Legitimacy Processes and Emotional Responses to Injustice

Abstract

People typically respond with negative emotions when they perceive an authority’s outcome distribution to be unjust. We argue, however, that legitimacy of the authority—“what others think” in terms of support coworkers and superiors extend to an individual occupying an authority position—acts as an opposing force, attenuating negative emotions and thus helping to sustain stratified orders. Likewise, legitimacy may stymie intentional displays of felt emotions. Our experiment examines the effects of perceived distributive injustice, legitimacy of the authority, and authority’s procedural fairness on members’ self-reported emotions and likelihood of intended emotional displays toward that authority, a superordinate authority, and coworkers in a work group context. Findings demonstrate that while perceived injustice arouses expected self-reported negative emotions, legitimacy (authorization as support by a superordinate authority and endorsement as support by coworkers) reduces such feelings. Also, strong authorization and endorsement indirectly affect intended emotional displays through self-reported negative emotions.

Keywords

legitimacy processes emotions justice processes authority group processes

Many studies of emotions examine the range of positive and negative emotions that individuals experience at work as well as the display of those feelings toward others, such as coworkers, subordinates, or superiors (e.g., Erickson and Ritter 2001; Lively 2000; Lively and Powell 2006; also see Lively and Weed 2014; Wharton 2014).¹ One focus of such studies pertains to negative emotional responses to injustice in work groups (see Hegtvedt and Parris 2014). Understanding such negative emotions cannot be underestimated because workers may experience psychological harm when they lack opportunities to express emotions resulting from perceived injustice. Given that negative emotions at work, felt and/or displayed, may increase job dissatisfaction and counterproductive work behaviors, understanding such processes is also vital for organizations (Clay-Warner 2006; Fox, Spector, and Miles 2001). Moreover, emotions are powerful forces in maintaining and changing social orders (Wingfield 2010; also see Foy et al. 2014; Lively 2008; Schwalbe and Shay 2014; Turner and Stets 2005).

Despite the importance of understanding negative emotions in workplace settings, studies have largely ignored the potential impact of coworkers and superiors on individuals’ perceptions of and emotional responses to injustice. We address that gap by investigating what individuals who perceive an injustice believe other people think about the situation and how that affects their experience and expression of negative emotions. “What others think” is central to legitimacy—a key process through which patterns of behavior or any cultural/social object gain social support (Johnson, Dowd, and Ridgeway 2006). When something, such as a rule or practice (e.g., family leave policy), a particular person in a position (e.g., a marketing firm manager), a position or structure of positions (e.g., an organizational hierarchy), or a distribution of resources (e.g., a factory wage scale), has legitimacy, there exists taken-for-granted support of this social object by a real or an implied social audience. Weber’s ([1924] 1978) classic formulation of legitimacy reveals that because actors perceive other people’s responses to support and accept an existing social order, the order seems as though it is a valid, objective feature of social life. Here, we argue that “what others think” may attenuate or heighten individuals’ self-reported negative emotions in response to an unjust work situation.

Undoubtedly, an individual who perceives that an authority’s distribution of outcomes is unjust will experience a negative emotional reaction (Homans 1974). Indeed, many studies demonstrate that unfair under-reward stimulates feelings of anger and resentment (see Cropanzano, Stein, and Nadisic 2011; Hegtvedt and Parris 2014). Emotional responses, however, involve the actual experience of feeling and its expression (Hochschild 1983), an important distinction given that felt emotions are not always expressed (Erickson and Ritter 2001; Turner and Stets 2005). Despite the numerous studies confirming negative emotional responses to unfair distributions, the literature focuses primarily on felt emotions (see Cropanzano et al. 2011; Hegtvedt and Parris 2014). Failure to address emotional expressions may stem from a lack of explicit consideration of the social context in which the distributions occur (except Kahn et al. 1982; see work on observers’ responses by Skarlicki and Kulik 2005).

