 61 The Interaction of Empathic Concern, Moral Development, and Moral Identity in Predicting Interactionally Fair Communication 63 The Four Component Model of Ethical Decision Making and Behavior 63 Summary of Study 2 Hypotheses 68 Study 2 Method ..." 69 Participants 69 Procedure..' 69 Measures 73 Manipulation checks 73 Predictor variables '. 74 Moral development 74 Moral Identity • 74 Criterion variables 74 Interactional justice 74 Control variables 75 Trait Empathic Concern • 75 Demographic Variables 76 Source 76 Study 2 Results 76 Study 2 Discussion 79 Chapter 4: Overall Discussion 91 Theoretical Implications 92 V Implications for Practice 96 Strengths and Limitations 97 Future Research 100 Conclusion 102 References 104 Appendix A: Consent Form for Study 1 121 Appendix B: Consent Form for Study 2 123 vi LIST OF TABLES Table 1..... 125 Table 2 126 Table 3 127 Table 4 128 Table 5 ; 129 Table 6 130 Table 7 131 Table 8 132 Table 9 133 Table 10 134 Table 11 135 Table 12 136 Table 13 137 Table 14 138 Table 15 139 Table 16 140 Table 17 141 Table 18 142 Table 19 143 V l l Table 20 144 Table 21 145 Table 22 146 Table 23 147 V l l l LIST OF FIGURES Figure 1. Factors leading to interactionally fair communication 148 Figure 2. Interaction of trait empathic concern and moral development in predicting interactional justice in Study 1 149 Figure 3. Figure 3. Interaction of empathic concern and moral development in predicting interactional justice in low moral identity condition of Study 2 150 Figure 4. Interaction of empathic induction and moral development in predicting interactional justice in high moral identity condition of Study 2 151 Figure 5. Interaction of empathic induction and moral development in predicting interactional justice (politeness) in low moral identity condition of Study 2 152 Figure 6. Interaction of empathic induction and moral development in predicting interactional justice (politeness) in high moral identity condition of Study 2 153 Figure 7. Interaction of empathic induction and moral development in predicting interactional justice (dignity and respect) in low moral identity condition of Study 2. 154 Figure 8. Interaction of empathic induction and moral development in predicting interactional justice (dignity and respect) in high moral identity condition of Study 2 155 1 CHAPTER 1: INTRODUCTION Among the many roles that leaders and managers have in organizations, delivering bad news to employees is arguably one of the most difficult. Bad news refers to decisions, evaluations, or events that can be viewed by employees as negative or unfavorable. For example, a manager may need to inform an employee of a poor performance rating, that the resources available to him or her have been decreased, or that the organization has gone bankrupt. Reactions to bad news can range from accepting the information and cooperating with authorities to rejecting the bad news and harming both the messenger and the organization from which it came. The latter reaction can include such behaviors as organizational retaliation behavior (Skarlicki & Folger, 1997), theft (Greenberg, 1990a), and litigation (Lind, Greenberg, Scott & Welchans, 2000). From an organizational justice perspective, the way in which decisions are communicated can be as important to recipients' perceptions of fairness and subsequent reactions as the content itself. Moreover, research suggests that, especially when outcomes are deemed unfavorable, individuals look to aspects of the way in which the news was delivered when making fairness judgments (Greenberg, 1994; Shapiro, Buttner, & Barry, 1994). Despite the empirical evidence regarding the importance of delivering bad news in ways that demonstrate dignity and respect (Bennett, Martin, Bies, Brockner, 1995), evidence suggests that managers can communicate bad news in an impolite fashion, provide curt explanations, and distance themselves from victims. For example, managers implementing layoffs underutilize practices that display sensitivity to the victims (Folger & Skarlicki, 2001). Insensitive and impersonal treatment compounds the bad news because the supervisor's treatment can be viewed as a negative outcome in its own right 2 (Cropanzano & Ambrose, 2001). The problem is made worse by the fact that managers appear especially likely to distance themselves from victims and use insensitive communication strategies at those very times when fair interpersonal treatment plays its most important role: when outcomes are negative (Folger & Skarlicki, 1998). In contrast, studies show that sensitive communication and adequate explanations can contribute to fairness perceptions even when outcomes are unfavorable (Brockner, 1988). How can we explain this tendency of managers to "make bad times worse" by the way they deliver bad news? In this dissertation I apply an organizational justice perspective to examining the way that negative news is communicated. Though early justice research tended to focus on distributive and procedural justice (Adams, 1965; Thibaut & Walker, 1975; Leventhal, 1976), the importance of interactional justice to overall fairness perceptions has been increasingly recognized. Bies (2001) defined interactional justice as the quality of interpersonal treatment during the enactment of organizational procedures. The importance of this social dimension of justice is shown by the fact that, when asked to give examples of fair and unfair behavior, people tend to think mostly of acts related to politeness, and consideration for the needs and feelings of others (Messick, Bloom, Boldizer, & Samuelson, 1985; Mikula, Petrick, & Tanzer, 1990). Many aspects of interactional justice relate specifically to how an outcome - especially a negative outcome - is communicated. Greenberg (1993) suggested that interactional justice includes two distinct facets, each with independent effects: respectful interpersonal treatment (interpersonal justice) and adequate explanations (informational justice). Interpersonal justice refers to the 3 sensitivity with which a recipient of negative news is treated, in terms of politeness, dignity, and concern, and affects reactions to a decision by making people feel better about an unfavorable outcome. Informational justice, which focuses on the adequacy of the explanations provided (in terms of detail, thoroughness, timeliness, and sincerity), allows recipients of negative news to feel better about the procedures followed. In situations where negative news is communicated, these interactional dimensions of justice have been shown to be positively related to evaluations of the organization, outcome satisfaction, and less negative reactions to unfavorable outcomes (Colquitt, Conlon, Wesson, Porter, & Ng, 2001). Because both interpersonal and informational justice relate to important aspects of how negative news is communicated, I focus on interactional justice (the umbrella term). In this dissertation I explore whether specific individual characteristics predict the tendency to deliver bad hews in ways regarded as interactionally (un)fair. Specifically, I focus on five factors that could relate to how a manager communicates negative news: the communicator's empathy, self-esteem, moral development, emotional intelligence, and moral identity. First, I consider the ways in which communicator empathy for the news recipient predicts interactionally fair communication. In particular, Ipropose that empathy focusing on other-oriented feelings of concern for the bad news recipient relates positively to interactionally fair communication, whereas self-focused feelings of distress prompted by the recipient's plight will relate negatively to interactionally fair communication. Second, I consider how a communicator's feelings of self-worth influence interactionally fair communication directly, as well as moderate the relationships between empathy and communication behavior. I propose that 4 communicator self-esteem relates to interactionally fair communication. I also argue that the relationship between empathy and interactionally fair communication is stronger for individuals with low (versus high) self-esteem. Third, I propose a positive relationship between communicator moral development and interactional fairness, and discuss how a manager's level of moral development can moderate the effects of empathy on interactionally fair communication. Fourth, I explore whether emotional intelligence is positively related to interactionally fair communication. Fifth, I propose that moral identity will relate positively to interactional justice. An understanding of these factors and how they interact can help explain why some managers might not demonstrate respect and concern for the recipients in the way the managers communicate bad news. In this dissertation I contribute to the research literature in several ways. From a justice perspective, considerable research attention has focused on questions related to understanding fairness as an independent variable (i.e., What are the effects of fairness perceptions on organizational outcomes?). In contrast, I explore justice as a dependent variable (i.e., What predicts a manager's fair treatment of other organizational members?). This line of inquiry is relatively under-researched. Second, by understanding what predicts interactional justice, I have the potential to contribute to justice theory, by explaining why managers don't always "do the right thing" when delivering bad news. Third, this research question can have implications for managerial practice. Assuming organizational leaders and managers wish to enhance fairness perceptions in the workplace, there is a need to understand the factors that might prevent this from occurring. Moreover, numerous findings in organizational study are not readily adopted by managers—a phenomenon that researchers in Industrial and Organizational 5 Psychology have labeled the Scientist-Practitioner gap (Dunnette, 1990). This dissertation represents an attempt at understanding why, despite the considerable benefits of workplace fairness reported in the literature, some aspects of organizational justice remain underused by managers. In this sense, this line of research attempts to also understand how to better extend fairness principles to the workplace. This dissertation consists of two studies using different contexts and methodologies. Study 1 involves 173 practicing managers. The managers provided written messages in response to a scenario exercise in which a layoff was to be communicated. In Study 1,1 tested the effects on interactional fairness of communicator empathy, self-esteem, moral development, and emotional intelligence, and their interactions. Study 2 builds on and extends findings from Study 1 in several respects. First, an experimental approach was used to provide enhanced control. Second, I manipulated empathic concern and primed moral identity to explore whether interactional justice could be enhanced via an intervention. I also investigated the interactive effects of empathic induction, moral development, and moral identity on interactional justice. Moreover, Study 2 extended my investigation to a context where the news contained feedback and was communicated face-to-face (versus in writing). 6 CHAPTER 2: STUDY 1 Especially in challenging economic times, the role of the manager as representative of the organization and bearer of bad news is not an easy one. Like the ill-fated messenger in the Persian fable, who met an untimely end after giving bad news, a manager can often be held responsible for the outcomes he or she communicates, even when he or she is hot to blame (Manis, Cornell, & Moore, 1974). Bies (1987) suggested that when an outcome is likely to be regarded as unfair, its very communication can be considered a violation of norms of justice and fairness. To be responsible for such communication creates a "predicament of injustice" whereby the communicator fears reactions to the injustice and possible reputational consequences. Communicating bad news involves potential tradeoffs between communicator and recipient. In choosing how to convey the message, the communicator considers strategies that protect and maintain his or her own social identity, that of the news recipient, or both. Such situations have also been discussed in terms of social predicaments (Gonzales, Manning, & Haugen, 1992), face-threatening acts (Brown & Levinson, 1987), and failure events (McLaughlin, Cody, & O'Hair, 1983). These terms have the following features in common. First, a negative message needs to be communicated. For example, a manager may need to inform an employee of a negative evaluation. Second, the message and its communication pose potential threats to the preferred identities of both communicator and news recipient. These situations are highly problematic for managers in that claims to moral character and concern for others are significantly threatened by the need to communicate negative news (Gonzales et al., 1992). 7 The Link Between Interactional Fairness and Communicating Bad News The way in which bad news is communicated is an especially powerful indicator of interactional justice. The process of delivering bad news has been the focus of justice research in two regards. First, most studies involving interactional justice have focused on situations involving negative outcomes. Substantial empirical support shows that process considerations, including interpersonal treatment, become most important when outcomes are negative or severe (e.g. Brockner & Wiesenfeld, 1996; Folger, 1993; Shapiro et al., 1994; Shapiro, 1991). Second, ever since Bies and Moag (1986) coined the term interactional justice, interpersonal communication has played a central role in its theorizing and measurement. For example, the four criteria of interactional fairness originally proposed by Bies and Moag (1986) all focus on "communication criteria of fairness": truthfulness, respect, propriety of questions, and justification of decisions. In further elaborating interactional justice, Bies (1987) identified another aspect of interpersonal communication: the explanations that accompany outcomes that are likely to be regarded as unfair. Subsequent research has broadened interactional justice to include additional aspects of communicative behavior, including sincerity, personalization, timeliness, and communication channel (Shapiro et al., 1994). In summary, the way in which negative decisions are communicated to individuals affected by the decision plays a central and important role in their perceptions of interactional justice. The importance to effective communication of adequate explanations and communicating with sensitivity, concern, and respect has been explored in literatures outside of the organizational justice context, including: politeness theory (Brown & 8 Levinson, 1987) and account-giving strategies (Schonbach, 1980). I use the term interactionally fair communication (or fair communication) to refer to communication which is likely to be regarded by the recipient of negative news as interactionally just, in terms of both interpersonal and informational fairness. A review of the literature suggests that the fairness of communication of negative news can be evaluated in several ways. First, were the explanations provided truthful and candid? People view as interactionally unfair attempts by communicators to distort or conceal the truth (Shapiro, 1991). Second, were adequate explanations provided? People feel morally entitled to know the reasons for unjust outcomes (Bies, 2001), and because excuses and justifications for bad news "lessen expressions of moral outrage," their provision is important to perceptions of interactional justice. Third, were efforts taken to demonstrate politeness, respect, and concern for the recipient of bad news? The group value model of justice predicts that how a recipient of an unfair outcome is treated interpersonally is important in demonstrating his or her social status and worth within the group (Tyler & Lind, 1992). When communicators can potentially threaten the social identities of others (for example, by communicating negative news), politeness behaviors are often used to soften the blow (Morand, 2000), thereby enhancing perceptions of interpersonal justice. Fourth, was the news communicated in a timely fashion? Untimely communication of bad news is likely to demonstrate less communicator concern and to cause anger and frustration which can make the explanation provided seem less reasonable (Shapiro et al, 1994), thereby reducing interpersonal and informational fairness. Finally, was the communication channel used likely to be regarded by the news recipient as personal and appropriate? For example, the use of a personal communication 9 channel may increase perceptions of interpersonal justice by conveying that the communicator is sincere, and that an effort has been made to demonstrate respect and concern for the recipient (Shapiro et al, 1994; Sitkin, Sutcliffe, & Barrios-Chopin, 1992). The more that communication of negative news satisfies the above criteria the more likely it will be deemed as interactionally fair communication. It is important to distinguish interactional justice from moral action in terms of the perspective generally taken by each, the domains to which they apply, and the behaviors they encompass. First, moral behavior involves the application of universal rules (Kohlberg, Levine, & Hewer, 1983). Whether behavior is moral is evaluated by asking whether a specific rule was followed or a specific type of reasoning engaged in, rather than by asking how another party was affected. In this way, moral action focuses on the behavior of the actor, and can involve criteria such as resistance to temptation or self-sacrifice, which need not benefit the recipient. In contrast, interactional justice focuses on the quality of treatment received by the recipient of a negative outcome, whether this is evaluated by the recipient, third parties, or even the transgressor. Interactionally just behavior can be prompted not only by the application of universal principles, but also, for example, by feelings in a specific situation for an individual, or even by instrumental concerns (e.g., the desire to avoid a negative organizational outcome, such as theft or retaliation). Further, interactional justice is suitable for both universalistic relationships, where a communicator owes no special duty of care to a recipient, and for more particularistic relationships. In contrast, because moral behavior involves the application of universal rules, it tends to be more appropriate for impersonal versus close relationships. For the above reasons, it would be possible for moral behavior (versus less 10 moral behavior) to lead to less interactional fairness when the application of universal principles could, for example, lead to less concern for the individual recipient, less personalization of the message, or explanations that may be regarded as less reasonable by a bad news recipient. In this dissertation, I focus on interactional justice as my criterion variable because of its strong communication focus, its applicability to close relationships (such as manager-employee relationships), its diverse motivations, and its focus on the quality of treatment received. Moral judgment is considered as one predictor of interactional justice. Obstacles to Interactionally Fair Communication Although the focus of this dissertation is on individual factors that can predict managers' communication, a number of situational and practical reasons can also contribute to interactionally unfair communication of negative news. One reason why managers might provide incomplete and impersonal explanations for adverse events might be efforts to minimize litigation (Sitkin & Bies, 1993). Bies and Tyler (1993) noted the dramatic rise in employees taking legal action against their organization, and the resulting growing impression among managers of a "litigation mentality." They suggested that many organizations attempt to minimize litigation by focusing their efforts on implementing policies which emphasize due process, including the defensibility and consistent application of fair procedures. Bies and Tyler (1993) pointed out, however, that an increasing formalization of procedures can constrain the flexibility and responsiveness which managers can demonstrate in individual situations. This can lead to communication that is less personalized, and can therefore be regarded as less sincere and as demonstrating less respect and concern. Organizations that fear litigation may also 11 encourage managers to provide less detailed, less open explanations for negative news, for fear that the information provided could be used against them in a legal suit. Security considerations, in terms of protecting company property and information, can also explain curt communication, and hasty implementation, of such events as employee dismissals (Dubose, 1994). An organization may want to take steps to ensure that a possibly angry ex-employee cannot damage, steal, or disseminate company property, whether intellectual or otherwise. In such a case, one way to make sure that an ex-employee cannot access company property and information, is to communicate details of their layoff to them offsite and to subsequently restrict their access to the organization. Another reason for low interactional fairness pertains to organizational culture. Dutton, Frost, Worline, Lilius, and Kanov (2002) argued that whereas some leaders cultivate environments which encourage compassionate treatment of employees, other organizations cultivate an environment in which employee hardship is either ignored or responded to in a detached and unsympathetic manner. It is likely that in the latter environment, efforts by a manager to sensitively address the pain of an employee receiving bad news will go unacknowledged and unrewarded, and can even be seen as wasting valuable company time and resources. Practical considerations, such as a large and/or dispersed group of employees might also make it difficult for managers to communicate with each employee in an individualized, sensitive fashion. Folger and Skarlicki (2001) noted that the sheer size of some layoffs might make it difficult to personalize communication. For this reason, not all mass-produced responses, such as standard severance packages and the hiring of 12 outplacement services, indicate a lack of managerial concern for employees who are victims of bad news. The above examples illustrate some of the ways in which a desire to provide employees with reasonable and detailed explanations for negative news in a personal, recipient-oriented fashion can face practical challenges. Nonetheless, research on fairness (Bies, 1987), the M U M effect (Tesser & Rosen, 1975), impression management (Greenberg, 1990b), and politeness theory (Brown & Levinson, 1987) suggest that managers are also motivated by personal considerations. These include the need for communicators to safeguard their own reputation, maintain their self-esteem, and protect their own emotional well-being. In these respects, though managers may be guided in their communication by what they perceive the needs of their organization and of news recipients to be, they are also likely to be substantially influenced by the potential instrumental, relational, and identity costs to themselves. Investigations into the " M U M effect" (Tesser & Rosen, 1975) have provided several important insights into why people are generally reluctant to communicate bad news to affected parties. Specifically, the very act of communicating negative news results in relational costs (Manis et al, 1974), and expectations by the communicator of a negative evaluation. It is therefore not surprising that as a result the reluctance to transmit bad news increases when the negative news becomes more serious (Tesser, Rosen, & Batchelor, 1972) and when the communicator is more concerned about the impression he or she makes (Johnson, Conlee, & Tesser, 1974). The fact that the M U M effect was found to apply even in situations where the communicator remains anonymous (Rosen & Tesser, 1972), suffers no material or physical costs (Tesser & Rosen, 1975), and is not 13 responsible for the news conveyed (Tesser et al, 1972), suggests that there may be considerable perceived psychological costs of communicating negative news. However, whereas research into the M U M effect focused on whether individuals would directly communicate negative news to the affected party, this research looks at the how the news is communicated - and specifically, with how much interactional fairness - when its communication cannot be avoided. Theory and Hypotheses for Study 1 In Study 1 I focus on four factors that are predicted to affect communication of negative news. First, I consider the effect on fair communication of a communicator's feelings, in terms of empathic concern and personal distress, prompted by the recipient of negative news. Second, I consider whether a communicator's self-esteem influences how he or she responds to a particular social threat: the anticipated interpersonal rejection and relational devaluation by the recipients of bad news. Third, I consider the effect of a communicator's level of moral development on fair communication. Last I consider the effect on fair communication of elements of emotional intelligence, namely strategic emotional intelligence. As shown in Figure 1, the communicator's empathy for the victim, self-esteem, moral development, and emotional intelligence are all expected to directly influence interactionally fair communication. The link between empathy and fair communication is expected to vary with communicator self-esteem and moral development. In the next section I explain each of these relationships in detail. Empathy and Communication of Negative News Some researchers have distinguished between the terms sympathy and empathy, whereas others have used them interchangeably. For example, Eisenberg and Miller 14 (1987) used the term sympathy to denote feelings of sorrow or concern that might accompany, and result from, matching the affective state of the target. They use the term empathy to refer specifically to affective matching without additional feelings for the person whose suffering is observed. Batson, Fultz, and Schoenrade (1987) referred to the same feelings of sympathy and compassion for the target person simply as empathy. In this paper, the term empathy is used to refer to a range of affective responses that an individual might experience in reaction to the suffering of another person, and that involve taking the perspective of another person and the vicarious sharing of the other's affect. Empathic concern and personal distress denote specific dimensions of empathy. The ability of communicators to empathize with message recipients has long been associated with effective interpersonal communication (Redmond, 1989). Athos and Gabarro (1978) suggested that empathy is especially important in business, where effective understanding of another person is essential to the quality of communication between them. Empathy generally refers to an emotional response that stems from another person's emotional state or condition, and that is congruent with the other's emotional state or condition (Eisenberg & Strayer, 1987). Empathic responses can also be prompted by the anticipated, but not yet experienced, emotional states of others (Redmond, 1989). Anticipatory empathy is particularly relevant to situations where the empathizer has advance knowledge of information that might negatively affect another person, such as the bad news that is to be communicated. Trait empathy refers to an individual difference and is usually measured via self-report. Individuals that are high (versus low) in trait empathy report a greater tendency to experience empathic feelings when they observe the suffering of another person. 