We propose that legitimacy of an individual in the authority position (i.e., strongly vs. weakly supported) may dampen negative emotional responses to perceived injustice resulting from under-reward. In effect, legitimacy acts as an opposing force to the emotions engendered by injustice. We extend the work on legitimacy that largely focuses on behavioral challenges to an authority’s actions (see Zelditch 2001, 2006). We consider the nature of affective responses to the actions of an authority figure who is strongly or weakly supported. We argue that legitimacy processes, under certain conditions, decrease the intensity of negative emotions that could lead to social action. Individuals, though unfairly treated, often accept the status quo because they believe that others around them condone it. Thus, legitimacy processes may act as a control mechanism by reducing the intensity of emotional responses to unfair distributions and in turn help to sustain stratified orders (Hegtvedt and Johnson 2000).

Specifically, we examine the effects of perceived distributive injustice and the legitimacy of the authority on self-reported felt emotions and intended displays of emotions toward the authority, the superordinate authority, and coworkers. The authority (group coordinator in our study) is the person who occupies an immediate authority position and is responsible for an inequitable reward distribution among group members. The superordinate authority is the experimenter, and coworkers are fellow group members.

We focus on a situation in which an individual in the authority position, who is weakly or strongly legitimated in her position, distributes rewards unfairly to task group members. We manipulate two collective sources of legitimacy: (1) authorization of the authority by a superordinate authority in the organizational structure in which the authority and coworkers are embedded and (2) endorsement of the authority by fellow coworkers. We expect that sources of legitimacy similarly affect felt emotions but differentially impact intentional displays to the various targets. In addition, we consider the effect of procedural fairness—the way the authority goes about deciding how to distribute rewards—given Tyler’s (2001, 2006) presumption that procedural justice is a means by which authorities achieve legitimacy.

In the following, we describe emotional responses to injustice and then argue why legitimacy may affect those responses as well but in an opposing direction. In so doing, we identify the scope conditions circumscribing the hypothesized effects of authorization, endorsement, and procedural justice on self-reported felt emotions and intentional displays in an unjust situation.

Distributive Injustice, Legitimacy, and Emotions

Emotional Responses to Distributive Injustice

Classic distributive justice models in the social exchange tradition (Adams 1965; Homans 1974; Walster, Walster, and Berscheid 1978) implicitly or explicitly share several assumptions. First, individuals are rational, motivated to maximize their rewards (material and social) and minimize their costs. Second, people accept an equity rule, dictating a distribution of resources or rewards commensurate to group members’ contributions as normative in exchange and work situations (Leventhal, Karuza, and Fry 1980). Third, to assess the fairness of rewards, individuals compare their shares to what is expected according to the distribution rule or to what others get. Personal motivations, beliefs, and interpretation of situational factors further shape the assessments (see Hegtvedt 2006; Jost and Kay 2010; Tyler et al. 1997). Thus, justice evaluations are subjective, reflecting the interests of the perceiver and mediating between actual circumstances and subsequent responses. Fourth, injustice perceptions cause “distress” (Adams 1965), represented in initial emotional responses. Fifth, individuals are motivated to relieve the distress, resulting in further cognitive or behavioral responses, tempered by situational circumstances (e.g., information availability, position of the victim, or presence of others; see Hegtvedt 2006). In keeping with the initial presumption that individuals are rational actors, Adams (1965) notes that people opt for the “least costly” response to injustice.

Scholars (e.g., Jasso 2006; Tyler et al. 1997) theorize that the more unjust individuals perceive a distribution to be, the more likely they are to experience negative emotions. Such a pattern is consistent with general arguments about affect: emotions are often likely to be evoked when people encounter the unexpected, lose or gain power or status, or where an identity is challenged in interaction (Turner and Stets 2005). With regard to injustice assessments, Homans (1974) posits that people disadvantaged by a distribution are likely to feel anger. Much research supports this proposition and also shows that individuals experience resentment and frustration as well (e.g., DeCremer and van Hiel 2010; Hegtvedt 1990; Krehbiel and Cropanzano 2000; see also Cropanzano et al. 2011; Hegtvedt and Parris 2014). Such negative emotions are relatively active and directed externally at those perceived responsible for the injustice. That activity, coupled with the external focus, is likely to shape individual behaviors and potentially mobilize social or collective action (e.g., Jasper 1998; Schieman 2006).²

Our study adopts scope conditions that characterize previous research on justice evaluations. First, we focus on the one-time distribution of group rewards by an individual in an authority position (i.e., group coordinator). As a consequence, group members are likely to focus on short-term, material rewards rather than long-term or social rewards as they make sense out of the situation. Second, to ensure perceived disadvantage, the focal actor is objectively more disadvantaged than the other group members. He or she receives less than others and less than the common justice principle in work situations—equity—would dictate (Leventhal et al. 1980), and thus it is likely that the focal actor subjectively evaluates the distribution as unfair. Third, the authority is a nonrecipient of the group reward, thereby ruling out her explicit self-interest in the distribution. And, fourth, the focal actor has full information on group pay distribution to facilitate assessments of fairness.