15 Research has distinguished between two categories of empathic responses: cognitive and affective reactions. From a cognitive perspective, empathy consists of being able to predict and recognize the feelings of others (Strayer, 1987), without necessarily experiencing specific feelings as a result. Affective empathic responses refer to the emotional reactions of empathizers to the suffering of another person. For example, a manager who is communicating news regarding organizational layoffs might vicariously experience fear or anxiety when observing the employees who are affected by the announcement. In addition, the manager might experience feelings, such as compassion and concern, that are less congruent with the negative reactions of the employees. My specific interests lie not only in a communicator being able to imagine a situation from another person's point of view (i.e., cognitive-empathy), but, rather, as in the above example, in the feelings evoked by doing so (i.e., affective empathy). These are likely to be a more immediate and powerful determinant of affective reactions to the situation, especially when they go beyond simply matching the other person's affect (Davis, 1994), and therefore more significantly influence interactionally fair communication. Affective Dimensions of Empathy Research has identified two distinct emotional empathic responses to another person's suffering: empathic concern and personal distress (Coke, Batson, & McDavis, 1978). Both reactions are provoked by seeing or anticipating the suffering of another person and are affected by the perceived magnitude of the need (Batson et al., 1987). However, empathic concern and personal distress differ in several important respects. First, whereas empathic concern is primarily an other-oriented response, involving 16 feelings focused on the person in distress, personal distress is a more self-oriented empathic response that focuses on the observer's feelings of anxiety and discomfort (Eisenberg, 2000). Second, empathic concern is a response which goes beyond matching the victim's affect. For instance, empathic responses to negative affect in another person can include positively-valenced emotions, such as warmth and compassion (Davis, 1994). In contrast, personal distress reactions generally involve only feelings that are congruent with the victim's negative emotions (Davis, 1994). Third, empathic concern and personal distress differ in the motivations and behavior they are associated with. Specifically, empathic concern leads to altruistic motivations to reduce the suffering of the target and tends to lead to more prosocial behavior. Personal distress, on the other hand, results in egoistic motivation to reduce the distress of the observer and more aggressive behavior, in terms of resentment, hostility, and suspicion (Davis, 1994). In the following section, I propose that empathic concern will be positively related, whereas personal distress will be negatively related, to interactionally fair communication. Empathic Concern and Interactionally Fair Communication As noted above, the term empathic concern denotes a specific dimension of empathy: feelings of warmth, compassion and concern for another person aroused by observing their need (Davis, 1980). Empathic concern is other-oriented, in that it goes beyond merely feeling what the other person feels and involves feelings of concern and compassion for the other person (Eisenberg, 2000). Most studies have assessed empathic concern by means of a set of adjectives, such as "tender" and "sympathetic." Situational empathic concern has been induced through asking the subject to imagine how the other feels. For example, Batson, Earley, and Salvarani (1997) asked subjects in the empathic 17 concern condition to "imagine how the person ... feels about what has happened and how it has affected his or her life" (p. 753). Most investigations have found a positive relationship between imagine-the-other perspective-taking and feelings of empathic concern (Davis, 1994). Eisenberg, Shea, Carlo, and Knight (1991) proposed that initial matching of the victim's affect is transformed into empathic concern when the observer focuses on the other's feelings. Empathic concern is associated with several motivations and outcomes related to interactionally fair communication of negative news. Empathic concern is strongly associated with selflessness and concern for others (Davis, 1983). The stronger the feelings of empathic concern experienced by the observer, the stronger their motivation to reduce the suffering of the victim (Batson, 1987; 1991). Empathic concern is associated with prosocial, helping, and altruistic behavior (Batson, 1998; Underwood & Moore, 1982). The relationship between empathic concern and helping behavior has been found to hold even when the observer's own discomfort could be alleviated through other means, such as escaping the situation (Batson et al., 1987). For instance, Batson and colleagues (1997) asked participants to listen to a bogus radio interview with a young woman in serious need. One third of the participants were asked to remain objective when listening, one third were asked to imagine their own feelings in that situation, and one third were asked to imagine the feelings of the person in distress. As predicted, imagine-the-other instructions led to greater empathic concern for the victim and a greater motivation to reduce the victim's suffering. Empathic concern has also been associated with several measures of a more considerate social style (Davis, 1994), 18 including warmth, positive outlook, and effective communication, in terms of both communicating openly and readily listening (Davis & Oathout, 1987). As mentioned previously, communicating negative news sensitively and with detailed explanations (i.e., in an interactionally fair manner) can be costly to the communicator, in terms of both requiring more effort and exposing the communicator more significantly to the recipient's negative reaction. Why would a communicator go to the extra effort to communicate fairly, rather than in a brusque or more self-defensive manner? One important reason is the desire, prompted by feelings of concern and compassion for the other person, to reduce the other person's suffering, whether or not the communicator is responsible for the harm caused. Greater (versus lesser) empathic concern for the target is expected to lead to attempts to reduce (or at least not add to) the suffering of the news recipient in several ways related to interactional fairness. One strategy is through politeness which attempts to recognize and support the social identity of the news recipient (Brown & Levinson, 1987). Another way of softening the blow of the bad news is through providing mitigating explanations for the bad news, such as apologies and excuses, rather than aggravating explanations, which deny or justify the harm to the news recipient, or no explanations. Providing negative news in interactionally fair ways often involves greater exposure to the negative reaction of the news recipient, which can be more psychologically costly to the communicator. Because of the other-oriented motivations of the communicator and their genuine concern for the bad news recipient, empathic concern is expected to be associated with communication strategies directed at alleviating 19 the suffering of the bad news recipient even when communicators could avoid or distance themselves from that suffering through less interactionally fair communication strategies. Hypothesis 1: Empathic concern for the recipient(s) is positively related to interactionally fair communication of bad news. Personal Distress and Interactionally Fair Communication As noted earlier, a second affective aspect of empathy that has received a great deal of empirical attention is personal distress, a self-focused aversive reaction resulting from observing, or anticipating, the negative emotion of another person (Davis, 1994). Feelings of personal distress increase with the intensity of the negative feelings vicariously experienced (Eisenberg, 2000). When a person observes suffering, personal distress shifts their concern away from the other party's suffering to his or her own feelings of personal discomfort prompted by that distress, a process that Hoffman (2000) termed egoistic drift. Personal distress has usually been measured by means of a set of adjectives, including the degree to which an individual is feeling worried, upset, and distressed. A common procedure for inducing personal distress has been to ask study participants to imagine what their own feelings would be in a given negative situation. For example, Batson and colleagues (1997) asked participants in the personal distress condition to "imagine how you yourself would feel if you were experiencing what has happened to the person .. . and how this experience would affect your life" (p. 753). Whereas empathic concern is primarily other-oriented, personal distress involves primarily self-oriented feelings. Unlike empathic concern, which is positively related to cognitive perspective-20 taking (Davis, 1983), personal distress is not significantly related to trying to see the situation from the other person's point of view (Toi & Batson, 1982). Feelings of personal distress are associated with distinctive motivations and outcomes associated with interactionally fair communication of negative news. The greater the personal distress experienced, the greater becomes the egoistic motivation to reduce it (Batson et al , 1987). Whether this egoistic motivation leads to behavior directed at reducing the victim's discomfort depends on the ease or difficulty of reducing one's own personal distress feelings by other means. When it is easier or less costly to alleviate the victim's suffering than to avoid it, personal distress can lead to prosocial behavior (Eisenberg, 2000; Eisenberg & Miller, 1987). On the other hand, when it is easier or less costly to avoid the victim's suffering than to allay it, personal distress is likely to lead to attempts to reduce one's exposure to the suffering. In both cases, personal distress involves an egoistic motivation to reduce one's own distress. Personal distress has also been associated with hostility toward the victim in terms of aggressive behavior, resentment, and suspicion (Davis, 1994; Miller & Eisenberg, 1988). Personal distress leads to an egoistic motivation to reduce one's own primarily self-oriented feelings of discomfort. Self-focused attention can lead communicators to focus on the rewards and costs to themselves of alternative courses of action, rather than on the standard of behavior which they "ought" to display. Consequently, I expect communicators of bad news who experience strong feelings of personal distress to focus on the material, social, or emotional costs to themselves. Though communication strategies, such as politeness and personalization, that show respect and concern for the recipient can soften the blow of bad news, communicators of bad news are rarely able to 21 alleviate all of the suffering of the news recipient. If the bad news is severe, the recipient is likely to experience pain and can react negatively. Empathic feelings of personal distress are more likely to lead a communicator to attempt to avoid the suffering of the recipient of bad news, as a means of reducing the communicator's own feelings of distress. Such avoidance or distancing can take several forms. First, the communicator might avoid acknowledging the harm caused (e.g., by not using mitigating explanations, such as apologies or excuses, which accept some responsibility for the negative outcome). Second, in order to reduce the duration of their exposure to the recipient's suffering, communicators might provide shorter, curter explanations (Folger & Skarlicki, 1998). Third, the communicator of bad news might derogate the victim as a way of reducing personal feelings of distress and sustaining a belief in a just world (Lerner, 1980). Hypothesis 2: Personal distress is negatively related to interactionally fair communication of bad news. Self-Esteem and Communication of Negative News Self-esteem refers to the favorability of an individual's self-evaluations (Coopersmith, 1967). In this section I focus on ways in which self-esteem affects how a person responds to having to communicate negative news, in terms of communicative behavior. Because I expect that experiencing a negative news situation from the victim's perspective can allay a defensive reaction in low self-esteem communicators, I also consider whether a person's self-esteem moderates the effect of empathic concern for a recipient of bad news on the interactional fairness with which the news is communicated. 22 Self-esteem has been discussed as a trait (i.e., as a predisposition) and state (i.e., situation-specific). Although traits have been shown to predict behavior, George (1991) proposed that states are more immediate precursors of behavior than are traits. In this research I focus on how state self-esteem affects the perception of and responses to perceived social threats associated with the need to communicate bad news. Trait and state self-esteem are both logically and empirically related (Brockner, 1988; Rosenberg, 1979). For example, a person's overall, or average, sense of self-esteem will generally help predict of feelings of self-worth in specific contexts. Because both forms of self-esteem affect how communicators respond to the social threats associated with the need to communicate bad news, I also incorporate into my discussion relevant research on trait-based self-esteem, realizing full well that an average or aggregate level of feelings of self-worth can be expected to vary considerably between specific contexts. Substantial evidence shows that self-esteem predicts how individuals deal with organizational challenges. Specifically, how employees perceive and respond to social and identity threats are often directed at enhancing or maintaining positive self-evaluations (Brockner, 1988). In general, activities that threaten a person's self-esteem tend to be avoided, relative to those activities that enhance one's self-esteem (or threaten it less substantially). Communicating negative news can pose significant threats to the communicator's self-esteem for two reasons. First, communicators fear being negatively evaluated by the recipient of the bad news, as this is likely to threaten a preferred self-impression of communicators (e.g., Folger & Skarlicki, 1998; Johnson, Conlee, & Tesser, 1974; O'Sullivan, 2000). Second, communicators can experience responsibility for the bad news they convey. Communicators who expect to be blamed for the negative news 23 communicated are likely to also anticipate social rejection by the recipient of the negative news (Leary, 2001). Research has shown that people desire social attachments, think often about their relationships, and put considerable effort into maintaining social bonds (Leary, 2001). Sommer (2001) suggested that interpersonal rejection represents a significant threat to a person's self-esteem. Even tacit, imagined, or anticipated interactions can activate a fear of negative evaluation and social rejection (Leary, Koch, & Hechenbleiker, 2001; Sommer & Baumeister, 2002). When faced with such a threat, a strong and secure belief in one's worth and efficacy is an important resource which leads high and low self-esteem individuals to respond very differently to the threat of failure and rejection. There are several reasons why communicating bad news can threaten the communicator's self-esteem. Because fair communication can sometimes call for close personal contact and more time with news recipients, it affords a perceived greater opportunity for the recipient to blame the communicator and directly express negative reactions. I suggested earlier that more interactionally fair communication is associated with more mitigating account strategies, such as excuses and justifications for actions and decisions, and greater use of politeness strategies (i.e., greater attention paid to the "face" of the news recipient). However, insofar as these strategies entail affirming the social identity of the news recipient, acknowledging the harm caused to them, and possibly accepting some of the responsibility for the negative event (e.g., by apologizing, or stating that the news recipient is not at fault), they are more likely to pose a threat to the positive social identity of the communicator than would strategies aimed at denying fault, derogating the victim, and maintaining the preferred self-impression of the 24 communicator. Thus, "distancing" oneself from the news recipient (i.e., communicating in interactionally unfair ways) allows the manager to reduce threats to his or her sense of self-worth. For example, by choosing to spend less time with a recipient of bad news a manager can reduce the victim's opportunity to criticize or blame him or her. Self-esteem and Expectations of Social Rejection The threat of negative evaluation by a recipient of negative news is likely to affect low and high self-esteem individuals differently due to differences in expectations for failure and interpersonal rejection, motivations when faced with possible negative evaluation, and behavior in response to actual, or possible, interpersonal rejection. High self-esteem is associated with higher expectations for performance in evaluative situations (Brockner, 1988), whereas low self-esteem is associated with lower expectations of success and the belief that future events will work out badly (Rosenberg, 2001). One way this is manifested is in the greater tendency of low self-esteem individuals to expect and perceive rejection (Sommer & Baumeister, 2002). Brown and Dutton (1995) suggested that low self-esteem individuals, who will generally have experienced more pain from esteem-threatening events than their high self-esteem peers, become hypervigilant to signs of social rejection. As a result, they are likely to perceive intentional rejection in the ambiguous behavior of relational partners (Downey & Feldman, 1996). In contrast, high self-esteem individuals are more confident that others regard them highly, which provides a buffer against occasional failures. Even when faced with actual or potential failure, high (versus low) self-esteem individuals have more. resources to refute threats to their self-esteem (Tice, 1993) and more confidence in their ability to perform esteem-enhancing behaviors (Brockner, 1988). 25 Self-esteem and Motivation Although both low and high self-esteem individuals want social approval, success, and belonging (Blaine & Crocker, 1993; Tice, 1993), the motivations underlying their behavior differ substantially when faced with social threats. First, because low self-esteem individuals are more likely to expect failure, their primary goal in most situations is to protect themselves against threats. This is manifested in a self-protective orientation, with a focus on minimizing the self-esteem consequences of failure or rejection, and protecting themselves against further distressing outcomes (Tice, 1993). One way in which low self-esteem individuals begin dealing with threats before they arise is by preparing for possible failure (Blaine & Crocker, 1993). In contrast, high self-esteem individuals are less concerned about possible failure (which they deem unlikely), and so they view evaluative situations more as opportunities for looking good and proving themselves. For this reason, high self-esteem individuals are more likely to take chances to enhance their self- and social image (Campbell & Lavallee, 1993). In contrast, low self-esteem individuals tend to avoid risks in order to, above all, protect themselves against further distressing outcomes (Tice, 1993). Another reason for the more self-protective, defensive orientation of low self-esteem individuals is the more substantial cost to them of social rejection. Because low self-esteem individuals have less confidence in their own worth, they rely more on external cues for affirmation, including the approval of relational partners (Brockner, 1988). In addition, social rejection is also more costly to low self-esteem individuals because of the attributions they are likely to make for negative events. Just as low and high self-esteem individuals differ in their expectations of success, they also differ in the 26 attributions they make for failure. Low self-esteem individuals tend to view failure as a result of stable, internal, and generalizable factors (Brockner, 1988). As a result, low self-esteem individuals regard negative feedback as self-diagnostic, and revealing of other aspects of their self-concept. In contrast, high self-esteem individuals tend to attribute negative events to unstable, specific, and external circumstances (Blaine & Crocker, 1993), thereby avoiding self-blame. Self-esteem and Interactionally Fair Communication Although low and high self-esteem individuals perform equally well following success, they differ substantially in the persistence, confidence, and vigor with which they respond to esteem-threatening situations. For example, people with high self-esteem respond to negative evaluation with attempts to prove the opposite (i.e., that they are worthy of social acceptance). Thus, high self-esteem individuals are more persistent (Brockner, 1988), increase their efforts at a task (Sommer & Baumeister, 2002), and perform better (Brockner, Derr, & Liang, 1987) than low self-esteem individuals after receiving negative feedback. Low self-esteem individuals, in contrast, have been shown to decrease their efforts on tasks following negative feedback (Sommer & Baumeister, 2002). Brockner (1988) suggested that low and self-esteem individuals are both affected by anticipated esteem-threatening evaluations, but in opposite directions, thereby reaffirming their low and high self-evaluations, respectively. Second, low self-esteem individuals are more likely than their high self-esteem peers to respond to interpersonal rejection by becoming more self-focused (Brockner, 1992), and derogating (Sommer, 2001; Baumeister, 1993), retaliating against (Rosenberg, 2001), and distancing themselves from the other party (Leary, 2002). Twenge, 27 Baumeister, Tice, and Stucke (2001) showed that actual or anticipated interpersonal rejection leads low self-esteem people to behave more aggressively. Sommer (2001) found that low self-esteem individuals can preemptively devalue a relationship as a means of protecting themselves when they believe the relational partner may reject them. Rosenberg (2001) referred to such interaction restriction as one aspect of the "moat mentality," whereby low self-esteem people protect themselves from possible rejection and negative evaluation. This contrasts with the tendency of high self-esteem individuals to both inflate their partner's regard for them and increase their commitment to the relationship when faced by the possibility of relational devaluation (Sommer, 2001). Building on the previous research, I propose that self-esteem affects communicators' expectations, motivations, and behavior in respects associated with interactional fairness. First, low self-esteem communicators are more likely than their high self-esteem counterparts to expect negative evaluation from the recipients of bad news, and more likely to expect negative evaluation to lead to interpersonal rejection. Second, given that people tend to avoid esteem threatening situations, I expected that low self-esteem communicators adopt a more defensive orientation, such that their communication strategies are aimed primarily at avoiding failure, rejection, or other distressing outcomes. Hypothesis 3: Self-esteem is positively related to interactionally fair communication of bad news. 28 Self-Esteem as a Moderator between Empathic Concern and Interactionally Fair Communication The important differences between low self-esteem and high self-esteem communicators are also expected to affect the relationship between the empathic concern dimension of empathy and interactionally fair communication. As discussed above, empathic concern is (a) associated.with more active perspective-taking, and (b) characterized by feelings, specifically of warmth, sympathy and concern, for the victim. I propose that empathic concern has a stronger relationship with interactionally fair communication for individuals who are low (versus high) in self-esteem. Low self-esteem communicators are more likely (than their high self-esteem counterparts) to derogate and distance themselves from the recipients of bad news in order to protect their own feelings of self-worth. Given this tendency of low self-esteem communicators, empathic concern becomes an especially important predictor of interactionally fair communication. Because empathic concern involves a communicator adopting an other-orientation, it makes it more difficult for low self-esteem individuals to derogate or emotionally distance themselves from the news recipient, as they might otherwise do. In this way, the tendency for empathic concern to lead to more other-oriented behavior will counteract the self-protective tendencies of low self-esteem individuals. The relationship between empathic concern and interactionally fair communication is expected to be less strong for high (versus low) self-esteem communicators for several reasons. First, high self-esteem individuals will anticipate less negative reactions (to themselves) as a result of communicating the negative news. Folger and Skarlicki (1998) found that when communicators of bad news anticipated less blame 29 for bad news, they were less likely to distance themselves from the victim. Second, high self-esteem communicators are more confident in their ability to both communicate the bad news effectively, and to overcome the reputational consequences of being associated with or even blamed for the news. As a result, even in the absence of empathic concern, high self-esteem communicators will tend not to act defensively in terms of derogating the victim or devaluing the relationship. Rather, high self-esteem communicators will, regardless of their level of empathic concern, attempt to overcome the anticipated negative evaluation by the bad news recipient. For example, they are more confident than low self-esteem communicators that by offering detailed and reasonable explanations and showing concern and respect for the victim they can lessen the potential for victim outrage, strengthen the relationship, and avert interpersonal rejection. Hypothesis 4: Self-esteem moderates the relationship,between empathic concern and interactionally fair communication of bad news. Specifically, the positive relationship between empathic concern and interactionally fair communication is stronger for those individuals whose self-esteem is low versus high. Moral Development and Communication of Negative News Research has begun to examine the role played by the ethical and moral reasoning of individuals in their justice-related behavior. According to Folger's (2001) deonance model of justice, individuals care about justice not only because of economic or relational motives, but also as an important end in itself. Deonance is derived from the Greek word deon, which refers to one's moral obligations. Deonance theory treats justice as an "ought," and suggest that people value fair treatment because of a respect for moral rules by which all people are expected to live. From a deontological perspective, acts are 30 regarded as fair or unfair to the extent that they conform to a priori standards of ethical behavior, including principles of fair treatment. The moral principles underlying deontological justice judgments involve noncomparative justice, in the sense that what a person is due is determined by a universal standard, rather than by comparison with other people (Feinberg, 1974). Bies (2001) suggested that interactional justice is an important noncomparative form of justice, in that a person's sense of injustice is likely to be aroused by deception, disrespect, or abusive words or actions - even if the treatment accorded to other people is similarly unfair. To the extent that moral reactions to injustice have been considered in the research literature, they have been considered from the perspective of the justice recipient (i.e., the victim of an unfair outcome), or from the perspective of third-party observers (e.g., Skarlicki et al., 1998). For example, Cooley's (1902) classic research on queue jumping showed that people not only react strongly against a person who cuts in line in front of them, but also against someone who cuts in line behind them. My interest, however, lies in the perspective and motivations of the person whose behavior might be regarded by others as unfair. From this perspective, this research focuses on the factors that influence how a transgressor (i.e., the queue-jumper) perceives the fairness of his or her own behavior. In particular, I consider how a communicator's level of moral development can influence his or her own communication behavior, both directly and by affecting the relationship between communicator empathy and interactionally fair communication. Folger (1998) proposed that a link exists between Kohlberg's (1984) three levels of moral development and different models for justice motivations. The pre-conventional 31 level of moral development, at which individuals are motivated by material rewards and punishment to act in their own best interests, corresponds to models of justice which focus on self-interested, instrumental motivations, (e.g., Adams, 1965). The conventional level of moral development is characterized by following social rules and gaining the approval of others, as in the group value or relational approaches to justice (e.g., Lind & Tyler, 1988). The final stage of moral development, the post-conventional level (which corresponds to the deonance model of justice), is characterized by respect for universal principles of right behavior and a genuine interest in the wellbeing of others. Individuals tend to predominantly engage in types of moral reasoning that are consistent with the highest level of moral development they have attained (Kohlberg et al., 1983). Moral Development and Interactionally Fair Communication I propose that a communicator's moral development predicts the choices made when conveying negative news. Specifically, the increasing regard paid by individuals at higher levels of moral development to moral principles and to the general welfare of others will influence certain tradeoffs made by communicators of bad news between themselves and the news recipient. In particular, interactionally fair communication can acknowledge hardship, accept some responsibility for the harm caused, provide explanations for the harm caused, and thereby focus on the social identity of the news recipient. These aspects of fair communication can be costly to the communicator in terms of exposing them to the blame and moral outrage of the bad news recipient (Bies, 1986). In such a situation, the willingness of the communicator to communicate in an interactionally fair manner may depend in part on the importance to them of universal principles of ethical behavior. 32 Research has shown that higher level moral reasoning is associated with a variety of ethical actions and motivations, including prosocial behavior and altruistic goals (Eisenberg, 1986). Moral development is predicted to be associated with interactionally fair communication for two reasons. First, the greater regard for others' welfare that is characteristic of higher levels of moral development will be reflected in communication strategies which are directed at protecting the identity and emotions of the bad news recipient. Second, individuals at higher levels of moral development will be influenced more by universal rules regarding appropriate and respectful interpersonal treatment than will individuals at lower levels, for whom material, identity, and social considerations will be predominant. High moral development communicators will engage in interactional justice because maintaining the others' dignity and respect is the right thing to do. Hypothesis 5: Moral development is positively related to interactionally fair communication of bad news. Moral Development as a Moderator between Empathic Concern and Interactionally Fair Communication A communicator's level of moral development is also expected to moderate the effect of communicator empathy on interactionally fair communication. I consider that an individual's (in this case a communicator of bad news') care and concern for others' rights in general can sometimes supersede other motivations, including affective reactions to the suffering of others in a specific situation. Moral development is likely to reduce the influence on interactionally fair communication of a communicator's own feelings of empathic concern. Higher levels of 33 moral development will result in the application of ethical standards of behavior which are universally applicable (i.e., principles, regardless of feelings prompted by the specific individuals involved) (Kohlberg, 1983). For example, a person may believe that people ought to be honest in their interactions with others, no matter what the consequences are. When such a universal moral value is salient to the communicator, other-oriented feelings of warmth and concern for a specific individual (i.e., empathic concern) are likely to be less important influences on behavior. I also expect the moral aspect of the self-concept to exercise a more powerful influence on behavior (and vis-a-vis other motives for behavior) when individuals are at the postconventional level of moral development than when they are at the preconventional or conventional levels of moral development. Moral development in these individuals (i.e., their general concern for others and sense of principled moral obligations) is likely to dominate concern for themselves or for specific other individuals. Hence, moral development is expected to offset the effect of communicator empathic concern on interactionally fair communication of negative news. Hypothesis 6: Moral development moderates the relationship between empathic concern and the interactionally fair communication of bad news. Specifically, the positive relationship between empathic concern and interactional fairness is stronger for those individuals who are low versus high in moral development. Emotional Intelligence and Communication of Negative News A growing body of research has begun to explore the role of emotion in organizations and work life (e.g., Ashkanasy, Hartel, & Zerbe, 2000), including individual and organizational outcomes related to emotional intelligence. The idea of 34 emotional intelligence presumes that it is possible to be intelligent with and about emotions, and that the ability to accurately perceive and effectively work with emotions is an important individual difference. Emotional intelligence is a subset of social intelligence (Salovey & Mayer, 1990), a term first introduced by Thorndike in 1920 to indicate another domain (versus abstract reasoning), specifically focused on affect, in which intelligent reasoning could be manifested (Landy, 2005). Because the communication of negative news can be expected to arouse strong emotions in the communicator, I investigate how emotional intelligence relates to fair communication. Emotional intelligence was first defined by Salovey and Mayer (1990), as an individual's ability to accurately perceive emotions, to understand emotions, and to effectively manage emotions in self and others. However, following the popularization of the term emotional intelligence by Goleman (1995), some models of emotional intelligence broadened the term considerably, to include traits, competencies, and abilities that do not primarily involve reasoning with or about emotions. These models have been referred to as "mixed" models because they overlap personality and other models. For example, the Emotional Competence Inventory, Version 2 (Goleman, Boyatzis & McKee, 2002) includes measures of empathy and optimism, as well as competencies, such as flexibility in handling change, that are not necessarily emotional (McEnrue & Groves, 2006). Likewise, the EQ-i measure (Bar-On, 1997) includes elements, such as happiness and problem solving, that do not necessarily relate to the intelligent use of emotion. Research has found that mixed models such as these substantially correlate with personality (Rosete & Ciarrochi, 2005). In this research, I use the ability model which focuses strictly on problem-solving within the emotional domain. 35 The Ability Model of Emotional Intelligence According to the ability model of emotional intelligence (Salovey & Mayer, 1990), emotional intelligence refers to an individual's ability to recognize the meanings of emotions and their relationships and to effectively use emotions for reasoning and problem solving. By this approach, emotional intelligence is measured as an ability to solve problems with and about emotions. The ability model of emotional intelligence originally proposed by Salovey and Mayer (1990) demonstrates solid convergent and discriminant validity to support its claims to be an intelligence (Daus & Ashkanasy, 2005), including low correlations with personality and low (positive) correlations with cognitive intelligence. The ability model proposes that four factors of emotional intelligence exist. The first factor, Perceiving Emotions, refers to the ability for a person to accurately perceive their own emotions and to accurately perceive how other people are feeling, through recognizing emotional signals in their facial expressions. The second factor, Using Emotions to Facilitate Thought, refers to a person's ability to effectively use emotions for reasoning and problem solving, including matching an appropriate mood to a particular task. The third factor, Understanding Emotions, includes the ability to label emotions, to recognize how different emotions are related and can combine, and to understand the antecedents of emotions and how they change over time. Managing Emotions, the fourth factor, refers to how well a person can manage his or her own and other people's emotions. The four factors combine into two distinct types of emotional skills: experiential emotional intelligence, and strategic emotional intelligence. Experiential Emotional 36 Intelligence comprises the first and second factors, and refers to a person's ability to perceive and use emotional information, without necessarily understanding it. As such, experiential EI focuses on basic-level processing of emotions. In contrast, Strategic Emotional Intelligence, which comprises the third and fourth factors, involves higher-level, conscious processing of emotions. Specifically, Strategic Emotional Intelligence refers to a person's ability to anticipate emotions, to understand how emotions arise, combine, and develop over time, and to manage one's own and others' emotions in an effective manner - without necessarily perceiving them accurately or fully experiencing them. Experiential emotional intelligence will help a person read the emotions of another person, and use his or her own emotions effectively. In contrast, strategic emotional intelligence is likely to be especially important where a person is imagining what his or her emotions and the emotions of another person are likely to be in a situation. Therefore, strategic emotional intelligence (versus experiential emotional intelligence) should better predict a person's reactions to a hypothetical situation in which emotions are stirred, such as the scenario exercise used in Study 1. Reasoning with emotions plays an important role in developing and maintaining workplace relationships (Ashkanasy, 2003; Lopes & Salovey, 2001). Higher emotional intelligence has been associated with improved interpersonal relationships at work (Wong & Law, 2002), and better peer and supervisor ratings of interpersonal facilitation (Lopes, Brackett, Nezlek, Schutz, & Salovey, 2004). Emotional intelligence can also be a key contributor to managerial, and especially leader, effectiveness (George, 2000). Mayer and Caruso (2002) suggest that leaders who are high (versus low) in emotional intelligence are able to communicate more effectively. A manager who perceives emotions more 37 accurately and reasons more effectively with emotions may often be in a better position to manage change, including the emotional reactions of employees (Mayer & Caruso, 2002). Experiential emotional intelligence is likely to be especially important in face-to-face situations when facial expressions can provide important emotional information. Because strategic emotional intelligence includes the ability to accurately anticipate emotions and predict how they are likely to evolve over time and can most effectively be managed, it is expected to be particularly appropriate for a situation in which an individual is anticipating the emotional reactions of another person and planning how best to manage them. Emotional Intelligence and Interactionally Fair Communication I propose that strategic emotional intelligence relates positively to interactional fairness in communicating negative news. First, individuals with higher strategic emotional intelligence are better able to anticipate the emotions likely to be felt by a recipient of negative news. Thus, an emotionally intelligent communicator is more accurate in anticipating how and why the recipient will emotionally react to negative news, and what emotional displays can ameliorate, versus exacerbate, the negative emotional reaction. As a result, a manager with high strategic emotional intelligence is . better able than a manager with low strategic emotional intelligence to plan and take steps to reduce a recipient's negative emotions in response to negative news, such as by providing reasonable explanations for the negative news, and by showing respect and concern for the recipient. Second, individuals with higher (versus lower) strategic emotional intelligence are better able to productively anticipate and productively manage their own emotions, 38 including fear, guilt, or anxiety they might expect to experience. Higher (versus lower) strategic emotional intelligence means that a person can regulate their emotions, and remain open to emotional information at important times and closed to it at other times. Because strategic emotional intelligence involves anticipating and working with feelings in judicious ways, rather than simply acting on feelings without considering their likely effect, it is likely to lead a communicator to use and to display emotions that reduce negative reactions from the recipient of bad news to the news and the communicator. Hypothesis 7: Strategic emotional intelligence is positively related to interactional fair communication of bad news. 39 Summary of Study 1 Hypotheses When delivering negative news to an employee a manager's: 1. Empathic concern for the recipient(s) is positively related to interactionally fair communication of bad news. 2. Personal distress is negatively related to interactionally fair cornmunication of bad news. 3. Self-esteem is positively related to interactionally fair communication of bad hews. 4. Self-esteem moderates the relationship between empathic concern and interactionally fair communication of bad news. Specifically, the positive relationship between empathic concern and interactionally fair communication is stronger for those individuals whose self-esteem is low versus high. 5. Moral development is positively related to interactionally fair communication of bad news. . 6. Moral development moderates the relationship between empathic concern and interactionally fair communication. Specifically, the positive relationship between empathic concern and interactional fairness is stronger for those individuals who are low versus high in moral development. 7. Emotional reasoning is positively related to interactionally fair communication of bad news. 40 Study 1 Methods In Study 1,1 tested the effect on interactionally fair communication of several important individual differences: empathy, self-esteem, moral development, and emotional intelligence. Study 1 involved a scenario exercise in which practicing managers were asked to write out how they would communicate negative news to an organizational member. Participants A list of organizations was obtained from the Coop Education office at the Sauder School of Business. I sent a letter describing the study to human resource managers at 12 companies and one professional association. I followed up by phone. The human resource managers made my email address available to their managers and the managers were invited to contact me if they were interested in participating. Two hundred and forty-seven managers from a broad range of industries, including resource firms, financial institutions, educational institutions, high-tech companies, and service organizations, volunteered to participate. One hundred and seventy-three managers (70.0%) completed the questionnaires. The participants in the final sample were from 148 different organizations. On average, they supervised 27.3 employees, were 44 years old, with 10.0 years of managerial experience, and 34.1% were women. Twenty-seven percent of the managers had completed a graduate degree, 44.2% an undergraduate degree, 23.6% a college diploma, and 5.5% percent high school. Procedure The managers were provided with a consent form to complete and return (see Appendix A) and with directions for completing the online questionnaires. Three 41 questionnaires containing measures of the predictors and a scenario exercise were then administered online in the following order to the managers. First, the managers completed the Defining Issues Test of moral reasoning (DIT-2; Rest, Narvaez, Thoma, & Bebeau, 1999). Second, the managers completed the MSCEiT V2.0 test of emotional intelligence (Mayer et al, 2003). Third, the managers completed a questionnaire containing demographic questions, and measures of empathic concern, personal distress, self-esteem, and the control variables. Participants then read a scenario in which negative news (i.e., a layoff) had to be communicated to an employee. They were asked to write out the message that they would use to communicate a layoff to the employee, whose name was Jim. The scenario was designed such that the organization (through poor managerial decisions) and the employee (through mediocre performance) could both be partly held to blame for the decision to lay Jim off. To avoid attribution biases, the news recipient was neither (a) an innocent victim, who thereby demanded sympathy, or (b) wholly to blame, which should lead to less interactional fairness (Feather, 1999). Thus, a situation was presented in which a range of communication strategies, both interactionally fair and unfair, could be considered reasonable under the circumstances. The scenario also explained that Jim could experience considerable hardship as a result of the layoff, because he and his wife had just put a down payment on a house and suitable alternative employment would be difficult to find. This was done to ensure that the managers understood that the news was likely to be negative and significant for the employee. 42 Measures Predictor variables Unless otherwise indicated, response sets consisted of 7-point Likert-type scale, ranging from 1 (strongly disagree) to 7 (strongly agree). For each variable, the individual item scores were averaged to form the variable score. Larger values signify greater amounts of the variables than lower numbers. Trait Empathic Concern. Participants indicated the extent to which they characteristically engage in the empathic concern dimension of empathy. Empathic concern was measured using 3 items from the Empathic Concern subscale of Davis' (1980) Interpersonal Reactivity Index: "I often have tender, concerned feelings for people less fortunate than me," "When I see someone being treated unfairly, I sometimes don't feel very much pity for them, (reverse coded)" and "Sometimes I don't feel very sorry for other people when they are having problems" (reverse coded). Trait Personal Distress. Participants indicated the extent to which they characteristically engage in the personal distress dimension of empathy. Personal distress was measured using 4 items from the Personal Distress subscale of the Interpersonal Reactivity Index (Davis, 1980): "In emergency situations, I feel apprehensive and ill-at-ease," "When I see someone get hurt, I tend to remain calm" (reverse coded), "When I see someone who badly needs help in an emergency, I go to pieces," and "I sometimes feel helpless when I am in the middle of a very emotional situation." State Self-esteem. State self-esteem was measured using 8 questions from Heatherton and Polivy's (1991) State Self-Esteem scale which ask how a person feels about himself or herself in a specific situation. The State Self-Esteem scale focuses on 43 actual self-esteem versus mood and is more sensitive to temporary changes in self-esteem than traditional self-esteem measures (Heatherton & Polivy, 1991). Items include: "Frustrated and rattled about my performance," "Like I am not doing well," "Confident about my abilities" (reverse-coded), "Confident that I understand this situation" (reverse-coded), "Worried about whether I am regarded as a success or a failure," "Self-conscious," "Worried about looking foolish," and "Worried about what other people think of me." Moral development. Moral development was measured using the N2 measure from the Defining Issues Test, Version 2 (DIT2) (Rest et al, 1999). The DIT2 is a measure of moral judgment derived from Kohlberg's theory (Kohlberg, 1984). The DIT2 consists of five paragraph-length hypothetical dilemmas, each followed by 12 issues representing thinking at different stages of moral reasoning. Participants rate and rank the 12 issues in terms of their importance to making a decision in response to the dilemma. The N2 score indicates the degree to which a subject uses higher stage moral reasoning, i.e., reasoning at the Postconventional stage, rather than at the Personal Interest or Maintaining Norms stages. The N2 score improves on the p-score (postconventional reasoning score) on six criteria for construct validity (Rest, Thoma, Narvaez & Bebeau, 1997). In addition to measuring the extent to which postconventional items are prioritized (i.e., the p-score), the N2 score measures the degree to which lower stage items selected are at the Personal level versus the Conventional level. Rest and colleagues (1999) reported Cronbach alpha reliability of .81 for the N2 index. Because ranking data is used in the calculation of the N2 score, individual items cannot be used as the unit of internal consistency. Therefore, the unit of internal reliability is at the story level. 44 Strategic Emotional Intelligence. The MSCEIT V2.0 (Mayer, Salovey & Caruso, 2002) was used to assess strategic emotional intelligence. The MSCEIT V2.0 is an ability measure of emotional intelligence which uses two tasks to measure each of the four dimensions of emotional intelligence: perceiving emotion, using emotion, understanding emotional information, or managing emotions. The MSCEIT V2.0 contains 141 items. To measure Strategic Emotional Intelligence, the MSCEIT V2.0 includes 61 questions about emotional vocabulary, how emotions blend together, how emotions change over time, and what would be the best course of action in specific social situations to improve a feeling or resolve a problem. Mayer and colleagues (2002) reported Cronbach alpha reliabilities of a = 0.91 for the full strategic emotional intelligence scale, and 0.77 and 0.81 for the emotional understanding and emotional management subscales, respectively, which together comprise Strategic Emotional Intelligence. Criterion variable Interactional justice. The interactional fairness of managers' written messages was content coded independently by two students who were blind to how participants responded to other survey measures. For each message, they rated on a 5-point Likert scale whether the communicator 1) was polite and courteous, 2) treated the recipient with respect, 3) expressed concern, 4) provided adequate explanations, and 5) accepted some responsibility for the negative outcome. The two coders were given three hours of training in using the guidelines for coding interactional justice. The coders then independently rated messages from a pilot sample of M B A students (A7 = 20). Their ratings on the pilot messages were compared. Areas of disagreements were discussed with the researcher and used to help clarify and 45 elaborate the coding guidelines. Each of the 173 messages from the participants in this study was then rated independently by the two raters. The scores for the two raters were averaged for each of the five items, and then the item scores were summed to form the interactional justice measure. Control variables Demographic variables. Three demographic variables were controlled: gender, age, and whether the managers had themselves experienced being fired. Communicator gender was controlled for because women tend to provide longer and more complex explanations for negative events and are more likely than men to accept responsibility (Gonzales et al., 1990). Thus, I expected interactional fairness to be greater for female managers. In addition, females score slightly higher across all ages on the DIT2. Female managers were coded 1, and male managers were coded 2. Because moral development tends to increase with age (Rest, 1994), I controlled for managers' age. Whether the managers had been laid off in the past (1 = yes, 0 = no) was assessed and controlled. I expected managers who had experienced a layoff to experience greater victim empathy than managers who had never been laid off (Kray & Lind, 2002). Study 1 Results The descriptive statistics and correlations between the measures are reported in Table 1. Scale reliabilities in terms of Cronbach's alpha are given on the diagonal. The level of agreement between the two raters for the 5-item interactional justice score, as measured by an intraclass correlation coefficient (McGraw & Wong, 1996; Shrout & Fleiss, 1979), was .89 (ICC). Interactional justice scores ranged from 6.00 to 24.00, with a mean score of 15.71 and a standard deviation of 4.26. 