Under these conditions, and consistent with the arguments of the classic justice theorists and previous empirical research, we predict:

Hypothesis 1: Perceived injustice of the authority’s pay distribution is positively related to the intensity of the focal actor’s self-reported negative emotions (anger, frustration, and resentment).

Perceived injustice of the pay distribution should also relate positively to the likelihood that these negative emotions will be displayed. The greater the perceived injustice, the stronger the intensity of the negative feelings and the more likely they will be expressed. Felt emotions may lead to action (e.g., Barclay, Skarlicki, and Pugh 2005; Khan, Quratulain, and Cranshaw 2013), which is a means to display feelings, under certain conditions (see Hegtvedt and Parris 2014). Markovsky (1985) illustrates how the disadvantaged express anger by actively complaining (through pressing a button) to the authority. The relationship between experienced and expressed emotions is consistent with Thoits’s (1989) classic approach to defining emotions that connects physiological sensations, coupled with cognitive appraisals and cultural labels, with affective displays. Thus, we argue that perceived injustice will positively impact the intended displays of negative emotions, similar to its impact on self-reported negative emotions. Specifically, we posit:

Hypothesis 2: Perceived injustice of the authority’s pay distribution is positively related to the likelihood of the focal actor’s intended displays of negative emotions toward the superordinate authority, coworkers, and the authority.

Thoits (1989) also notes that affective displays may be free or inhibited. The former suggests a close connection between emotional experiences and expressions, as complementary Hypotheses 1 and 2 suggest. The notion of inhibited, in contrast, implies the possibility of situational constraints. The authority’s perceived legitimacy may act as such a constraint.

Legitimacy Processes and Emotions

Legitimacy theories distinguish private, individual consent from collective processes (Dornbusch and Scott 1975; Weber [1924] 1978; Zelditch and Walker 1984). Private consent refers to propriety or an individual’s personal beliefs about what is desirable, proper, and appropriate. In contrast, the collective process of legitimacy captures people’s beliefs that they are obliged to obey the norms or requests from authorities even in the absence of personal approval of them. The collective process, generally referred to as validity, inspires compliance as a means to avoid formal and informal sanctions from others.

Legitimacy requires cooperation from others, leading actors to feel obliged “to obey” the authority. Two sources of this collective aspect of legitimacy are authorization and endorsement. Authorization refers to the perceived support from superiors or “uppers” in a group hierarchy who control formal sanctions. Endorsement encompasses the perceived support from peers and subordinates who control informal sanctions (Zelditch and Walker 1984).

Given these sources of legitimacy, a condition further defining the scope of our argument is that the situation represents at least three levels of authority: (1) the focal actor (i.e., the one disadvantaged by the distribution) who is embedded in a hierarchical structure, including other group members working on similar tasks; (2) a person in the authority position who allocates the group reward as compensation for work performances (i.e., the group coordinator); and (3) a superordinate authority who oversees group members and the authority. This condition allows for endorsement from peers (i.e., coworkers) and authorization from those “above” the authority (in our case, the experimenter).

Both sources of legitimacy imply active support of the authority in the form of provision of resources from those above or below when needed (Walker 2014). Individuals compare their own evaluations of a situation with what they believe others around them think. If they perceive that others support an authority, they assume that others will act in accord with the authority, which in turn affects their own reactions. Studies (see Walker and Zelditch 1993; Zelditch 2006; Zelditch and Walker 1984) show that the effects of authorization and endorsement are much stronger than those of propriety on suppressing challenges to unfair resource distributions. In addition, legitimacy of an authority by collective sources is stable, providing a basis for giving the authority the benefit of the doubt—leeway within limits—for particular acts, even those perceived to be unfair.