46 The hypotheses were tested using hierarchical regression analysis. Table 2 presents the results when interactional justice was regressed on all of the predictor variables and hypothesized interactions. Tables 3 through 10 present the results of the hierarchical regression analyses for individual predictors. The control variables were entered in at Step 1. Direct effects were entered in at Step 2. Hypothesized interactions, if any, were entered in at Step 3. For completeness, I report for each hypothesis two results: first, when all hypothesized main effects and interactions (and the control variables) are regressed on interactional justice, and, second, when only the individual predictor (and the control variables) is regressed on interactional justice. Hypothesis 1 predicted empathic concern is positively related to interactional justice. When only empathic concern is regressed on interactional justice, empathic concern is positively related to interactional justice, fi=.16,p< .05. See Table 3. However, as shown in Table 2, when all predictors and proposed interactions are regressed on interactional justice, empathic concern is not related to interactional justice, P = .16, ns. Thus, Hypothesis 1 was partially supported. Hypothesis 2 predicted that personal distress is negatively related to interactional justice. Personal distress was not related to interactional justice when only personal distress was included in the regression equation, = .03, ns, or when all of the predictors and proposed interactions were included in the regression equation, /? = -.01, ns. See Table 2 and table 4. Thus, Hypothesis 2 was not supported. Hypothesis 3 stated that self-esteem is positively related to interactional justice. Self-esteem was not related to interactional justice when only self-esteem was regressed 47 on interactional justice, /? = .03, ns, nor when all of the predictors and proposed interactions were included in the regression equation, fj = .06, ns. See Table 2 and 5. Thus, Hypothesis 3 was not supported. Hypothesis 4 predicted that the relationship between empathic concern and interactional justice would be stronger for communicators with low (versus high) self-esteem. Table 6 shows that Hypothesis 4 was not supported, whether all of the other predictors and proposed interactions were included in the regression equation, /? = .02, ris, or only self-esteem, empathic concern, and the proposed interaction were included in the regression equation, /? = .01, ns. Table 7 shows that self-esteem did not relate to interactional justice for either low or high empathic concern communicators. Hypothesis 5 predicted that moral development is positively associated with interactional justice. Table 8 shows that when only moral development is regressed on interactional justice, moral development is positively related to interactional justice, ft = A6,p<.05. However, as shown in Table 2, when all predictors and proposed interactions are regressed on interactional justice, moral development is not related to interactional justice, P = .12, ns. Thus, Hypothesis 5 was partially supported. Further, as predicted by Hypothesis 6, the two-way interaction between empathic concern and moral development in predicting interactional fairness was significant when all predictors and proposed interactions were included in the regression equation, ft = -.15, p < .05, and when only empathic concern, moral development, and the interaction between them were regressed on interactional justice, /? = -.15, p < .05. See Table 2 and Table 9. The interaction was probed using the procedures recommended by Aiken and West (1991). As shown in Figure 2 and Table 10, in the case of low moral development 48 communicators there is a positive relationship between trait empathic concern and interactional justice, £(172) = 2.85, p < .01. In the case of high moral development communicators, trait empathic concern did not predict interactional justice, £(172) = -.26, ns. Thus, Hypothesis 6 was supported. Hypothesis 7 predicted that strategic emotional intelligence is positively associated with interactional justice. Table 11 shows that when the individual predictor is regressed on interactional justice, strategic emotional intelligence is positively related to interactional justice,/? = .15,/? < .05. However, as shown in Table 2, when all predictors and proposed interactions are regressed on interactional justice, strategic emotional intelligence is not related to interactional justice, {3 = .13, ns. Thus, Hypothesis 7 was partially supported. Study 1 Discussion In Study 1,1 investigated the effects of four theoretically relevant individual difference factors on interactional fairness: communicator empathic concern and personal distress, self-esteem, moral development, and strategic emotional intelligence. I proposed that each of these factors can help explain why some managers communicate negative outcomes in an interactionally fair manner, whereas others will tend to show less respect and give less adequate explanations. Study 1 highlighted the important role that empathic concern plays in a manager's interactional fairness. As predicted, the tendency to feel sympathy and concern in response to another person's anticipated or observed suffering related positively to interactional fairness in communicating negative news. Empathic concern predicted interactional justice, in terms of sensitivity and respect and detailed explanations, in spite 49 of the fact that the comrnunicator could expect to partially be blamed for the harm to the recipient. Further, the desire to reduce the recipient's suffering would seem to be altruistically motivated, in that the more detailed and sensitive explanations which characterize interactional fairness would increase the personal costs to the communicator, in terms of their increased exposure to the recipient of the bad news. Whereas empathic concern was positively related to interactional fairness, there was no significant relationship between a manager's tendency to experience personal distress and how they communicated the negative news. There are several possible explanations for why personal distress did not relate to interactional justice. First, Hoffman (2000) suggested that personal distress results only when the empathic feelings are so strong and uncomfortable that they shift the observer's attention away from the victim to themselves. It is possible that the scenario exercise did not prompt sufficiently strong feelings to cause state personal distress, even in individuals who had a characteristic tendency to feel personal distress when faced with the suffering of another person. Second, trait personal distress and trait empathic concern were highly correlated in Study 1, r = .53, p < .001. This is not surprising given that both factors are caused by seeing a person in need and are affected by the perceived magnitude of the need (Batson et al., 1987). Both perspective-taking conditions could be expected to result in strong feelings for the recipient, though personal distress was also expected to elicit strong self-oriented feelings which shifted the focus of the communicator. Thus, any feelings of personal distress were likely to have been accompanied by feelings of empathic concern, leading to a mix of altruistic and egoistic motivation (Batson et al., 1997), versus a purely self-oriented reaction. Third, personal distress results in motivation to reduce one's 50 exposure to the other person's suffering. In the scenario exercise, contact with the recipient of the bad news could not be avoided by escaping the situation, which high personal distress individuals are wont to do. As a result, high personal distress communicators might have focused more on reducing the recipient's negative reaction to the news than they would have if the situation had allowed them to distance themselves more from the bad news recipients. The hypothesized positive relationship between communicator self-esteem and the interactional justice of their messages was not observed. At least three reasons might explain these results. First, given the tendency of high self-esteem individuals to expect to perform well, be positively evaluated, and avoid social rejection (Brockner, 1988), it may have been difficult for high-self esteem managers to easily relate to or imagine themselves being laid off for poor performance. Second, the scenario was such that blame for the negative outcome could be attributed to either the victim or the organization and the communicator. In order to protect their own positive self-regard, high self-esteem managers might have avoided accepting partial responsibility for the layoff, thereby reducing the detail and reasonableness of the explanations they provided for the news. Third, in the scenario, where the layoff would sever the relationship between the manager and the recipient of the news, the threat and consequences of a negative reaction to how the news was communicated might not have been sufficiently great or immediate to create a situation where high self-esteem individuals performed better, in working hard to refute and overcome a potential negative evaluation. 51 Study 1 Theoretical Implications Study 1 related a manager's interactional fairness to his or her level of moral reasoning, providing support for a deontological motivation for fair communication. The idea of justice as a moral virtue suggests that some people are fair simply because it is the right thing to do (Cropanzano, Rupp, Mohler, & Schminke, 2001). In the case of communicating negative news, the application of universal moral principles regarding how victims of bad news ought to be treated resulted in more sensitive strategies and more adequate explanations (i.e., interactional justice). The measure of moral development used, the Defining Issues Test 2, is a well-established measure. Support for the moderator hypothesis adds an important element to understanding when empathic concern for the news recipient will exercise a greater, positive impact on interactional fairness. Specifically, higher (versus lower) trait empathic concern was positively related to interactional fairness only when moral development was low (versus high). When moral development was high, trait empathic concern was not related to interactional fairness. Empathic concern and moral development functioned as substitutes for each other with respect to interactional fairness. There was no additional benefit in terms of interactional fairness when the communicator was high in both moral development and empathic concern. Figure 2 shows an ordinal interaction between empathic concern and moral development, with the interactional justice of high moral development communicators above that of low moral development communicators. Finally, the findings regarding the effect on interactional fairness of strategic emotional intelligence add to our understanding of the role emotions can play in justice decisions and behavior. The role of emotion in justice reasoning is underresearched (Barclay, Skarlicki, & Pugh, 2005), with no research to my knowledge directly relating emotional intelligence to interactionally fair treatment of others. Communicating serious negative news, such as a layoff, can be expected to stir emotions in both communicator and recipient, including strong negative emotions, such as anxiety, fear, anger, or guilt. Study 1 shows a manager's capacity to anticipate and understand the likely emotional outcomes related to their use of fair communication strategies, aimed at ameliorating their own and the recipient's negative emotions. The measure of emotional intelligence used, the MSCEJT V.2, focused on the ability to solve problems about and with emotions, and has been found not to correlate with personality measures or self-reported emotional intelligence (Daus & Ashkanasy, 2005). Further, the strategic emotional intelligence dimension specifically measures the ability to accurately anticipate emotions and predict how they are likely to evolve over time and can most effectively be managed, which is particularly appropriate to a situation in which an individual is planning future communication of negative news. In situations where a person is actually delivering the message face-to-face, and where the communicator has to respond to the emotional reactions of the bad news recipient, the effect on interactional justice of the ability to accurately perceive emotions and productively use one's own emotions (i.e., experiential emotional intelligence) should be investigated. Study 1 Strengths and Limitations One of the strengths of Study 1 pertains to the interactional justice measure. The high level of interrater agreement in the interactional justice of the messages suggests that the coding procedures provided a reliable measure of interactional justice. Second, participants in our study were asked to compose a message, rather than, for instance, in 53 other studies where they were asked to choose from an array of strategies provided for them. The approach taken in the present research has been found to more accurately predict communicative behavior (Clark, 1979). It is important, however, to acknowledge that written messages do not capture all aspects of interactional justice. For example, the tone and body language used to communicate negative news in face-to-face situations will also influence recipient perceptions of concern and respect, and the perceived sincerity and adequacy of the explanations provided. It is therefore important in future research to test the relationship between individual differences in managers and measures of interactional fairness which relate to face-to-face delivery of bad news. It is also important to note that the use of a scenario exercise might have overstated the effect on interactional justice of the factors tested for the following reasons. First, because a scenario exercise might not arouse the strong sentiments that a live situation could, the effect on interactional justice of cognitive reasoning structures might be increased, relative to situations in which moral principles can compete with strong, emotional reactions. Second, the ability to solve justice problems using moral principles is only one contributor to moral action. The Four Component Model (Rest, 1986) suggests that moral sensitivity, motivation, and character are necessary to translate a judgment of what is morally right or wrong into action that implements the judgment. For example, a manager who is high in moral reasoning might not interpret a situation as a moral one, and thereby avoid engaging in higher level moral reasoning. Alternatively, a manager might use postconventional reasoning to determine what is morally right, but lack the character or motivation to take moral action, especially if there is competition 54 from other personal values or identities. Thus, it is possible that moral judgment can have a greater effect on planned communication than on actual communication, which would require additional components of the Four Component Model to translate moral judgments and intentions into action. While the use of a scenario in Study 1 provided a controlled test of the hypotheses, it is a challenge to create vivid and realistic scenarios wherein the communication of negative news poses a threat to the self-image of participants. On the other hand, several reasons exist to expect the results to generalize to managers in organizational contexts. First, our participants were practicing managers from a variety of industries. Second, a recent meta-analysis showed that few justice relationships differ substantially in magnitude or direction between experimental and field studies (Cohen-Charash & Spector, 2001). Third, managers would not have been aware of the criteria by which their messages would be coded, making it unlikely that a demand effect or social desirability bias affected the results. The interactional fairness of a manager toward a subordinate can be influenced by the reactions of the subordinate (Korsgaard et al., 1998). Because the managers in this study did not have to face the likely negative and emotional reactions of the bad news recipient, they may have overstated how much interactional fairness they would use if actually communicating to the news recipient. An important next step would be to test how managers when faced by the need to convey bad news to an employee actually use the strategies, language, and explanations identified in these messages. Other potential methodological weaknesses warrant discussing. The data were cross-sectional which does not permit me to discuss direction of causality. The 55 Cronbach's alpha for the empathic concern scale was low, which is often found when measuring empathy (Davis, 1980). Cortina (1993) noted the importance of taking into account the average item intercorrelation and dimensionality when interpreting coefficient alpha. A principal component analysis of our scale revealed a single factor in the empathic concern variable, with all item loadings above .70. In addition, the average item intercorrelation for the empathic concern scale was an acceptable .30. Therefore, the alpha for our empathic concern variable should be interpreted in light of its item intercorrelation and its single-factor structure. Finally, low reliability is likely to increase sampling error variances and thereby reduce the magnitude of observed relationships from their true value (Viswesvaran, Ones, & Schmidt, 1996). In the present case, significant relationships were nonetheless found between empathic concern and interactional justice in spite of the low alpha of empathic concern. Last, the present study focused on the layoff context, which can differ in important regards from other situations in which negative news is communicated to employees, and involved partial communicator responsibility for the bad news. First, a layoff ends the relationship between the employee and manager, and managers may have been less motivated to maintain a good relationship than they would have been if they were going to continue working with the employee. Second, the layoff in the scenario presented had very serious negative consequences for the news recipient. Future research should investigate factors affecting interactional fairness in contexts in which the bad news needs communicating amid ongoing relationships and/or is less serious in nature. Further, the scenario was one in which participants were specifically told that they and the organization were partly to blame for the layoff. Folger and Skarlicki (1998) showed 56 that communicators are more likely to distance themselves from recipients of negative news when they expect to be blamed for it. Therefore, future research should investigate the effect on interactional fairness of empathy, self-esteem, moral development, and emotional intelligence in situations where the communicator of negative news is unlikely to be blamed for it. Study 2 was designed to build on many of Study 1 findings and overcome its limitations. I used an experimental design where participants delivered negative news to a recipient. I manipulated (rather than measured) empathic concern and moral identity. In this way I provided a strong test of the direction of causality. I extended the investigation to a different context (performance feedback), and coded filmed interactions as my measure of interactional justice. 57 CHAPTER 3: STUDY 2 One of my research goals was to consider what organizations can do to increase the likelihood that managers will utilize interactional justice when delivering bad news. Given findings from Study 1,1 tested whether an intervention (i.e., instructing communicators of bad news to be empathic) would be effective in this regard. I also tested whether one's moral identity could be activated as a way to increase interactional justice. In Study 2,1 build on Study 1 by investigating the effect on interactional justice of communicator empathic concern, moral development, and moral identity. There are several reasons for focusing on these variables. First, empathic concern and moral identity (versus, for example, strategic emotional intelligence) can be manipulated in a laboratory, making it possible to establish a causal link between our independent and dependent variables. Second, empathic concern and moral development were found in Study 1 to interactively predict interactional fairness. Researchers have proposed a close connection between higher order emotions, such as empathic concern, and moral behavior (e.g., Eisenberg, 2000; Eisenberg & Miller, 1987), which makes their interaction especially worth investigating. In Study 2,1 also explore the role played by moral identity in interactionally fair communication. Theory and Hypotheses for Study 2 In Study 2,1 investigated whether interactional fairness can be increased by (1) manipulating a communicator's empathic concern, and (2) priming a communicator's moral identity. I also explored whether moral identity interacts with an empathic concern induction and moral development to predict interactionally fair communication. In Study 58. 2, participants provided actual feedback to the news recipient (versus writing the message as in Study 1), thus providing a stronger test of my hypotheses. Empathic Induction and Interactionally Fair Communication Given the positive relationship between empathic concern and interactional justice, an important question becomes: What can be done to increase interactional fairness when delivering bad news? One possibility is to manipulate a communicator's empathic concern. Further, by manipulating empathic concern, versus measuring it, the causal role of empathic concern can be tested. Numerous studies have shown that empathic concern can be invoked by instructing people to step outside of their usual perspective on events and imagine how a distressed other feels (Toi & Batson, 1982). For example, Batson, Earley, and Salvarani (1997) found that asking subjects to "imagine how the person ... feels about what has happened and how it has affected his or her life" (p. 753) increased feelings of empathic concern prompted by witnessing the person's suffering. Most investigations have found that when people are asked to take the perspective of a "victim" they are more likely than people who were not given perspective-taking instructions to rate their emotional response as one of empathic concern (Davis, 1994). In summary, one way of increasing manager's interactional fairness is to encourage their empathic concern for the victim of bad news. Building on the theory presented in Study 1,1 made the following prediction. Hypothesis 1: Bad news is communicated with more interactional fairness when empathic concern is induced than when empathic concern is not induced. 59 Moral Identity and Interactionally Fair Communication Kohlberg's (1969) Cognitive-Developmental Theory assumes that higher levels of moral development relate to higher levels of moral behavior. It is increasingly, recognized, however, that cognitive moral development does not automatically lead to moral behavior. Hoffman (2000) and Eisenberg (1986) suggested that moral emotions, such as empathy or guilt, can play an important role in transforming abstract moral principles into other-oriented behavior. A number of approaches have also focused on the motivating role of an individual's moral identity (e.g., Blasi, 1980; Aquino & Reed, 2002), defined as the extent to which being moral is regarded as important to a person's sense of self (Blasi, 1983). Moral identity is generally seen as complementing the moral reasoning approach by identifying a social-psychological motivator of moral conduct (Aquino & Reed, 2002). Aquino and Reed (2002) developed and tested a measure of moral identity based on the self-importance to a person of a set of nine moral traits. Two dimensions of moral identity have been identified: internalization and symbolization (Aquino & Reed, 2002). Internalization reflects the extent to which the moral traits are central to a person's self concept. Symbolization reflects the importance to a person of publicly expressing the moral traits through their behavior. A person's moral identity is only one of their social identities, albeit one that can be expected to more strongly influence moral behavior than other social identities (Reed & Aquino, 2003). Aquino and Reed (2002) found that the self-importance of a person's moral identity related to moral intentions and behavior, such as donating to charities and self-reported volunteering. Research shows a link between a highly self-important moral 60 identity and a more expansive and inclusive circle of moral regard (i.e., the people who are entitled to moral treatment from an individual). For instance, moral identity has also been related to the extent to which people forgive, provide assistance to, and show concern for the welfare of outgroups (Reed & Aquino, 2003). The two dimensions of moral identity predict different attitudes and outcomes. Internalization appears to more directly tap into the self-importance of the moral characteristics whereas symbolization is concerned with the public display of a moral persona. Symbolization may be more predictive in situations that are highly public and in which the requirements for moral behavior are unambiguous than in less public situations and contexts in which there is less agreement regarding moral behavior. This can explain why symbolization (but not internalization) relates to religiosity, a highly established context (Aquino & Reed, 2002). In contrast, internalization relates more strongly to self-consistency and self-sanctions for violating one's own moral principles. Internalization should relate to moral behavior in situations where behavior is less public and the requirements of moral behavior are less prescribed. In such situations, the motivation for moral action would lie in the internal desire for self-consistency. Moral Identity Priming and Communicating Negative News Moral identity is only one of a person's multiple identities which can influence motivations and behavior. Blasi (1984) described moral identity as an individual difference which can change over time and between different contexts. Although a person's moral identity tends to be relatively stable over time, it can be activated or suppressed by situational variables (Aquino & Reed, 2002). Skitka's (2003) Accessible Identity Model proposes that moral identity will be activated (over material and social 61 identities) when a situation arouses moral thinking or emotions. When moral identity is activated, it is expected to dominate the self-concept over material or social concerns (Skitka, 2005). Reed, Aquino, and Levy (2006) found that a person's moral identity could be primed by, first, reflecting on the nine moral traits identified by Aquino & Reed (2002), and, second, writing about themselves in terms of the nine moral traits. Individuals whose moral identity was made more salient in this way were more likely (than individuals who wrote about themselves in terms of nine non-moral traits) to demonstrate caring for individuals in need, in terms of donating time to a charitable organization (Reed, Aquino, & Levy, 2006). When a person's moral identity is activated, they will be motivated to act in a way that is consistent with the traits associated with being a moral person. Several of the traits common to most people's moral self-definitions (Aquino & Reed, 2002) are likely to lead to interactionally fair communication, including: honesty, fairness, compassion, and caring. I expected that priming moral identity will increase communicator efforts, in terms of communication, to reduce the suffering of the negative news recipient through proving reasonable explanations and showing respect and concern. Hypothesis 2: Bad news is communicated with more interactional fairness when moral identity is primed than when moral identity is not primed. The Internalization Dimension of Moral Identity and Communicating Negative News I also expected that the internalization dimension of moral identity as a trait relates positively to the interactionally fair communication of negative news for several reasons. First, although the content and emphasis of moral identities will vary among 62 individuals, moral identity measures the self-importance to individuals of nine traits that are central to most people's moral self-definitions. In order to maintain consistency with their moral selves, individuals who are high in the internalization dimension of moral identity will seek to demonstrate these traits, which include honesty, fairness, and compassion, and to avoid communicating negative news in ways that would be seen as violating these traits. Second, in every situation, multiple identities and motives vie in compelling action (Eisenberg, 1986). However, a person's moral identity should be a greater predictor than other identities of treating recipients of negative news in a way that is consistent with the above moral traits. This should especially be the case for individuals high (versus low) in the internalization dimension of moral identity. Third, the internalization dimension of moral identity has been found to relate to actual moral behavior (versus self-reported behavior) (Aquino & Reed, 2002). Tyler, Boeckmann, Smith, and Huo (1997) suggested that people extend considerations of fairness only to individuals within their "moral community." Reed and Aquino (2003) found that the internalization dimension of moral identity related to a more expansive "circle of moral regard" (i.e., the individuals entitled to moral concern). Low moral identity individuals are more likely than high moral identity individuals to regard recipients of negative news as outside this circle, and therefore not entitled to moral decisions and treatment. As a result, low (versus high) moral identity individuals will use less moral behavior, in terms of interactional fairness, in communicating bad news. For the above reasons, I expected the internalization dimension of moral identity > to predict interactional fairness in the communication of negative news. 63 Hypothesis 3: The internalization dimension of moral identity is positively related to interactionally fair communication of bad news. The Interaction of Empathic Concern, Moral Development, and Moral Identity in Predicting Interactionally Fair Communication An important question arises: does empathic induction result in higher interactional justice for everyone? I argue that a three-way interaction among empathic induction (or control), moral development, and moral identity can predict interactionally fair communication of bad news. Specifically, the following higher order interaction is proposed: when both moral development and moral identity are high, the positive relationship between empathic induction and interactional fairness will be stronger than when either or both moral development or moral identity are low. The Four Component Model of Ethical Decision Making and Behavior Although higher level moral development has been linked to behavior that is consistent with postconventional moral reasoning (i.e., based on moral principles of fair treatment and reflecting a general concern for the wellbeing of others), the magnitude of the relationship has generally been modest (Blasi, 1984; Thoma, 1994). There are several reasons why moral reasoning ability might not automatically lead to interactional fairness. First, a decision to act consistently with a moral judgment requires that moral values be prioritized relative to other values, such as competing social or material demands that can outweigh a decision regarding what is morally right to do. Second, moral development does not always predict behavior because individuals do not always reason at their highest stage of moral reasoning (Higgins, Power & Kohlberg, 1984). Rather, individuals operate within a range of moral reasoning stages (Rest, 1979), and the 64 type of moral reasoning an individual engages in can also depend on characteristics of the moral issue (Jones, 1991; Reynolds, 2006), and contextual factors, such as whether an issue is defined in moral terms versus social or material terms (Skitka, 2003). Rest (1986) proposed a model of ethical decision making and behavior which regards moral actions as the result of four distinct processes. The first component, moral sensitivity, involves interpreting the issue as a moral one (i.e., a decision which can benefit or harm others) through processes of empathy and roletaking (Bergman, 2002). The second component, moral judgment refers to the type of reasoning (preconventional, conventional, or postconventional) that an individual engages in when they are deciding what is the right or wrong course of action in a justice dilemma. Moral motivation, the third component, involves prioritizing moral values relative to other values by deciding to act in accordance with a moral judgment. The final component, moral character, is having the character, persistence, and ability to act on a moral decision. The Four Component Model regards moral development as only one of several psychological processes necessary for moral action (Thoma, 1994). An individual's level of moral development is more likely to relate to interactional fairness when moral sensitivity, moral motivation, and moral character are also present (Rest, 1986). The communication of negative news will not always be regarded by the communicator as a moral issue, and therefore requiring moral reasoning, especially if other social or material identities are activated (Skitka, 2005). For example, in the context of a negotiation a person's material identity could be activated and the communication of negative news could be seen in terms of its influence on potential material gains or losses. Alternatively, if group membership is especially important to a person, interactions 65 within the group could activate his or her social identity, and communicating negative news could be seen less as a moral issue (i.e., what is the right thing to do here to be consistent with my principles?) and more as a social issue (i.e., what should I do to maintain my standing and relationships within the group?). In order for moral reasoning to lead to interactional fairness, the issue must be recognized as a moral one. An empathic induction is expected to increase a communicator's recognition of an issue as a moral one (i.e., one in which a person's actions can harm another person) by having the communicator imagine the full impact of negative news on the victim. An empathic induction thereby makes salient the moral aspects of a negative news situation, including the consequences for the news recipient. In this way, an empathic induction increases "the moral imperative in a situation," labeled moral intensity (Jones, 1991, p. 372). High moral intensity issues are more likely than low moral intensity issues to be recognized as moral issues, to stimulate more sophisticated moral reasoning, and to lead to actions based on moral judgments (Jones, 1991). Further, an empathic induction is likely to be a more potent influence on the interactional fairness of individuals that are high (versus low) in the internalization dimension of moral identity. Because these communicators attach greater importance (than low moral identity individuals) to their moral self-identity, their interactional fairness will be more affected by the increased moral intensity resulting from empathy induction. At least three reasons underlie why moral identity is expected to increase the positive effect of moral development on the relationship between empathic induction and interactional fairness. First, a person must be motivated to act consistently with a moral judgment. Specifically, a judgment regarding what is morally right must be followed by a 66 decision to act on the basis of that judgment. In order to maintain psychological self-consistency and not betray a central part of who they are, high moral identity individuals are more likely (than low moral identity individuals) to act in accordance with their moral decisions. Second, high moral identity individuals can be more motivated to expend cognitive resources in higher level moral reasoning when considering an ethical dilemma (Aquino & Reed, 2002), increasing the effect on behavior of a person's moral reasoning ability. Third, because the internalization dimension of moral identity involves a more expansive circle of moral regard, it is likely to decrease psychological distance between a communicator and the recipient of bad news. This can increase the moral intensity of a situation (Jones, 1991) and make it more likely that an individual, first, expends the cognitive resources to engage in higher level moral reasoning and, second, decides to act in accordance with a moral decision (Jones, 1991). In summary, empathic induction is expected to relate more strongly to interactional justice when it both activates principled reasoning (moral development must be high) and is accompanied by strong motivation to act consistently with moral decisions (moral identity must be high). When moral identity is low, moral development does not moderate the effect on interactional justice of an empathic induction, because the individual might not be motivated to act on a moral decision. When moral identity is high, however, high moral development will increase the positive effect on interactional justice of an empathic induction, as communicators are more likely to reason at their highest stage of moral development and to act consistently with their decisions made through moral reasoning. Thus, in the high moral identity condition I would expect the interactional justice of low (versus high) moral development communicators to be at a 67 lower level in the control condition, and to increase less as a result of the empathic induction. In summary, I propose the following three-way interaction. Hypothesis 4: Under conditions of high moral identity, moral development moderates the effect of empathy induction on interactionally fair communication of bad news. Under conditions of low moral identity, moral development does not moderate the effect of empathic induction on interactionally fair communication. 68 Summary of Study 2 Hypotheses 1. Bad news is communicated with more interactional fairness when empathic concern is induced than when empathic concern is not induced. 2. Bad news is communicated with more interactional fairness when moral identity is primed than when moral identity is not primed. 3. The internalization dimension of moral identity is positively related to interactionally fair communication of bad news. 4. Under conditions of high moral identity, moral development moderates the effect of empathy induction on interactionally fair communication of bad news. Under conditions of low moral identity, moral development does not moderate the effect of empathic induction on interactionally fair communication. 69 Study 2 Method This study was a 2x2 between-subject factorial design, with participants randomly assigned to empathic manipulation or control, and to high or low moral identity prime. The same experimenter ran all the sessions, and the same confederate received feedback from each participant. Participants Participants were 81 undergraduate students from a variety of faculties at the University of British Columbia, recruited via posters and end-of-class announcements at the Sauder School of Business. Participation was completely voluntary, and subjects were able to withdraw from the study at any time without penalty. Upon completion of the study, participants were paid $20.00. The average age of participants was 21.5, 74.5% were women, and 70.6% were commerce students. Procedure Study 2 involved two online questionnaires, and a laboratory experiment in which student participants provided face-to-face performance feedback to another student (actually a confederate). The feedback provided was videotaped and coded for interactional fairness by two independent coders. Students interested in participating were given a website address for two online questionnaires, and scheduled for a 1-hour lab appointment. The first online questionnaire contained demographic questions and measures of trait empathic concern and the internalization dimension of moral identity. The second online questionnaire was the Defining Issues Test of moral reasoning (D1T2; Rest et al., 1999) to measure level of 70 moral development. Both questionnaires were to be completed before participants arrived at the lab session. The following steps were followed when participants arrived at the lab. Step 1. The participants were welcomed to the lab, completed the consent form (See Appendix B), and were told that they were in an experiment about giving and receiving feedback. Step 2. Each participant was introduced to the other student (actually a confederate). The participant and the confederate ostensibly drew lots to see who would give feedback and who would receive feedback. The participant always ended up giving feedback to the confederate. The confederate was taken into an office to ostensibly complete a test on which she would receive feedback from the participant. The task consisted of completing a Pre-GMAT, a set of multiple choice questions testing verbal and analytic skills relevant to performance in a graduate business program. The task took between 10 and 15 minutes to complete. Step 3. The participant's moral identity was primed following procedures recommended by Reed and colleagues (2006). Participants were given a short writing exercise. The first part consisted of nine words, each listed on a separate row in the first column of a 9 x 5 matrix. Participants in the low moral prime condition were given the following words: carefree, compatible, favorable, generally, happy, harmless, open-minded, respectable, and polite. Participants in the high moral prime condition were given the following words: caring, compassionate, fair, friendly, generous, helpful, hardworking, honest, and kind. Each participant was asked to reflect for a few seconds on the meaning of each word and how it related to him or her, and then handwrite each word four times, across 71 the four empty cells in each row. On the page that followed the participant was asked to briefly reflect again on each word, to visualize each word as it was relevant to his or her life, and to spend 10 to 15 minutes writing a brief story (one or two paragraphs) about himself or herself which used each of the words at least once. Participants then completed a manipulation check. Step 4. The participant was then given details of the task - specifically that they would be marking a test in order that they could provide feedback to another person. Participants in the empathic induction condition were given specific perspective-taking instructions for giving feedback to the confederate. Specifically, participants were asked (via verbal instructions from the experimenter) to imagine how the person receiving the news/feedback would feel, an induction technique found to evoke empathic concern (Batson, Sager, Garst, Kang, Rubchinsky, & Dawson, 1997). Participants in the empathic concern condition were told "Put yourself in the other person's shoes ... try to imagine how they think and feel about the feedback. Try to feel the full impact of what the other person is going through." Participants in the control group were not given any directions regarding the perspective they should take in planning and delivering feedback. Step 5. The experimenter then brought the confederate's completed test to the participant and asked the participant to grade the task using the marking instructions provided to participants. In addition, the marking instructions explained that the Pre-GMAT was a good predictor of performance on the actual Graduate Management Admissions Test, which is necessary for admission to most M B A and other graduate business programs. In all cases the confederate did poorly on the test, with their performance placing them in the fourth (bottom) quartile. 72 In order to make the feedback that the participant provided to the confederate seem meaningful (to the confederate), the test included a Demographic and Motivation Questions section, which asked the ostensible test-taker (1) "Are you interested in pursuing an M B A degree?", and (2) "How interested are you in obtaining your test score on the Pre-GMAT?". The test always indicated that the confederate was very interested in pursuing an M B A degree (with a score of 6, where 1 = Definitely Not and 7 = Definitely), and that the confederate was very interested in obtaining her test score on the Pre-GMAT (with a score of 6, where 1 = Not Interested and 7 = Extremely Interested). Step 6. When the participant had graded the confederate's test and indicated that he or she was ready to deliver the performance feedback, the confederate was led by the experimenter into the room where the participant waited. The experimenter then left the room and closed the door. The participant delivered the performance feedback, which was videotaped. The confederate had been trained to respond consistently to the feedback, e.g., by saying they felt they had done well, if asked, by not showing strong positive or negative emotions, by not asking questions to the participant, and by showing the same physical demeanor in terms of sitting upright, attentive, and slightly away from the table. The experimenter was informed by the confederate when the feedback was finished, and the confederate was taken into another room to ostensibly complete a follow-up questionnaire. Step 7. The experimenter administered a follow-up questionnaire to the participant consisting of a manipulation check, which asked participants the instructions that they received for giving feedback. 73 Step 8. The experimenter debriefed the participant, questioned him or her regarding the deception, and asked the participant not to discuss the study with other students. Measures Unless otherwise indicated, response sets consisted of 7-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree). For each variable, the individual item scores were summed to form the variable score. Larger values signify greater amounts of the variables than lower numbers. Manipulation checks In order to check on the empathic concern manipulation, after completing all other measures and providing the feedback, participants were asked to indicate on the follow-up questionnaire which instructions they recalled receiving for the performance feedback. Participants indicated whether they recalled being verbally given that particular instruction, on a 7-point Likert scale, ranging from 1 (Definitely Not) to 7 (Definitely). Four statements related to the empathic concern induction: "To imagine how the other person receiving the news would feel," "To try to feel the full impact of what the other person is going through," "To try to think of what concern and sympathy I felt for the other person," and "To take the other person's perspective and think how the news would affect his or her life." In order to check on the moral identity prime condition, participants were asked to reread the story they had written about themselves and to indicate to what extent it reflected how they saw themselves as a) a student, b) a member of an organization, c) a moral person, and d) safety-conscious. 74 Predictor variables Moral development. Moral development was measured using the Defining Issues Test, Version 2 (DIT2) (Rest et al, 1999). See Study 1 for a description. Moral Identity. Participants indicated how important it was to them to have 9 characteristics, which have been found to be associated with moral identity (Aquino & Reed, 2002): caring, friendly, helpful, compassionate, generous, honest, fair, hardworking, and kind. Moral identity was measured using the 5-item internalization subscale of Aquino and Reed's (2002) moral identity instrument. The items were as follows: "It would make me feel good to be a person who has these characteristics," "Being someone who has these characteristics is an important part of who I am," "I would be ashamed to be a person who has these characteristics" (reverse coded), "Having these characteristics is not really important to me" (reverse coded), and "I strongly desire to have these characteristics." Criterion variables Interactional justice. Interactional justice was measured in two ways. First, the confederate rated the interactional justice of each participant. Second, the interactional justice of the participants' feedback to the confederate was content coded independently by two graduate students who were blind to conditions and to how participants responded to other survey measures. For each videotaped message, they rated whether the communicator 1) was polite and courteous, 2) treated the recipient with dignity and respect, 3) refrained from improper remarks and comments, 4) gave a reasonable explanation, 5) thoroughly explained the procedures followed, and 6) tailored his or her communication to the recipient's needs. 75 Two coders were trained in using the guidelines for coding interactional justice. The coders independently rated messages from a small pilot sample (N = 10). Their ratings on the pilot messages were compared. Areas of disagreements were discussed with the researcher and used to help clarify and elaborate the coding guidelines. Each of the 81 messages from the participants in this study was then rated independently by the two raters for each of the 6 interactional justice items on a 5-point Likert scale where 1 = Strongly disagree and 5 = Strongly agree. The level of agreement for the 6-item interactional justice score, as measured by an intraclass correlation coefficient (McGraw & Wong, 1996; Shrout & Fleiss, 1979), was .95 (ICC). The scores for the two raters were averaged for each of the six items, and then the item scores were summed to form the interactional justice measure. Control variables Trait Empathic Concern. Participants indicated the extent to which they characteristically engage in the empathic concern dimension of empathy. Empathic concern was measured using the 7-item Empathic Concern subscale of the Interpersonal Reactivity Index (Davis, 1983). The items were as follows: "I often have tender, concerned feelings for people less fortunate than me," "When I see someone being treated unfairly, I sometimes don't feel very much pity for them, (reverse coded)" and "Sometimes I don't feel very sorry for other people when they are having problems" (reverse coded), "When I see someone being taken advantage of, I feel kind of protective toward them," "I would describe myself as a pretty soft-hearted person," "Other people's misfortunes do not usually disturb me a great deal" (reverse coded), and "I am often quite touched by things that I see happen." 76 Demographic Variables. Gender and age were controlled because of their possible influence on variables of interest to this study. For example, females tend to report greater levels of empathy, especially the empathic concern dimension, than males do (Lennon & Eisenberg, 1987). Males and females have also been found to differ in the obligation they feel to communicate negative news (Tesser et al., 1972), and in their responses to anticipated interpersonal rejection (Kelly, 2001). Females also score slightly higher on the DIT2 (Rest, 1994). Because greater age is associated with higher moral development scores (Rest, 1994), age was controlled. Source. In order to control for the possibility that students from different sources responded differently to the variables of interest, I controlled for student source. I dummy-coded (1 = yes, 0 = no) each of the four classes recruited from and the group recruited via posters. Three of the four classes were commerce majors only, and one of these was a higher level human resource management class that included performance feedback as a topic. The majority of the students responding to the poster were from faculties other than commerce. Study 2 Results The manipulation check for empathic concern manipulation showed that participants in the empathic induction condition were more likely (than participants in the control condition) to indicate the empathic concern induction instructions as ones they had been given, M = 5.99, SD = 1.15, F = 108.76, p < .001, versus in the control condition, M = 2.35, SD = 1.92. The results showed that the manipulation was effective and in the right direction. 77 The manipulation check for moral identity prime showed that participants in the high moral identity prime condition, however, were not more likely than participants in the low moral identity prime condition to indicate that the story they wrote about themselves reflected how they saw themselves as a moral person, M = 5.40, SD = 1.34, F = .213, ns, versus in the control condition, M = 5.54, SD = 1.32. Thus, the results suggest that the moral identity prime might not have been effective. The descriptive statistics and correlations between the measures are reported in Table 12. Scale reliabilities in terms of Cronbach's alpha are given on the diagonal. Interactional justice scores as rated by the two coders from videotapes were on average 21.46 and ranged from 14.00 to 30.00. The confederate ratings of interactional justice were on average 22.39, ranged from 12.00 to 29.00.1 refer below to the former, coder-ratings of videotaped interactions, as interactional justice, and the latter, confederate ratings of the interactional justice they experienced, as confederate interactional justice. Interactional justice and confederate interactional justice were significantly correlated, r = .42, p < .01. For the reasons given below in the Study 2 Discussion, I use interactional justice (independent ratings of videotapes by two coders) as my criterion variable, but for completeness I report results for both interactional justice and confederate interactional justice. An analysis of covariance (ANCOVA) was conducted, with participant class, gender, age, and trait empathic concern as covariates, to determine whether significant differences in interactional justice existed between the control and empathic induction conditions. As Hypothesis 1 predicted, empathic induction was positively related to interactional justice, F=11.29, p<.01. As shown in Table 13, interactional justice was 78 lower in the control condition (M = 20.15, SD = 2.54) than in the empathic induction condition (M = 22.18, SD = 2.80). Empathic induction was also positively related to confederate interactional justice, F = 11.46, p < .001. As shown in Table 14, confederate interactional justice was lower in the control condition (M = 20.96, SD = 3.01) than in the empathic induction condition (M - 23.71, SD = 2.66). Hypothesis 2 predicted that a moral identity prime increases interactional justice. As shown in Table 13, interactional justice was not significantly different in the low moral identity prime condition (M = 21.31, SD = 3.16) than in the high moral identity prime condition (M = 21.09, SD = 2.52), F = .05, ns. As shown in Table 14, confederate interactional justice was also not significantly different in the low moral identity prime condition (M =22.44, SD = 3.12) than in the high moral identity prime condition (M = 22.34, SD = 3.19), F = .01, ns. Thus, Hypothesis 2 was not supported. Hypothesis 3 predicted that the internalization dimension of moral identity is positively related to interactional justice. Table 15 shows moral identity internalization was not related to interactional justice, /? = .04, ns. Table 16 shows moral identity internalization was also not related to confederate interactional justice, /? = -.06, ns. Thus, Hypothesis 3 was not supported. Hypothesis 4, which predicted a three-way interaction between empathic induction, trait empathic concern, and moral development, was tested using hierarchical regression analysis. The control variables were entered in at Step 1, main effects were entered in at Step 2, second-order interactions were entered in at Step 3, and the third-order interaction was entered in at Step 4. Table 17 shows a significant three-way interaction among empathic induction, moral development, and moral identity 79 internalization, /? = .26, p < .05. The interaction was probed using the procedures recommended by Aiken and West (1991). As shown in Figure 3 and Figure 4 and Table 19, only in the case of high moral identity internalization communicators and when moral development was high was there a positive relationship between empathic concern induction and interactional justice, t(Sl) = 3.16, p< .01. No relationship between empathic concern induction and interactional justice existed when moral identity was low and moral development was either low, ?