Authorization, then, implies what is appropriate for an authority in the situation. When an authority acts in an unexpected way, the unexpected also stimulates emotions. Yet, a high level of authorization of the individual in the authority position provides the benefit of the doubt with regard to the authority’s behavior (Hegtvedt and Johnson 2000). Strong support from above operates as a buffer for the authority, creating uncertainty in the focal actor’s view of the situation. As a result, a focal actor may grow less certain about the extent of the unfairness of the outcome distribution created by the authority if that authority is highly authorized.

Likewise, endorsement from fellow coworkers also provides a buffer for the authority. Previous research using a vignette study shows that a high level of endorsement of an authority reduces negative emotional reactions of workers to a conflict over what could be construed as an unfair outcome (i.e., refusal to give a pay increase to the focal group of workers, which included the study participant) (Johnson, Ford, and Kaufman 2000). Study participants who believed that fellow coworkers supported the authority reported less intense negative emotions about the unfair outcome compared to those who believed that their coworkers did not endorse the authority. Similar to Johnson et al. (2000), we reason that legitimacy creates uncertainty about people’s own feelings when they compare their beliefs to what they think their peers think. Thus, we argue that uncertainty created by a high level of authorization or endorsement may also create uncertainty in feelings and thus dampen the intensity of emotions resulting from perception of an unfair act:

Hypothesis 3a: The level of authorization of the authority is negatively related to the intensity of the focal actor’s self-reported negative emotions (anger, frustration, and resentment).

Hypothesis 3b: The level of endorsement of the authority is negatively related to the intensity of the focal actor’s self-reported negative emotions (anger, frustration, and resentment).

Hypotheses 3a and 3b suggest that owing to uncertainty in one’s own assessments created by knowing what others think about the individual in the authority position, authorization and endorsement as the sources of legitimacy act as forces suppressing, and possibly alleviating, self-reported felt distress typically experienced in reaction to perceived distributive injustice. A parallel process is likely for the intentional display of negative emotional responses to injustice. We propose, however, that the mechanism stimulating the suppression of emotional displays is not simply uncertainty but also fear of sanctions.

Legitimacy researchers argue that when an authority is highly legitimated, individuals who are affected may consciously or unconsciously consider the social implications of failing to also provide support. When an authority is strongly supported from above, people face consequences that are within the power of the authority to mete out. Also, when an authority is highly endorsed, failing to go along with peers increases the likelihood of facing social ostracism (see Zelditch 2006; Zelditch and Walker 1984). Potential sanctions, in other words, carry social and in some cases material costs (e.g., job loss, lower wages), which individuals may attempt to avoid by complying with an authority’s decision. Anticipation of such sanctions affects what people express in response to the situation. Thus, we argue that intended displays of emotion should be dampened when the authority is highly authorized or endorsed. Our argument, moreover, is consistent with results reported by Lively (2000; see also Lively 2008) showing that paralegals avoided expressing their anger toward attorneys, their legitimated superiors.

The types of legitimacy should also correspond to the target (superordinate authority, authority, or coworkers) of the emotional expression. As authorization rests with support for the work group authority from the superordinate authority, fear of sanctions from that source should bolster the negative effect of a high level of authorization on intended displays to that “upper” authority. Similarly, with strong endorsement rooted in the beliefs of coworkers, fear of informal sanctions enhances the negative impact of endorsement on emotional displays to peers. High levels of authorization and endorsement should also negatively affect intended displays toward the authority. Knowing that the superordinate authority or coworkers support the authority increases difficulty in expressing these emotions toward the authority. Thus:

Hypothesis 4a: The level of authorization of the authority is negatively related to the likelihood of the focal actor’s intended displays of negative emotions toward the superordinate authority.

Hypothesis 4b: The level of endorsement of the authority is negatively related to the likelihood of the focal actor’s intended displays of negative emotions toward the coworkers.

Hypothesis 4c-1: The level of authorization of the authority is negatively related to the likelihood of the focal actor’s intended displays of negative emotions toward the authority.

Hypothesis 4c-2: The level of endorsement of the authority is negatively related to the likelihood of the focal actor’s intended displays of negative emotions toward the authority.