(81) = 1.30, ns, or high, ?(81) = 1.46, ns. No relationship existed between empathic concern induction and when moral identity was high and moral development was low, f(81) = -.17, ns. However, the pattern of results in the interaction was not as predicted. Specifically, in the control condition when moral identity was high, low (versus high) moral development related to interactional justice. Thus, Hypothesis 4 was not supported. As shown in Table 18, the interaction among empathic induction, moral development, and moral identity internalization did not predict confederate interactional justice, ft = .04, ns. Study 2 Discussion In Study 2,1 was interested in exploring whether interactional justice could be increased via increasing empathic concern and moral identity. Specifically, I investigated the effect of empathic concern, moral development, and moral identity on the interactional fairness with which negative news is communicated. I build on Study 1, in which an interaction between moral development and trait empathic concern was found to predict organizational justice. Empathic concern was manipulated in Study 2, via an empathic induction, in order to test for a causal relationship with interactional justice. 80 Moral identity was both measured and primed in Study 2 because it was expected to interact with empathic concern and moral development in predicting interactional justice. I sought to extend the findings from Study 1 and to provide a more stringent test of my predictors than in Study 1. Specifically, Study 2 involved a context in which the news was delivered face-to-face, believed to be real, and contained performance feedback. Study 2 showed the positive effect of an empathic induction on interactional justice. Communicators who were asked to imagine how the recipient would be affected by the negative news (versus communicators who received no perspective-taking instructions) provided more detailed explanations and treated the recipient with more dignity and respect. The relationship in Study 1 between empathic concern and interactional justice held when empathic concern was manipulated and the news was communicated face-to-face. Thus, empathic concern played an important role in a face-to-face situation, where a verbal message could be accompanied by gestures, facial expression, vocal tone, and body language that demonstrated sympathy and compassion. It is important, however, to consider whether the effect of the empathic induction on interactional justice might have been a product of experimental demand. I doubt this was the case as participants would not have known how the instructions in their condition (i.e., empathic concern versus control group) differed from those in the other condition. Nor would participants have been aware of the criteria by which their messages would be coded, making it unlikely that a demand effect or social desirability bias operated on their responses. It is possible, however, that a less blatant induction would not have had as great an effect on interactional justice. Future research should examine the effect on interactional justice of more subtle ways of inducing empathic concern, such as via 81 instructions that do not specifically refer to communicator feelings of warmth and concern for the recipient, or via questions posed to the communicator. I also investigated the effect of a moral identity prime, in order to explore the effect of moral identity on interactionally fair communication of negative news. No significant relationship between the moral identity prime and interactional justice was observed. In Study 2, the manipulation check for moral identity priming showed that the moral identity prime appeared not to be effective in making people in the high moral identity prime condition think about themselves as moral people. One reason could be that the prime was not sufficiently strong to influence behavior in a face-to-face situation. In a face-to-face situation, a communicator's social identity could be activated by sitting across from the recipient of negative news or at the prospect of doing so. This could have reduced the effect of a moral prime earlier administered. Second, participants' close attention to the face-to-face instructions used in the empathic induction could have overpowered the effect on interactional justice of the moral identity prime, which was a written prime. This suggests that a person's moral identity may be difficult to make salient in the presence of strong emotional cues and instructions. Third, participants were aware in advance that they would be giving or receiving meaningful feedback. This could already have prepared participants to think of the situation as one in which their actions could harm or benefit another individual (i.e., a moral one), thereby reducing the effect of a moral identity prime. The hypothesized main effect of the internalization dimension of moral identity and interactional justice was also not observed. At least three reasons might explain this result. First, the internalization dimension of moral identity motivates a person to be true 82 to his or her self-conception as a moral person. In a situation where negative feedback must be delivered, however, people can reasonably vary in what they regard as moral behavior. Even if an individual is motivated to think of themselves as a moral person, he or she may reason at the preconventional or conventional level to determine a right course of action. For example, an individual who is low in moral development might have, regardless of the self-importance of his or her moral identity, believed that the right thing to do in that situation is to follow social norms or minimize social losses. Reynolds and Ceranic (2006) found that high moral identity can in some circumstances increase commitment to behavior that would normally be regarded as unethical (in that case, cheating by students) if the behavior is common and does not violate social or personal norms. When it comes to communicating negative news, people at different stages in moral reasoning can reasonably differ with regard to what they see as moral behavior. This might explain why a main effect of moral identity on interactional justice was not observed. A second possibility for why moral identity did not relate to interactional justice could be that communicating negative news might have been especially threatening to high moral identity individuals. Communicating outcomes that are likely to be regarded as unfair creates a "justice dilemma" (Bies, 1987), whereby communicators fear negative reactions from the news recipient, relational costs, and reputational harm. Manis and colleagues (1974) showed that communicators of negative news often feel blamed even when they are not responsible. In the case of high moral identity communicators, feeling blamed for negative news could threaten a person's highly self-important impression of himself or herself as fair, compassionate, or caring. High moral identity communicators 83 might have been able to maintain a preferred self-impression by framing a situation that is not explicitly moral, such as communicating negative performance feedback, as a nonmoral one which has lower moral intensity. This could have reduced the effect of moral identity on interactional fairness. Although a three-way interaction between empathic induction, moral development, and moral identity predicted interactional justice, it differed from the hypothesized interaction. As hypothesized, the effect on interactional justice of the empathic induction was stronger when both moral development and the internalization dimension of moral identity were high. When moral identity was low, moral development did not moderate the relationship between empathic induction and interactional justice. However, contrary to what I hypothesized, in the control condition when moral identity (internalization) was high, moral development related negatively to interactional justice. This raises an interesting question: why, when empathic concern was not induced, might individuals who were high in moral development and moral identity have used less interactional fairness in communicating negative news? There are several possible explanations for this finding. First, it is possible that high moral development communicators were less concerned with relational goals than were communicators who reasoned at the conventional level of moral development. High moral development communicators can be expected to be more concerned about following universal rules of what one ought to do, and to be less concerned about following social rules, maintaining a positive relationship with the recipient, and allaying the recipient's negative reactions. This is especially likely to be the case when high moral identity reinforces a communicator's dominant style of moral reasoning. In such a case, 84 an empathic concern induction would have been required to shift high moral development communicators away from principled reasoning and toward becoming concerned with the feelings and reactions of the news recipient. When moral identity was low, although higher moral reasoning may have resulted in principled reasoning, communicators may have lacked the motivation to give priority to universal principles of moral behavior over other (e.g., social or relationship ) considerations. This is consistent with Study 2 findings. The above reasoning suggests, however, that the universal principles applied by individuals high in moral reasoning and high in moral identity do not always relate positively to higher interactional justice, and may, in fact, relate negatively to interactional justice. For example, it is possible that in Study 2 (where the communicator had to communicate that the recipient achieved a very low score on a pre-GMAT test, and that the low score indicated that the recipient was unlikely to be admitted to an M B A program), high moral development communicators believed that the most ethical course of action was to give direct feedback which did not "sugarcoat" the negative news or raise false hopes in the recipient. Such an approach, though moral, could be regarded as less interactionally fair in terms of the respect and politeness shown, and of the adequacy and personalization of explanations. I further explored the three-way interaction by investigating whether a similar three-way interaction predicts individual items within the interactional justice measure. I reasoned that it was possible that communicators who are high in both moral identity and moral development might be particularly low in specific aspects of interactional justice. As shown in Tables 20 and 22, the three-way interaction was significant for two of the 85 six interactional justice items, both relating to interpersonal treatment: interactional justice (politeness), and interactional justice (dignity and respect). The pattern of results for the two individual items, which were probed using the procedures recommended by Aiken and West (1991), were similar to the interaction observed with interactional justice as the dependent measure. When moral identity was high, in the control condition communicators who were high (versus low) in moral development showed less interactional justice (politeness) and less interactional justice (dignity and respect). When moral identity was high, an empathic induction raised the level of both interactional justice items (politeness, and dignity and respect) for high moral development communicators to the same level as that of low moral development communicators, whose score on both items remained stable across control and empathic induction conditions. Figure 5 and Figure 6 and Table 21 show that only in the case of high moral identity internalization communicators and when moral development was high was there a positive relationship between empathic concern induction and interactional justice (politeness), £(81) = 2.59, p < .05. No relationship between empathic concern and politeness existed when moral identity was low and moral development was either low, £(81) = 1.52, ns, or high, £(81) = .20, ns. No relationship existed between empathic concern and interactional justice (politeness) when moral identity was high and moral development was low, £(81) = -.34, ns. Figure 7 and Figure 8 and Table 23 show a similar pattern of results when interactional justice (dignity and respect) is regressed on the three-way interaction. Only in the case of high moral identity internalization communicators and when moral 86 development was high was there a positive relationship between empathic concern induction and interactional justice (dignity and respect), r(81) = 2.86,/? < .05. No relationship between empathic concern and politeness existed when moral identity was low and moral development was either low, £(81) = 1.10, ns, or high, £(81) = .25, ns. No relationship existed between empathic concern interactional justice (dignity and respect) when moral identity was high and moral development was low, £(81) = -.14, ns. Although caution should be exercised in interpreting the results for individual items, the above analysis suggest that when empathic concern was not induced communicators who were high in moral development and in moral identity appear to be less interactionally fair in specific respects: being polite, and showing dignity and respect for the news recipient. This interaction was not observed for the other aspects of interactional justice measured: refraining from improper remarks, giving reasonable explanations, explaining procedures thoroughly, and tailoring communication to the recipient. This could suggest that in the absence of an empathic induction high moral development communicators with a self-important moral identity might be less focused on the person (versus task-) aspects of communicating negative news. Whereas refraining from improper remarks and crafting a detailed, reasonable, and personalized message might be necessary for clear communication of negative news, being polite and demonstrating dignity and respect relate more to maintaining a good relationship with the news recipient, and affirming the worth of the news recipient as an individual. It is possible that individuals operating at a conventional level score higher in these "relationship" aspects of interactional justice. Moral reasoning based on universal principles (versus moral reasoning based on social rules and demands) might favor clear 87 cornmunication over communication that is focused on the social relationship and the feelings of the negative news recipient. Future research should examine whether high moral development, when it is accompanied by high moral identity, leads to moral behavior that can fall outside the measurement of interactional justice, such as perhaps avoiding sugarcoating news, providing realistic advice, and treating all news recipients in a less personalized and more uniform manner. A great deal of research has explored the effects of moral judgment and moral identity separately (Reynolds & Ceranic, 2006). This is one of few studies that has looked at their interactive effects. An important role of empathy could be that it transforms abstract moral reasoning into interactional fairness. According to the Four Component Model (Rest, 1986), in order for high moral development to lead to moral behavior, a person has to recognize the issue as a moral one (moral sensitivity). When empathic concern was induced, the communicator was more likely to appreciate the possible harm to the recipient of the news, thereby increasing his or her interpretation of the situation as a moral one. Thus, empathic induction appears to make more salient aspects of a situation which are likely to activate moral reasoning and moral identity. This is consistent with Hoffman's (1970) suggestion that moral emotion (i.e., empathy) can transform abstract moral principles into prosocial behavior. An empathic induction might be especially important for drawing people's attention to the moral aspects of an issue in situations, such as performance feedback, which are not necessarily seen as involving moral decisions. Study 2 results also suggest that an empathic induction can be especially important for communicators who are high in moral development and moral identity 88 internalization in terms of the more interpersonal aspects of interactional justice: politeness, and demonstrating dignity and respect. Moral identity as a trait appeared to play a complementary role to moral development, as the Four Component Model would predict. Whereas empathic induction can lead to greater recognition of an issue as moral in nature (i.e., as one that can harm or benefit another person), moral identity can provide the motivation to act in accordance with moral decisions. In addition, when a moral issue is recognized, higher moral identity is related to a more expansive circle of moral regard, whereby the communicator is more likely to see the news recipient as deserving of moral reasoning and treatment. However, the three-way interaction observed in Study 2 suggests that high moral identity internalization can also lead high moral development communicators to use less interactional fairness, perhaps because a more self-important moral identity increases the use of strategies regarded as more moral, in terms of following universal principles of behavior, but likely to be perceived as less interactionally fair. Research has shown that high moral identity can increase the effect on behavior, both positive and negative (in terms of being moral), of the type of ethical reasoning a person characteristically uses (Reynolds & Ceramic, 2006). Thus, it is not surprising that moral development had a less significant effect on interactional justice in the low versus high moral identity condition. Specifically, in the low moral identity condition, moral development did not relate to interactional justice in the control condition or in the empathic induction condition. However, in the high moral identity condition, moral development had a significant effect on interactional justice in the control condition, but not when empathic concern was induced. Thus, it was when the internalization dimension 89 of moral identity was high and empathic concern was not induced that moral development appeared to have the greatest effect on interactional justice. A strength of Study 2 was the quality of the dependent variable. Interactional justice was rated by the confederate immediately after receiving feedback, and then the videotaped feedback was independently coded for interactional justice by two coders. The two measures were correlated, showing that the coding of the videotapes provided a reasonable measure of the interactional justice experienced by the news recipient. I used the coding of the videotapes as my interactional justice measure, as the interrater reliability could be assured, and the quality of the confederate ratings might have been adversely affected (for example, by contrast effects) by receiving the performance feedback from over 80 individuals. Bies (1986) defined interactional justice as the quality of interpersonal treatment in workplace interactions. This should include more than simply the words used to convey a message. The term interactional justice suggests that two parties interact during the communication and enactment of procedures. Interactional justice, for example, includes listening to a person's concerns, and refraining from improper questions or comments. By videotaping and coding face-to-face interactions, I captured important and rich aspects of interactional fairness that would not be captured through written messages or through a communicator selecting among strategies. The tone, body language, and facial expressions with which negative news is delivered can influence perceptions of fair interpersonal treatment and of message reasonableness and sincerity. Further, in a face-to-face situation (versus in a hypothetical and/or written situation) our participants would have more fully experienced the "justice dilemma" (Bies, 1987) of having to 90 communicating an outcome that can be regarded as unfair. In a face-to-face situation, communicators have to anticipate and deal with a possible negative reaction from the recipient of the bad news, who can hold them accountable for it, even if they are not responsible (Manis at al., 1974). There was considerable variation in what communicators said, in spite of the nature of the negative news: performance feedback on a multiple choice ability test. For example, some communicators that were low in interactional justice gave incorrect information to the recipient, omitted important details, questioned the new recipient's effort on the task, smirked or laughed during the feedback, and did not address the recipient by name, rise out of their seat to greet the recipient, or give the recipient the opportunity to respond or ask questions. On average, participants spent between two and three minutes communicating the negative feedback, with some spending less than a minute and others spending more than six minutes in giving the negative news to the recipient. Thus, given that all participants were asked to deliver the same feedback to the same individual, it appears that individual differences played a significant role in how the negative news was communicated. 91 CHAPTER 4: OVERALL DISCUSSION This dissertation seeks to increase understanding of an important aspect of managerial behavior: the interactional fairness with which bad news is communicated to employees. In Study 1 I found that interactional justice was predicted by a manager's trait empathic concern, moral development, and strategic emotional intelligence when interactional justice was regressed on these variables individually. Self-esteem and trait personal distress did not relate to interactional fairness. Further, moral development moderated the effect of empathic concern on interactional justice when all the predictors were included in the regression equation. In Study 2,1 found that a communicator's interactional fairness could be increased by an empathic induction. An interesting interaction among empathic induction, moral development, and moral identity was observed in Study 2. For high moral identity communicators, the effect of an empathic induction on interactional justice was greater for high (versus low) moral development communicators, who used less interactional justice than low moral development communicators when empathic concern is not induced. This is an important area of study that merits research as most managers will need at some point to impart unfavorable news to employees. The ability to do so in an interactionally fair way, which engenders employee acceptance and perceptions of fairness, may well be a key managerial skill. Bad news can concern a range of topics, such as layoffs or plant shutdowns, negative performance appraisals, reductions in resources, or company mergers or takeovers. In all cases, the way in which unfavorable outcomes are communicated will significantly influence how employees respond to them, in terms of decision acceptance, perceptions of fairness, and retaliation against the 92 organization. Interactional justice is especially relevant because managers often do not have control over the outcomes or the decision-making procedures leading to them, but they do have relatively greater control over their behavior. This can explain why, as Folger (2001) suggests, managers are most likely to be held morally accountable for interactional (versus distributive or procedural) fairness. For the above reasons, it is important to closely examine the individual variables that influence interactionally fair communication. Theoretical Implications This dissertation focuses on justice as a dependent variable, and from the perspective of the party behaving in a fair or unfair manner. Specifically, I examined individual differences that appear to encourage or impede fair treatment by communicators of bad news, in terms of interactionally fair communication. In contrast, most organizational justice research has tended to look at fair treatment as an independent variable influencing the perceptions and responses of the victim. As a result of previous studies we have learned a great deal about how the victims of adverse events respond to fairness in terms of outcomes, procedures, and treatment. However, we know far less about the factors that motivate and encourage fair behavior on the part of the manager. Specifically, why, in spite of the significant benefits of interactional fairness, might managers convey bad news in ways that violate expectations of fair interpersonal treatment? The findings of this research make several contributions to theory. First, in this paper I show the importance of communicator empathic concern to interactional justice. In Study 1, trait empathic concern related positively to interactional justice. In Study 2, 93 inducing empathic concern increased interactional justice. However, the effect on interactional justice of empathic concern depended on an individual's level of moral development. It appears from Study 1 that trait empathic concern is less important for individuals that are high in moral development, and more important for individuals that are low in moral development, suggesting a substitution effect. However, in Study 2, empathic induction increased interactional justice only for communicators high in moral development (who in the control condition were less interactionally fair than low moral development communicators), and only when moral identity was also high. These findings suggest that when a communicator's moral identity is high (versus low), high moral reasoning increases the positive effect on interactional justice of empathic concern. Thus, when moral identity is high, high moral development can increase the effect of, rather than substitute for, empathic induction. As discussed below, the interactive effects of empathic concern and moral development on interactional justice might also differ according to aspects of the communication (i.e., whether communication is face-to-face or written) and the moral intensity of the negative news (i.e., whether the news clearly involves harm to another person). Given the importance accorded to morality-based justice in philosophical writing (e.g., Feinberg, 1974) and psychological research (e.g., Folger, 2001; Skitka, 2005), the factors influencing non-self-interested motivations for fairness is an area that merits further attention. Relationships between moral development and interactional fairness should not be surprising given the strong justice-focus of Kohlberg's (1969) developmental model of moral reasoning, in which each stage indicated a different type of reasoning used to make decisions injustice dilemmas. In Study 1, a manager's moral 94 development predicted interactional fairness in communicating bad news. In Study 2, a communicator's moral development moderated the effect of empathic induction on interactional fairness so long as communicator moral development and moral identity were both high. It appears that without high moral identity (which can provide moral motivation), empathic induction (which can increase recognition of an issue as moral) can be insufficient to transform cognitive moral reasoning into a decision to implement interactional fairness when delivering bad news. This dissertation highlights the importance of research that explores the interactive effects of moral reasoning, moral identity, and moral emotions, such as empathic concern. We see in both Study 1 and Study 2 that moral development moderates the relationship between empathic concern and interactional justice, but in very different ways. In Study 1, low (versus high) moral development related to lower interactional justice unless trait empathic concern was high. In Study 2, when moral identity was high, high (versus low) moral development related to lower interactional unless empathic concern was induced. There are at least two possible explanations for these different findings. First, perhaps in situations where negative news is communicated in writing (versus face-to-face) moral development may be sufficient to predict interactional justice. However, in situations where negative news is communicated face-to-face, moral development can relate negatively to interactional justice when the use of principled reasoning is reinforced by high moral identity. It appears that high moral development can relate negatively to interpersonal aspects of interactional justice, such as being polite and showing dignity and respect, that can become more important when negative news is delivered face-to-face. In such situations (e.g., Study 2), empathic concern for the 95 recipient might be necessary to focus the attention of high moral development communicators on the feelings of the individual recipient (versus on universal principles of behavior). Second, the relationship between moral development, empathic concern, and interactional justice might also depend on the moral intensity of the news. When an issue is clearly moral, in terms of involving possible harm to another person (as in Study 1, which involved a layoff), a communicator may not require empathic concern to view the situation as a moral one and imagine the possible consequences of their action. However, when (as in Study 2, which involved performance feedback) a situation involves less obvious harm to another person, empathic concern may be necessary for a high moral development communicator to focus on the feelings of the news recipient versus on the principled Tightness of their own behavior. The latter is more likely to be the case when a high moral development communicator's moral reasoning approach is reinforced by a self-important moral identity. Thus, it appears that the relationship between moral development and interactional fairness is less straightforward than initially hypothesized. The three-way interaction suggests that moral development only matters in terms of interactional justice (in face-to-face situations) when it is reinforced by a communicator's high moral identity. Finally, not only was moral development not positively associated with interactional justice in the high moral identity - control condition of study 2, it was associated with less interactional fairness. This interesting finding suggests that moral development might be leading to communication strategies that are considered moral by communicators but interactionally unfair by news recipients. Future research should investigate whether high moral development might lead to strategies that can be regarded as moral by a 96 communicator in terms of being in the perceived best interests of the news recipient (e.g., frankness) but can regarded as interactionally unfair from the perspective of the news recipient. Further exploration in this area might reveal situations in which interactional justice can lead to less positive recipient outcomes than communication that is moral but interactional unjust. This dissertation raises an interesting question regarding an important aspect of interactional justice: its noncomparative nature. Feinberg (1974) suggested that what is regarded as interactionally fair depends on universal standards, rather on comparisons with how others are treated in similar circumstances. However, even if interactional justice is determined according to universal standards of fair treatment, it appears that whether interactional fairness is used by the communicator of negative news can depend on whether empathic concern is felt for the recipient. Implications for Practice From a practical perspective, given that most managers will need at some point to impart unfavorable news to employees, this area of study can potentially provide benefits for practice and training. The ability to deliver negative news in an interactionally fair. way, which engenders employee acceptance and perceptions of fairness, is a critical managerial skill (Whetton & Cameron, 2004). Despite its importance, however, managers might not always "do the right thing." I provide insights for managers and organizations that face the need to convey bad news to employees. For managers and organizations that face the need to convey bad news to employees, empathic concern for recipients of bad news appears to be crucial in predicting the interactional fairness of the communication strategies they employ. Thus, one strategy to increase interactional justice is to select as communicators of bad news individuals that are high in trait empathic concern. However, this may not always be possible because the task of communicating the bad news might fall to somebody, such as the manager of a department, who is selected for the role on the basis of a variety of characteristics. An alternative is to encourage managers to take the perspective of employees who are about to be informed of a bad outcome, and to imagine how the bad news can affect the employees' lives. Particular efforts should be made to focus the attention of managers who are high in moral development and moral identity on the need to consider the feelings of the individual news recipient in terms of using politeness toward them and affirming their dignity and respect. The best strategy for increasing interactional fairness may be a combination of selecting managers who are high in trait empathic concern and telling them to consider the plight of the victim. Strengths and Limitations One of the strengths of this dissertation lies in the ways in which interactional justice was measured. Whereas in Study 1 the interactional justice of written messages was coded, in Study 2 the interactional justice of face-to-face treatment was coded. Bies and Moag (1986) defined interactional justice as "the quality of interpersonal treatment they receive during the enactment of organizational procedures" (p. 44). More recently, Bies (2001) called for broader conceptions of interactional justice, to include,more than social accounts and to reflect the richness of the concept. Accordingly, in Study 2 I videotaped the way in which bad news was conveyed face-to-face. Further, because my interest lies in predicting interactionally fair behavior (versus how people think or say, 98 they would behave) I designed an experiment in which participants believed the news to be real and consequential to the news recipient. An experimental approach allowed me in both studies to control for extraneous variables and to ensure that communicators faced the same task. Although it is a challenge to create vivid and realistic scenarios wherein the communication of negative news elicits the moral reactions and emotional reactions that a real situation would, care was taken to make both situations realistic, feasible, and ones in which the communicator would act as themselves. Bies and Moag (1986) argued that one reason communication might have been neglected injustice research was the difficulty of separating the communication from distributive and procedural aspects. In this dissertation, by controlling the outcomes communicated and the procedures leading to the outcomes I specifically focused on the individual differences influencing the way in which news was conveyed. Though this dissertation provides valuable insights into why managers may neglect interactional fairness in communicating bad news, a few methodological weaknesses are noted. As noted earlier, Study 1 has potential limitations. First, the data were cross-sectional, making causality difficult to determine. Second, written messages might not capture all aspects of interactional justice. For instance, the concern, respect, and sincerity of explanations which are critical to interactional fairness can be more fully and richly conveyed face-to-face than through a written scenario. However, there will be situations, especially given the increased use of electronic mail for both routine and nonroutine messages (Crowther, 2001), when negative news is communicated in writing. Third, a layoff situation can differ from other organizational contexts because the 99 negative news is severe and relationship ending. Fourth, my measure of empathic concern in Study 1 exhibited low scale reliability. Although Study 2 was designed to address many of the issues raised in Study 1, it also has potential limitations. First, Study 2 used a relatively homogeneous sample of undergraduate students. This could have decreased within cell-variance and thus contributed positively to overall F-statistic size. In addition, the way in which students responded to a communication task may not generalize to work settings, to more heterogeneous populations, such as working managers, and to dyads in which the communicator has higher status than the news recipient. In neither study did communicator gender have the expected positive effect on interactional justice, which is contrary to previous research showing that when giving bad news women lied less often, offered more concessions and fewer refusals, and paid more attention to the other person's needs than men (e.g., Gonzales et al., 1992). However, these findings generally related to non-organizational contexts. It is possible that the contexts and samples in my studies reduced the effect of gender on communication. In Study 1, participants were experienced managers communicating in a professional context. It is possible that, first, female managers may have become socialized to communicate in the same ways as their male counterparts, and, second, female managers may have been observing norms for communicating in a business context (Fagenson, 1990; Posner & Schmidt, 1984). In Study 2, the majority of participants were upper-level business students who might similarly have been influenced in their communication by perceived norms in a business school context for being more assertive and direct, and less recipient-oriented and interpersonally sensitive. 100 Taken together, the multiple methods used in this dissertation should provide confidence in the findings. First, a scenario exercise and a sample of practicing managers were used to used to investigate the relationship between several important individual differences and managerial interactional fairness in communicating negative news. In Study 1, empathic concern, personal distress, self-esteem, moral development, and strategic emotional intelligence were measured so causal effects on interactional justice could not be established. Second, Study 2 involved a laboratory experiment in which the communication task was believed to be real, face-to-face communication was used, and empathic concern and moral identity were manipulated, in order that causality could be determined. Future Research To date, relatively little is known about why individuals choose to act in a fair manner (Korsgaard et al., 1998). In this dissertation, I have found that individual differences predict, directly or interactively, the interactional fairness with which negative news is communicated. I discuss below several promising areas for further investigation, including other psychological and situational factors that influence interactional fairness and additional facets of fair communication. Future research should investigate additional personal characteristics of the communicator which may influence the effect of empathy, moral identity, and moral development on interactional justice. The Four Component Model (Rest, 1986) proposed that moral sensitivity, reasoning, motivation, and character all contributed to moral behavior. The influence of moral character (i.e., a person's character, persistence, and ability to act on a moral decision) on the relationships observed should be investigated as 101 it might provide another condition for moral development to translate into interactional justice. Previous research has shown that training managers in organizational justice principles can increase the fairness with which their subordinates are treated (Skarlicki & Latham, 1996). An important next step would be a field study in the form of a quasi experiment in which managers were trained in empathy strategies, moral reasoning, and organizational justice principles. In this dissertation I focused on the effects on interactional justice of several individual differences between communicators and an empathic induction. However, group- and organization-level influences on interactionally fair communication may also be considerable. For example, organizations vary considerably in the environments they create to encourage or discourage the compassionate treatment of employees, as well as the support they provide for managers who incur the personal costs of communicating in a less defensive and more other-oriented fashion (Frost, 2002). In addition, teams within organizations teams develop norms for communication over time that constrain and guide both communicative behavior, as well as perceptions of what is regarded as acceptable communication strategies (Schmitz & Fulk, 1991). For the above reasons, future research should also examine the factors that cause differences between organizations, and groups within organizations, in how negative news is communicated. I considered the effect of several individual differences and two interventions on several important aspects of interactionally fair communication: politeness, respect and concern, refraining from improper comments, providing reasonable and thorough explanations, and tailoring the communication to the news recipient. However, the effects on additional aspects of interactional fairness should also be investigated. These could 102 include the extent to which communicators delegate the delivery of bad news to subordinates or outside parties, or the extent to which a message violates privacy concerns of the bad news recipient (Bies, 2001). In addition, future research should investigate situations in which behavior that is regarded as moral may be regarded as interactionally unjust. Finally, a great deal of communication within organizations takes place within ongoing relationships. The way in which recipients of negative news behave is an important influence on the interactional fairness of managers (Korsgaard et al., 1998). I would also expect that a manager's communication behavior is influenced by how employees have previously responded to negative news. It may be worth investigating interactionally fair communication, and responses to it, within work groups and specific dyads over time. Conclusion This dissertation does not take the position that managers who underutilize sensitive communication strategies are ignorant regarding recipient needs or expectations for interactional fairness, oblivious to the pain they can inflict, or malicious. Rather, the research-practitioner gap underscores findings that managers do not demonstrate behaviors that reflect scientific research. In this dissertation I explore the challenge of having to communicate negative news, and the complex interplay of moral and emotional factors that lead some communicators to accord fair treatment to recipients, and others to distance themselves from the victims of bad news. Negative workplace news can be difficult, sometimes impossible, to avoid. We can help managers mitigate the negative effect, on employees and on the organization, of the negative outcome. One reason for 103 doing so is that using interactional fairness is pragmatic and relatively inexpensive. But treating recipient of bad news with respect and sincerely explaining to them the procedures followed and the reasons for adverse outcomes is also the right thing to do. Bies (2001) said "We, as justice researchers, must reflect on our responsibilities to conduct research that achieves not only "statistical" significance but also "moral" significance" (p. 107). With a better understanding of why managers often make bad times worse by communicating with the victims of bad news in ways that are likely to be seen as interactionally unfair, we can begin to propose strategies for motivating communication that is more interpersonally and informationally fair. 104 REFERENCES Adams, J. S. (1965). Inequity in social exchange. In L . Berkowitz (Ed.), Advances in experimental social psychology, 2 (pp. 267-299). New York: Academic Press. Aiken, L. S. & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage. Aquino, K., Ray, S., & Reed, A. (2003). Moral identity as a predictor of lying in negotiations. Paper presented at the annual meeting of the Academy of Management, Seattle, WA. Aquino, K., & Reed II, A. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423-1440. Ashkanasy, N. M . (2003). Emotions in organizations: A multilevel perspective. In F. Dansereau, & F. J. Yammarino (Eds.), Research in multi-level issues: Multi-level issues in organizational behavior and strategy (Vol. 2, pp. 9-54). Oxford, UK: Elsevier Science. Ashkanasy, N. M . , Hartel, C. E. J., & Zerbe, W. (Eds.) (2000). Emotions in the workplace: Research, theory, and practice. Westport, CT: Quorum. Athos, A. G., & Gabarro, J. J. (1978). Interpersonal behavior: Communication and understanding in relationships. Englewood Cliffs, NJ: Prentice-Hall. Barclay, L. J., Skarlicki, D. P., & Pugh, D. S. (2005). Exploring the role of emotions in injustice perceptions and retaliation. Journal of Applied Psychology, 90, 629-643. Bar-On, R. (1997). Bar-On emotional quotient inventory: Technical manual (EQ-i). Toronto, Canada: Multi-Health Systems. Batson, C. D. 1987. Self-report ratings of empathic emotion. In N. Eisenberg & J. Strayer (Eds.), Empathy and its development (pp. 356-360). Cambridge University Press: Cambridge, UK. 105 Batson, C. D. 1991. The altruism question: Toward a social-psychological answer. Hillsdale, NJ: Lawrence Erlbaum Associates. Batson, C. D. (1998). Altruism and prosocial behavior. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology, 2 (pp. 282-316). Boston: McGraw-Hill. Batson, C. D., Early, S., & Salvarani, G. (1997). Perspective taking: Imagining how another feels versus imagining how you would feel. Personality and Social Psychology Bulletin, 23, 751-757. Batson, C. D., Fultz, J., & Schoenrade, P. A. (1987). Adults' emotional reactions to the distress of others. In N. Eisenberg & J. Strayer (Eds.), Empathy and its development (pp. 163-184). Cambridge, UK: Cambridge University Press. Batson, C D . , Sager, K., Garst, E. , Kang, M . , Rubchinsky, K., & Dawson, K. (1997). Is empathy-induced helping due to self-other merging? Journal of Personality and Social Psychology, 73, 495-509. Baumeister, R. F. (1993). Understanding the inner nature of low self-esteem: Uncertain, fragile, protective, and conflicted. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 201 - 218). New York: Plenum Press. Bennett, N., Martin, C. L. , Bies, R. J., & Brockner, J. (1995). Coping with a layoff: A longitudinal study of victims. Journal of Management, 21(6), 1025-1034. Bergman, R. (2002). Why be moral: A conceptual model from developmental psychology. Human Development, 45, 104-124. 106 Bies, R. J. (1987). The predicament of injustice: The management of moral outrage. In L. L. Cummings & B. M . Staw (Eds.), Research in organizational behavior (Vol. 9, pp. 289-318). Greenwich, CT: JAL. Bies, R. J. (2001). Interactional (injustice: The sacred and the profane. In R. Cropanzano & J. Greenberg (Eds.), Advances in organizational justice (pp. 89-118). Stanford, CA: Stanford University Press. Bies, R. J., & Moag, J. S. (1986). Interactional justice: Communication criteria of fairness. In R. Lewicki, B. H. Sheppard. & M . H. Bazerman (Eds.), Research on negotiation in organizations (Vol. 1, pp. 43-55). Greenwich, CT: JAI Press. Bies, R. J., & Tyler, T. (1993). The "litigation mentality" in organizations: A test of alternative psychological explanations. Organization Science, 4, 352-366. Blaine, B., & Crocker, J. (1993). Self-esteem and self-serving biases in reactions to positive and negative events: An integrative review. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 55 - 85). New York: Plenum Press. Blasi, A. (1980). Bridging moral cognition and moral action: A critical review of the literature. Psychological Bulletin, 88, 1-45. Blasi, A. (1984). Moral identity: Its role in moral functioning. In W.M. Kurtines & J.L. Gewirtz (Eds.), Morality, Moral Behavior, and Moral Development. New York: John Wiley. Brockner, J. (1988). Self-esteem at work: Research, theory, and practice. Lexington, MA: D.C. Heath and Company. Brockner, J., Derr, W. R., & Liang, W. N. (1987). Self-esteem and reactions to negative feedback: Towards greater generalizability. Journal of Research in Personality, 21,318-333. 107 Brockner, J., & Wiesenfeld, B. M . (1996). An integrative framework for explaining reactions to decisions: Interactive effects of outcomes and procedures. Psychological Bulletin, 120, 189-208. Brown, J. D., & Dutton, K. A. (1995). The thrill of victory, the complexity of defeat: Self -esteem and people's emotional reactions to success and failure. Journal of Personality and Social Psychology, 68, 712-722. Brown, P., & Levinson, S. (1987). Politeness: Some universals in language usage. Cambridge, UK: Cambridge University Press. Campbell, J. D., & Lavallee, L. F. (1993). Who am I?: The role of self-concept confusion in understanding the behavior of people with low self-esteem. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 3 - 20). New York: Plenum Press. Clark, R. A. (1979). The impact on selection of persuasive strategies of self-interest and desired liking. Communication Monographs, 46, 257-273. Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A meta-analysis. Organizational Behavior and Human Decision Processes, 86 , 278-321. Coke, J. S., Batson, C. D., & McDavis, K. (1978). Empathic mediation of helping: A two-stage model. Journal of Personality and Social Psychology, 36, 752-766. Cooley, C. H. (1902). Human nature and the social order. New York: Scribners. Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: W. H. Freeman. Colquitt, J. A., Conlon, D., Wesson, M . , Porter, C , & Ng, Y. (2001). Justice at the millennium: A meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychology, 86(3), 425-445. Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: W. H. Freeman. 108 Cortina, J. M . (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98-104. Cropanzano, R., & Ambrose, M . L. (2001). Procedural and distributive justice are more similar than you think: A monistic perspective and a research agenda. In J. Greenberg & R. Cropanzano, (Eds.), Advances in organizational justice (pp. 119-151). Stanford: Stanford University Press. Cropanzano, R., Rupp, D. E. , Mohler, C. J., & Schminke, M . (2001). Three roads to organizational justice. In J. Ferris (Ed.), Research in personnel and human resources management (Vol. 20, pp. 1-113). Greenwich, CT: JAI Press. Crowther, C. (2001). Face-to-face or email: The medium makes a difference. Communication World, 118,23-24. Daus, C. S., & Ashkanasy, N. M . (2005). The case for the ability-based model of emotional intelligence in organizational behavior. Journal of Organizational Behavior, 26, 453— 466. Davis, M . H. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10(4), 85. Davis, M . H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44(1), 113-126. Davis, M . H. (1994). Empathy: A social psychological approach. Boulder, Colorado: Westview Press. 109 Davis, M . H., & Oathout, H. A. (1987). Maintenance of satisfaction in romantic relationships: Empathy and relational competence. Journal of Personality and Social Psychology, 53, 397-410. Downey, G., & Feldman, S. I. (1996). Implications of rejection sensitivity for intimate relationships. Journal of Personality and Social Psychology, 70(6), 1327-1343. Dubose, C. (1994). Breaking the bad news. HRMagazine, 39(4), 62-64. Dunnette, M . D. (1990). Blending the science and practice of industrial and organizational psychology: Where are we and where are we going? In M . D. Dunnette and L. M . Hough (Eds.) Handbook of industrial and organizational psychology (Vol. 1, pp. 1-27). Palo Alto, CA: Consulting Psychologists Press. Dutton, J. E. , Frost, P. J., Worline, M . C , Lilius, J. M . , & Kanov, J. M . (2002). Leading in times of trauma. Harvard Business Review, 80(1), 55-61. Eisenberg, N. (2000). Emotion, regulation, and moral development. Annual Review of Psychology, 51, 665-697. Eisenberg, N. (1986). Altruistic emotion, cognition, and behavior. Hillsdale, NJ: Erlbaum and Associates. Eisenberg, N. (2000). Emotion, regulation, and moral development. Annual Review of Psychology, 51, 665-697. Eisenberg, N., & Miller, P. A. (1987). Empathy and prosocial behavior. Psychological Bulletin, 101, 91-119. Eisenberg, N., Shea, C. L. , Carlo, G., & Knight, G. P. (1991). Empathy-related responding and cognition: A "chicken and the egg" dilemma. In W. Kurtines & J. Gewirtz (Eds.), Handbook of moral behavior and development. Volume 2: Research, pp. 63-88. Hillsdale, NJ: Lawrence Erlbaum Associates. Eisenberg, N., & Strayer, J. (1987). Critical issues in the study of empathy. In N. Eisenberg & J. Strayer (Eds.), Empathy and its development (pp. 3-16). Cambridge University Press: Cambridge, UK. Fagenson, E. A., 1990. At the heart of women in management research: Theoretical and methodological approaches and their biases. Journal of Business Ethics, 9, 261-21 A. Feather, N. T. 1999. Values, achievement, and justice: Studies in the psychology of deservingness. New York: Kluwer Academic / Plenum Publishers. Feinberg, J. 1974. Noncomparative justice. Philosophical Review, 83: 297-338. Folger, R. (1993). Reactions to mistreatment at work. In K. Murnighan (Ed.), Social psychology in organizations: Advances in theory and research (pp. 161-183). Englewood Cliffs, NJ: Prentice-Hall. Folger, R. (1998). Fairness as a moral virtue. In M . Schminke (Ed.), Managerial ethics: Morally managing people and processes (pp. 13-34). Mahwah, NJ: Lawrence Erlbaum Associates. Folger, R. (2001). Fairness as deonance. In S. Gilliland, D. Steiner and D. P. Skarlicki (Eds.), Research in social issues in management (pp. 3-33). Greenwich, Conn.: Information Age Publishing. Folger, R., & Skarlicki, D. P. (1998). When tough times make tough bosses: Managerial distancing as a function of layoff blame. Academy of Management Journal, 41(1), 79-87. I l l Folger, R., & Skarlicki, D. P. (2001). Fairness as a dependent variable: Why tough times can lead to bad management. In R. Cropanzano (Ed.), Justice in the workplace: Volume II -From theory to practice (pp. 97-118). Mahwah, NJ: Erlbaum. Frost, P. J. (2002). Toxic emotions at work: How compassionate managers handle pain and conflict. Boston: Harvard Business School Press. George, J. M . (1991). State or trait: Effects of positive mood on prosocial behaviors at work. Journal of Applied Psychology, 76(2), 299-307. , George, J. M . (2000). Emotions and leadership: The role of emotional intelligence. Human Relations, 53, 1027-1055. Goleman, D. (1995). Emotional intelligence: Why it can matter more than IQ. London: Bloomsbury Publishing. Goleman, D., Boyatzis, R., & McKee, A. (2002). Primal leadership: Realizing the power of emotional intelligence. Boston: Harvard Business School Press. Gonzales, M . H , Manning, D. J., & Haugen, J. A. (1992). Explaining our sins: factors influencing offender accounts and anticipated victim responses. Journal of Personality and Social Psychology, 62: 958-971. Greenberg, J. (1990a). Employee theft as a reaction to underpayment inequity: The hidden cost of pay cuts. Journal of Applied Psychology, 75, 561-568. Greenberg, J. (1990b). Looking fair vs. being fair: Managing impressions of organizational justice. Research in Organizational Behavior, 12, 111-157. Greenberg, J. (1993). The social side of fairness: Interpersonal and informational classes of organizational justice. In R. Cropanzano (Ed.), Justice in the workplace: Approaching fairness in human resource management (pp. 79-103). Hillsdale, NJ: Lawrence Erlbaum. 112 Greenberg, J. (1994). Using socially fair treatment to promote acceptance of a work site smoking ban. Journal of Applied Psychology, 79(2), 288-297. Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60(6), 895-910. Higgins, A., Power, C., & Kohlberg, L. (1984). The relationship of moral atmosphere to judgments of responsibility. In W. M . Kurtines & J. L. Gewirtz (Eds.), Morality, moral behavior and moral development (pp. 74-106). New York: Wiley. Hoffman, M . L. (2000). Empathy and moral development: Implications for caringand justice. New York, NY: Cambridge university Press. Johnson, R. E. , Conlee, M . C , & Tesser, A. (1974). Effects of similarity of fate on bad news transmission: A reexamination. Journal of Personality and Social Psychology, 29, 644-648. Jones, T. M . (1991). Ethical decision making by individuals in organizations: An issue-contingent model. Academy of Management Review, 16, 366-395. Kohlberg, L. (1969). Stage and sequence: The cognitive developmental approach to socialization. In D. A. Goslin (Ed.), Handbook of socialization theory and research (pp. 347-380). Chicago, IL: Rand McNally. Kohlberg, L. (1983). Moral stages: A current formulation and a response to critics. Basel, Switzerland: Karger. Kohlberg, L. (1984). The psychology of moral development: The nature and validity of moral stages. San Francisco: Harper & Row. Kohlberg, L. , Levine, C , & Hewer, A. (1983). Moral stages: A current formulation and a response to critics. Basel, Switzerland: Karger. Korsgaard, M . A., Rymph, R. D., & Roberson, L. (1998). What motivates fairness? The role of subordinate assertive behavior on managers' interactional fairness. Journal of Applied Psychology, 83,131-1AA. Kray, L. J., & Lind, E. A. (2002). The injustices of others: Social reports and the integration of others' experiences in organizational justice judgments. Organizational Behavior and Human Decision Processes, 89, 906-924. Landy, F. J. (2005). Some historical and scientific issues related to research on emotional intelligence. Journal of Organizational Behavior, 26, 411-424. Leary, M . R. (2001). Toward a conceptualization of interpersonal rejection. In M . R. Leary (Ed.) Interpersonal Rejection (pp. 3-20). New York, NY: Oxford University Press. Leary, M . R., Koch, E. J. & Hechenbleiker, N. R. (2001). Emotional responses to interpersonal rejection. In M . R. Leary (Ed.) Interpersonal rejection (pp. 145-166). New York, NY: Oxford University Press. Lerner, M . J. (1980). The belief in a just world: A fundamental delusion. New York: Plenum. Leventhal, G. S. (1976). The distribution of rewards and resources in groups and organizations. In L. Berkowitz & E. Walster (Eds.), Advances in experimental social psychology (Vol. 9, pp. 91-131). New York: Academic. Lind, E . A., Greenberg, J., Scott, K. S., & Welchans, T. D. (2000). The winding road from employee to complainant: Situational and psychological determinants of wrongful termination claims. Administrative Science Quarterly, 45(3), 557-590. Lind, E. A., & Tyler, T. R. (1988). The social psychology of procedural justice. New York: Plenum. Lopes, P. N., Bracket!, M . A., Nezlek, J. B., Schutz, A., Sellin, I., & Salovey, P. (2004). Emotional intelligence and social interaction. Personality and Social Psychology Bulletin, 30, 1018-1034. Lopes, P., & Salovey, P. (2001). Emotional intelligence and social-emotional learning: Assessing emotional intelligence and developing skills and flexibility. CEIC Review, 10, 12-13. Manis, M . , Cornell, S. D., & Moore, J. C. (1974). The transmission of attitude-relevant information through a communication chain. Journal of Personality and Social Psychology, 30, 81-94. Mayer, J. D., & Caruso, D. (Nov/Dec 2002). The effective leader: Understanding and applying emotional intelligence. Ivey Business Journal, 1. Mayer, J., Salovey, P., & Caruso, D. (2002). MSCEIT user's manual. Toronto: Multi-Health Systems. McEnrue, M . P., & Groves, K. (2006). Choosing among tests of emotional intelligence: What is the evidence? Human Resource Development Quarterly, 17, 9-42. McGraw, K. O., & Wong, S. P. (1996). Forming inferences about some intraclass correlation coefficients. Psychological Methods, 1(1), 30-46. McLaughlin, M . L, Cody, M . J., & O'Hair, H. D. (1983). The management of failure events: Some contextual determinants of accounting behavior. Human Communication Research, 9, 208-224 Messick, D. M . , Bloom, S., Boldizer, J. P., & Samuelson, C. D. (1985). Why we are fairer than others. Journal of Experimental Social Psychology, 21, 480-500. 115 Mikula, G., Petrick, B., & Tanzer, N. (1990). What people regard as unjust: Types and structures of everyday experiences of injustice. European Journal of Social Psychology, 20, 133-149. Miller, P. A. & Eisenberg, N. (1988). The relation of empathy to aggressive and externalizing/antisocial behavior. Psychological Bulletin, 103(3), 324-344. Morand, D. A. (2000). Language and power: An empirical analysis of linguistic strategies used in superior-subordinate communication. Journal of Organizational Behavior, 21, 235-248. O'Sullivan, P. B. (2000). What you don't know won't hurt me: Impression management functions of communication channels in relationships. Human Communication Research, 26(3), 403-431. Posner, G. N., & Schmidt, W. H. (1984). Sex effects on managerial value systems. Human Relations, 37,909-921. Redmond, M . V. (1989). The functions of empathy (decentering) in human relations. Human Relations, 42(1), 593-605. Reed II, A., & Aquino, K. (2003). Moral identity and the expanding circle of moral regard toward out-groups. Journal of Personality and Social Psychology, 84, 1270-1286. Reed II, A., Aquino, K., & Levy, E. (2006). Moral identity and judgments of charitable behavior. Manuscript submitted for publication. Rest, J. (1979). Development in judging moral issues. Minneapolis: University of Minnesota Press. Rest, J. (1986). Moral development: Advances in research and theory. New York: Praeger. Rest, J. R. (1994). Background: Theory and research. In J. R. Rest & D. Narvaez (Eds.) Moral development in the professions: Psychology and applied ethics (pp. 1-26). Hillsdale, NJ: Lawrence Erlbaum. Rest, J. R., Narvaez, D., Thoma, S. J., & Bebeau, M . J. (1999). D1T2: Devising and testing a revised instrument of moral judgment. Journal of Educational Psychology, 91, 644-659. Reynolds, S. J. (2006). Moral awareness and ethical predispositions: Investigating the role of individual differences in the recognition of moral issues. Journal of Applied Psychology, 91, 233-243. Reynolds, S. J., & Ceranic, T . L . (2006). The effects of moral judgment and moral identity on moral behavior: An empirical investigation of the moral individual. Manuscript submitted for publication. Rosen, S., & Tesser, A. (1972). Fear of negative evaluation and the reluctance to transmit bad news. Journal of Communication, 22, 124-141. Rosenberg, M . (1979). Conceiving the self. New York: Basic Books. Rosenberg, M . (2001). Low self-esteem people: A collective portrait. In T. J. Owens, S. Stryker, and N. Goodman (Eds.), Extending self-esteem theory and research: Sociological and psychological currents (pp. 400-436). Cambridge, UK: Cambridge University Press. Rosete, D., & Ciarrochi, J. (2005). Emotional intelligence and its relationship to workplace performance outcomes of leadership effectiveness. Leadership & Organization Development Journal, 26, 388-399. Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition, and Personality, 9, 185-211. 117 Schmitz, J. & Fulk, J. (1991). Organizational colleagues, media richness, and electronic mail. Communication Research, 18, 487-523. Schonbach, P. (1980). A category system for account phases. European Journal of Social Psychology, 10, 195-200. Shapiro, D. L. (1991). The effects of explanations on negative reactions to deceit. Administrative Science Quarterly, 36, 614-630. Shapiro, D. L. , Buttner, E. H., & Barry, B. (1994). Explanations: What factors enhance their perceived accuracy. Organizational and Human Decision Processes, 58, 346-368. Shrout, P. E. , & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86(2), 420-428. Sitkin, S. B., & Bies, R. J. (1993). The legalistic organization: definitions, dimensions, and dilemmas. Organization Science, 4, 345-351. Sitkin, S. B., Sutcliffe, K. M . , & Barrios-Chopin, J. (1992). A dual capacity model of communication media choice in organizations. Human Communication Research, 18, 563-598. Skarlicki, D. P., Ellard, J. H., & Kelln, B. R. C. (1998). Third-party perceptions of a layoff: Procedural, derogation, and retributive aspects of justice. Journal of Applied Psychology, S3,119-127. Skarlicki, D. P., & Folger, R. (1997). Retaliation in the workplace: The roles of distributive, procedural and interactional justice. Journal of Applied Psychology, 82, 434-443. Skarlicki, D. P., & Latham, G. P. (1996). Increasing citizenship behavior within a labor union: A test of organizational justice theory. Journal of Applied Psychology, 81, 161-169. 118 Skitka, L. J. (2003). Of different minds: An accessible identity model of justice reasoning. Personality and Social Psychology Review, 7, 286-297. Skitka, L. J. (2005). An Accessible Identity Approach to understanding fairness in organizational settings. In K. van den Bos, S. W. Gilliland, D. D. Steiner & D. P. Skarlicki (Eds.), What motivates fairness in organizations? (pp. 105-128). Greenwich, CT: Information Age Publishing. Sommer, K. L. (2001). Coping with rejection: Ego-defensive strategies, self-esteem, and interpersonal relationships. In M . R. Leary (Ed.) Interpersonal rejection (pp. 167-188). New York, NY: Oxford University Press. Sommer, K. L. , & Baumeister, R. K. (2002). Self-evaluation, persistence, and performance following implicit rejections: The role of trait self-esteem. Personality and Social Psychology Bulletin, 28(1), 926-938. Strayer, J. (1987). Affective and cognitive perspectives on empathy. In N. Eisenberg & J. Strayer (Eds.), Empathy and its development (pp. 218-244). Cambridge University Press: Cambridge, UK. Tesser, A., & Rosen, S. (1975). The reluctance to transmit bad news. Advances in Experimental Social Psychology, 8, 193-232. Tesser, A., Rosen, S., & Batchelor, T. (1972). Some message variables and the M U M effect. Journal of Communication, 22, 239-256. Thibaut, J., & Walker, L. (1975). Procedural justice: A psychological analysis. New York: Lexington. Thoma, S. (1994). Moral judgments and moral action. In J. R. Rest & D. Narvaez (Eds.) Moral development in the professions: Psychology and applied ethics (pp. 199-211). Hillsdale, NJ: Lawrence Erlbaum. Tice, D. M . (1993). The social motivations of people with low self-esteem. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 37-53). New York: Plenum Press. Toi, M . , & Batson, C. D. (1982). More evidence that empathy is a source of altruistic motivation. Journal of Personality and Social Psychology, 43, 281-292. Trevino, L. K. (1986). Ethical decision making in organizations: A person-situation interactionist model. Academy of Management Review, 11, 601-617. Twenge, J. M . , Baumeister, R. F., Tice, D. M . , & Stucke, T. S. (2001). If you can't join them, beat them: Effects of social exclusion on aggressive behavior. Journal of Personality and Social Psychology, 81(6), 1058-1069. Tyler, T. R., & Lind, E. A. (1992). A relational model of authority in groups. Advances in experimental social psychology (Vol. 25, pp. 115-191). New York: Academic Press. Tyler, T., Boekmann, R. J., Smith, H. J., & Huo, Y. J. (1997). Social justice in a diverse society. Boulder, CO: Westview Press. Underwood, B., & Moore, B. (1982). Perspective-taking and altruism. Psychological Bulletin, 91(1), 143-173. Viswesvaran, C , Ones, D. S., & Schmidt, F. L. (1996). Comparative analysis of the reliability of job performance ratings. Journal of Applied Psychology, 81, 557-574. Whetton, D., & Cameron, K. (2005). Developing management skills (6th ed). Upper Saddle River, NJ: Pearson/Prentice Hall. 120 Wong, C. S., & Law, K. S. (2002). The effect of leader and follower emotional intelligence on performance and attitude: An exploratory study. Leadership Quarterly, 13, 243-274. 125 Table 1 Means, Standard Deviations, Correlations, and Reliability Estimates for Study la Mean SD 1 2 3 4 5 6 7 8 9 1. Trait empathic concern 4.28 1.26 (.57) 2. Trait personal distress 1.75 .70 .53** (.78) 3. Self-esteem 5.60 1.25 -.28** -.50** (.84) 4. Moral development 41.07 13.90 .07 -.01 .04 (.81) 5. Strategic emotional intelligence 51.27 18.98 .01 -.11 .08 .22** (.77) 6. Interactional justice 15.71 4.26 .15* .03 .04 .19* .18* (.72) 7. Gender 1.66 .48 -.08 -.02 .07 .30** .16* .11 n.a. 8. Experience being fired 1.65 .48 -.07 -.05 .10 -.04 .09 -.02 .03 n.a. 9. Age 44.51 8.35 -.13 .03 -.02 -.12 -.14 -.12 -.09 .07 n.a. *N = 172 * p < .05 Note. Trait empathic concern, trait personal distress, state self-esteem, moral development, and interactional justice could range from 1 to 7, with higher scores representing greater amounts of the measure. Table 2 Study 1 - Results of Hierarchical Regression Analyses for All Predictors" Interactional Justice Variable Control variables Gender Age Experience being fired Predictor variables Trait Empathic Concern Moral development Strategic emotional intelligence Trait Personal Distress State Self-Esteem Interaction Trait Empathic Concern x Self-Esteem Trait Empathic Concern x Moral Development R 2 F Change in R 2 a7V= 172 *p< .05, ** p < .01 Step 1 Step 2 Step 3 .10 .06 .10 -.11 -.06 -.11 -.02 -.02 -.02 .16 .16 .12 .12 .13 .13 -.01 - .01 .06 .06 .02 -.15* .03 .09 .11 1.44 1.87 2.13* .06 .02 Table 3 Results of Hierarchical Regression Analyses for Empathic Concern in Study la Interactional Justice Variable Step 1 Step 2 Control variables Gender 1 0 - 1 2 Age -11 --09 Experience being fired "-02 -01 Predictor variables Trait Empathic Concern .15* R 2 .03 .05 F 1.46 2.05 Change in R 2 .02_ aN= 172 *p< .05, ** p< .01 Table 4 Results of Hierarchical Regression Analyses for Personal Distress in Study la Interactional Justice Variable Step 1 Step 2 Control variables Gender - 1 0 - 1 0 Age -11 Experience being fired --02 --02 Predictor variables Trait Personal Distress .03 R 2 .03 .03 F 1.46 1.13 Change in R 2 .00_ aN= 172 *p< .05, ** p < .01 Table 5 Results of Hierarchical Regression Analyses for Self-Esteem in Study la Interactional Justice Variable Step 1 Step 2 Control variables Gender 1 0 1 0 Age Experience being fired -.11 -.11 -.02 -.02 Predictor variables State Self-Esteem .03 R 2 .03 .03 F 1.46 1.13 Change in R 2 .00 a7V=172 *p< .05, ** p < .01 130 Table 6 Results of Hierarchical Regression Analyses for Interaction between Self-Esteem and Empathic Concern in Study la Interactional Justice Variable Step 1 Step 2 Step 3 Control variables Gender .10 .11 .11 Age -.11 -.09 -.09 Experience being fired -.02 -.02 -.02 Predictor variables Self-Esteem .08 .08 Trait Empathic Concern .17* .17 Interaction Self-Esteem x Trait Empathic Concern .01 R 2 .03 .05 .05 F 1.45 1.85 1.53 Change in R 2 .03 .00 a N=172 * p < .05, ** p < .01 Note. All variables were mean-centered, following the procedures recommended by Aiken and West (1991). Table 7 Standard Errors and t Tests for Simple Slopes of Regression of Interactional Justice on State Self-Esteem in Study la Self-Esteem Simple Slope SE t(172) Low 0.19 0.12 1.56 High 0.17 0.10 1.65 *N=\12 *p< .05, ** p < .01 Table 8 Results of Hierarchical Regression Analyses for Moral Development in Study la Interactional Justice Variable Step 1 Step 2 Control variables Gender 1 0 0 6 Age --10 Experience being fired " - ^ --01 Predictor variables Moral Development . 16* R 2 .03 .05 F 1.46 2.07 Change in R 2 CJ2_ a N = 172 * p < .05, ** p < .01 133 Table 9 Results of Hierarchical Regression Analyses for Interaction between Moral Development and Empathic Concern in Study 1" Interactional Justice Variable Step 1 Step 2 Step 3 Control variables Gender Age Experience being fired .10 .07 .05 -.11 -.08 -.08 -.02 .00 .01 Predictor variables Trait Empathic Concern .15 .14 Moral Development .14 .14 Interaction Moral Development x Trait Empathic Concern -.15* R 2 .03 .07 .09 F 1.44 2.30 2.60 Change in R 2 .04 .02 a N = 172 * p < .05, ** p < .01 Note. All variables were mean-centered, following the procedures recommended by Aiken and West (1991). Table 10 Standard Errors and t Tests for Simple Slopes of Regression of Interactional Justice on Trait Empathic Concern in Study I Moral Development Simple Slope SE t(172) Low 0.30 0.11 2.85** High -0.03 0.11 -0.26 aN = 172 * p < .05, **p<.01 Table 11 Results of Hierarchical Regression Analyses for Strategic Emotional Intelligence in Study la Interactional Justice Variable Step 1 Step 2 Control variables Gender - 1 0 - 0 8 Age --09 Experience being fired --03 Predictor variables Strategic Emotional Intelligence .15* R 2 .03 .05 F 1.45 2.05 Change in R 2 Tj2_ aN= 172 * p < .05 136 Table 12 Means, Standard Deviations, Correlations, and Reliability Estimates for Study 2a Mean S.D. 1 2 - 3 4 5 6 7 8 9 10 11 12 13 14 1. Group 1 .11 .32 n.a. 2. Group 2 .10 .30 -.12 n.a. 3. Group 3 .23 .43 -.20 -.18 n.a. 4. Group 4 .17 .38 -.17 -.15 -.25* n.a. 5. Group 5 .38 .49 -28* -.26* __44** -.36 ** n.a. 6. Gender 1.74 .44 -.06 .10 .20 -.10 -.11 n.a. 7. Age 22.02 1.41 .18 .09 - 30** .08 .03 -.14 n.a 8. Trait Empathic Concern 3.87 .57 -.02 .01 .18 -.05 -.11 .13 -.01 (.74) 9. Moral Identity Prime .49 .50 .12 .09 -.02 .01 -.11 .08 -.08 -.07 n.a. 10. Empathic Induction .52 V .50 -.05 -.01 .01 .11 -.06 .05 -.02 .12 -.04 n.a. 11. Moral Development 33.31 13.71 .05 .05 .09 .12 -.24* .08 -.02 .23* -.15 -.08 (.81) 12. Moral Identity (Int.) 4.44 .48 -.06 .16 .24* -.01 -.26* .09 -.10 .45 ** .26* .10 :20 (.70) 13. Int. Justice - Coders 21.46 2.80 .04 -.11 .06 -.06 .03 -.12 -.08 .08 -.05 .35** .03 .05 (.81) 14. Int. Justice - Conf. 22.39 3.14 .09 -.11 .22* -.11 -.10 .11 -.15 .30** -.06 40** .03 .11 .42** (.68) N = 81, * p < .05, ** p < .01. Note. Trait empathic concern, moral identity (internalization) could range from 1 to 5, with higher scores representing greater amounts of the measures. Interactional justice could range from 6 to 30, with higher scores representing greater amounts of the measures. Moral development is the N2 score, as provided by the Center for the Study of Ethical Development at the University of Minnesota. 137 Table 13 Cell Means for Interactional Justice in Study 2 Condition Low Moral Identity High Moral Identity aiV=81 * p < .05, ** p < .01 Empathic Concern Manipulation Control Condition M SD M SD 22.41 3.12 20.04 2.78 21.93 2.45 20.25 2.37 Table 14 Cell Means for Confederate-Rated Interactional Justice in Study 2 Condition Low Moral Identity High Moral Identity a/V=81 *p< .05, ** p < .01 Empathic Concern Manipulation Control Condition M SD M SD 24.02 2.91 20.61 2.28 23.38 2.39 21.30 3.59 139 Table 15 Results of Hierarchical Regression Analyses for Moral Identity Internalization in Study 2a Interactional Justice Variable Step 1 Step 2 Step 3 Source and demographic variables Group 1 Group 2 Group 3 Group 4 Group 5 Gender Age .04 -.08 .04 -.06 -.13 .09 .04 -.08 .02 -.06 -.14 .10 .04 -.09 .01 -.06 -.13 .09 Other control variables Trait Empathic Concern Moral Identity Manipulation .09 .04 .07 -.05 Predictor variables Moral Identity Internalization .04 F Change in R 2 .04 .49 .05 .45 .01 .05 .40 .00 a iV=81 *p< .05, ** p < .01 140 Table 16 Results of Hierarchical Regression Analyses for Moral Identity Internalization in Study 2a Confederate-Rated Interactional Justice Variable Step 1 Step 2 Step 3 Source and demographic variables Group 1 .14 .15 .15 Group 2 -.09 -.09 -.08 Group 3 .16 .12 .13 Group 4 -.06 -.06 -.06 Group 5 Gender -.18 -.16 -.16 Age .13 .15 .15 Other control variables Trait Empathic Concern .20 .23 Moral Identity Manipulation -.03 -.02 Predictor variables Moral Identity Internalization -.06 R 2 .13 .17 .18 F 1.90 1.89 1.69 Change in R 2 .04 .00 a/V=81 *p< .05, **p< .01 141 Table 17 Results of Hierarchical Regression Analyses for Three-way Interaction in Study 2 Interactional Justice Variable Step 1 Step 2 Step 3 Step 4 Control variables Group 1 -04 .04 .02 -.00 Group.2 - - 0 8 --°9 "13 -.13 Group 3 -04 .01 .04 .06 Group 4 - ° 6 "-I2 --21 -21 Gender -.13 -.16 -.18 -.16 Age .09 .09 .05 .10 Independent and predictor variables Trait Empathic Concern .03 .05 .04 Moral Identity manipulation -.02 -.02 -.02 Empathic Concern Induction .38** .37** .33** Moral Development .08 .03 .07 Moral Identity Internalization .00 -.05 -.12 Two-Way Interactions Empathic Induction x Moral Development .29* .25* Empathic Induction x Moral Identity Int. -.08 .03 Moral Development x Moral Identity Int. -.21 -.35* Three-Way Interaction .26* Empathic Induction x Moral Development x Moral Identity Internalization R 2 .04 .18 .30 .34 F .49 1.39 2.01* 2.22* Change in R 2 .14 .12 .04 a N = 81,*/?<. 05, **/?<. 01 Note. All variables were mean-centered, following the procedures recommended by Aiken and West (1991). 142 Table 18 Results of Hierarchical Regression Analyses for Three-way Interaction in Study 2 Confederate-Rated Interactional Justice Variable Step 1 Step 2 Step 3 Step 4 Control variables Group 1 .14 .16 .15 .15 Group 2 -.09 -.07 -.09 -.09 Group 3 .16 .14 .14 .15 Group 4 -.06 -.10 -.14 -.14 Gender .18 .14 .16 .16 Age .13 .14 .14 .15 Independent variables Trait Empathic Concern .20 .20 .20 Moral Identity manipulation .00 -.02 -.02 Empathic Concern Induction .43** .43** .42** Moral Development -.02 -.03 -.03 Moral Identity Internalization -.09 -.15 -.16 Two-Way Interactions Empathic Induction x Moral Development .03 .02 Empathic Induction x Moral Identity Int. -.06 -.05 Moral Development x Moral Identity Int. -.17 -.19 Three-Way Interaction Empathic Induction x Moral Development x Moral Identity Internalization .04 R 2 .13 .36 .39 .39 F 1.90 3.55** 3.00** 2.77** Change in R 2 .23 .03 .00 N = 81, * p < .05, ** p < .01 Note. All variables were mean-centered, following the procedures recommended by Aiken and West (1991). Table 19 Standard Errors and t Tests for Simple Slopes of Regression of Interactional Justice on Empathic Concern Induction in Study 2a Condition Moral Development Simple Slope SE t(81) Low Moral Identity Low 1.02 .78 1.30 Low Moral Identity High 1.74 1.20 1.46 High Moral Identity Low -.06 .96 -.17 High Moral Identity High 3.00 .96 3.16* aJV=81 * p < .05, ** p < .01 144 Table 20 Results of Hierarchical Regression Analyses for Three-way Interaction in Study 2 Interactional Justice - Politeness Variable Step 1 Step 2 Step 3 Step 4 Control variables Group 1 -.03 -.03 -.04 -.06 Group 2 -.02 .00 -.03 -.03 Group 3 .13 .16 .19 .22 Group 4 -.02 -.05 -.12 -.12 Gender -.21 -.22 -.24 -.22 Age .08 .08 .05 .11 Independent and predictor variables Trait Empathic Concern -.08 -.06 -.07 Moral Identity manipulation .04 .03 .03 Empathic Concern Induction .28** .28** .23* Moral Development , .02 -.02 .03 Moral Identity Internalization -.09 -.13 -.21 Two-Way Interactions Empathic Induction x Moral Development .21 .16 Empathic Induction x Moral Identity Int. -.07 .07 Moral Development x Moral Identity Int. -.16 -.33* Three-Way Interaction .35* Empathic Induction x Moral Development x Moral Identity Internalization R 2 .07 .15 .22 .29 F .87 1.11 1.29 1.73 Change in R 2 .08 .07 .07 N = 81, * p < .05, **/?<.01 Note. All variables were mean-centered, following the procedures recommended by Aiken and West (1991). 145 Table 21 Standard Errors and t Tests for Simple Slopes of Regression of Interactional Justice(Politeness) on Empathic Concern Induction in Study 2a Moral Simple Condition Development Slope SE t(81) Low Moral Identity Low .15 .10 . 1.52 Low Moral Identity High .03 .15 .20 High Moral Identity Low -.04 .12 -.34 High Moral Identity a iV=81 * p < .05, ** p < .01 High .30 .12 2.59* 146 Table 22 Results of Hierarchical Regression Analyses for Three-way Interaction in Study 2 Interactional Justice - Respect Variable Step 1 Step 2 Step 3 Step 4 Control variables Group 1 .08 .10 .08 .06 Group 2 -.09 -.09 -.11 -.12 Group 3 .12 .10 .13 .16 Group 4 .01 * -.03 -.09 -.09 Gender -.12 -.13 -•14 -.12 Age .08 .08 .06 .12 Independent and predictor variables Trait Empathic Concern -.04 -.01 -.01 Moral Identity manipulation -.11 -.11 -.11 Empathic Concern Induction .27* .26* .21 Moral Development .06 .01 .06 Moral Identity Internalization .01 -.04 -.12 Two-Way Interactions Empathic Induction x Moral Development .21 .16 Empathic Induction x Moral Identity Int. -.03 .10 Moral Development x Moral Identity Int. -.19 -.35* Three-Way Interaction .31* Empathic Induction x Moral Development x Moral Identity Internalization. R 2 .05 .13 .21 .26 F .64 .95 1.21 1.53 Change in R 2 .08 .07 .06 a N = 81,*/?< .05, ** p < .01 Note. All variables were mean-centered, following the procedures recommended by Aiken and West (1991). Table 23 Standard Errors and t Tests for Simple Slopes of Regression of Interactional Justice (Dignity and Respect) on Empathic Concern Induction in Study 2a Moral Simple Condition Development Slope SE t(81) Low Moral Identity Low .10 .09 1.10 Low Moral Identity High .03 .14 .25 High Moral Identity Low -.02 .11 -.14 High Moral Identity a iV=81 * p < .05, ** p < .01 High .30 .11 2.86* 148 Figure 1. Factors leading to interactionally fair communication. Moral Identity Self-Esteem Communicator Concern (+) ^ \ Empathic Concern Personal t Distress (-) Interactionally Fair Communication Moral Development Emotional Reasoning 149 Figure 2. Interaction of trait empathic concern and moral development in predicting interactional justice in Study 1. 35-1 30 25 20 Low Moral Development High Moral Development Low High Trait Empathic Concern 150 Figure 3. Interaction of empathic concern and moral development in predicting interactional justice in low moral identity condition of Study 2. 151 Figure 4. Interaction of empathic induction and moral development in predicting interactional justice in high moral identity condition of Study 2. Low Moral Development High Moral development Control Empathic Induction Empathic Concern Condition 152 Figure 5. Interaction of empathic induction and moral development in predicting interactional justice (politeness) in low moral identity condition of Study 2. I 0 r ' L ' o 3 ^ 6 Low Moral Development High Moral Development Control Empathic Induction Empathic Concern Condition 153 Figure 6. Interaction of empathic induction and moral development in predicting interactional justice (politeness) in high moral identity condition of Study 2. iD U 3 "8? 101 8 M 6 Low Moral Development High Moral Development Control Empathic Induction Empathic Concern Condition 154 Figure 7. Interaction of empathic induction and moral development in predicting interactional justice (dignity and respect) in low moral identity condition of Study 2. o Control Empathic Induction Empathic Concern Condition Low Moral Development High Moral Development 155 Figure 8. Interaction of empathic induction and moral development in predicting interactional justice (dignity and respect) in high moral identity condition of Study 2. 10 Low Moral Development High Moral Development Control Empathic Induction Empathic Concern Condition 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092981/manifest

Comment

Related Items