Although our focus is on perceived injustice of the distribution created by a weakly or strongly legitimized authority, we recognize the potential impact of how an authority “acts” in terms of use of fair or unfair procedures in arriving at a reward distribution for group members. Tyler (2001, 2006) theorizes and provides empirical evidence that procedural justice facilitates development of perceptions of an authority’s legitimacy. Such a perspective stresses it is how the authority acts that enhances perceptions of legitimacy, not simply “what others think.” Additionally, much research shows that people use information on the decision-making process to assess outcome fairness—the so-called fair process effect (van den Bos et al. 1997; see review by van den Bos 2005). Fair procedures attenuate the severity of perceived distributive injustice, thereby reducing the likelihood of felt or expressed negative emotions. Thus, we predict:

Hypothesis 5: The authority’s use of fair procedures is negatively related to the intensity of the focal actor’s self-reported negative emotions (anger, frustration, and resentment).

Hypothesis 6: The authority’s use of fair procedures is negatively related to the likelihood of the focal actor’s intended displays of negative emotions toward any of the targets.

We note, however, that van den Bos (2005) emphasizes that the fair process effect is most likely to emerge under conditions of uncertainty characterized by knowing only one’s own outcomes and not those of other group members. In a rare study (given that most procedural justice studies including those focused on the emergence of legitimacy provide only information on the perceiver’s own outcomes), van den Bos et al. (1997) show that procedural justice has far less impact when full distribution information is available. Given that our study likewise offers group members information on the distribution of outcomes, the role of procedural justice in reducing uncertainty and shaping distribution evaluations and emotional responses may be minimized.

Finally, in addition to examining the opposing effects of distributive justice perceptions and sources of legitimacy on self-reported felt emotions and intended displays, we assess whether the former mediates the relationship between legitimacy factors and intended emotional displays toward the three targets. Consistent with Thoits’s (1989) definition of emotions, for emotions to be displayed, they must also be felt. It may be that authorization and endorsement indirectly affect intentional displays by reducing the self-reported negative emotions. Without these sources of legitimacy, felt negative emotions will be stronger and will increase the likelihood of actors displaying these emotions toward the superordinate authority, coworkers, and the authority.

The Experiment

Overview

We created a computerized experimental situation to emphasize performance in a work group. As a member of a four-person work group, each study participant took on an independent task that contributed to the completion of the overall group task. Although participants believed that they were working with others, in reality, each was the focal actor who was disadvantaged by the distribution, creating an unfair situation. In the experiment, we created high and low levels of authorization and endorsement of the individual in the authority position (i.e., group coordinator) and high (fair) and low (unfair) levels of procedural justice used by the authority.

Study Participants

We recruited students in various courses (e.g., sociology, political science, and biology) at a private southeastern university to participate in a series of studies about “work groups.” We told potential participants that they would be compensated $6 to $16 depending on their performance and asked them to indicate on the recruitment form their gender, race, major, availability, and contact information. We used only women to control the gender of participants. Our experiment included 142 female participants, with 12 to 15 subjects per cell of the 2 × 2 × 2 design.

Procedures

An experimenter greeted study participants as they arrived at the social psychology laboratory. In a face-to-face meeting, she explained to the four participants that they would work on a group task, with three people completing individual tasks and one, chosen by the experimenter based on previous work experience and anticipated career, acting as the group coordinator who would distribute rewards. She stressed that more detailed instructions would be delivered via the computer. After providing the overview, the experimenter reviewed the consent form and answered questions.

Once study participants were in their randomly selected rooms, the computer displayed further instructions and provided the means for delivering information to manipulate experimental factors. When it was necessary to make it appear as if the computer was busy with particular activities, we built in appropriate “pauses.” We provided a sheet for participants to write down information as they received it.

To lay the groundwork for the legitimacy manipulations, participants completed a background experience form in the privacy of their computer rooms. They provided information on: (1) up to three previous jobs and/or nonpaid (volunteer) work activities in which they held leadership or highly responsible positions, including position title, pay (if applicable), and duties; (2) their intention (if any) for a career in health or medicine; and (3) the areas of health that are of most interest to them (e.g., medical ethics, genetic engineering, cancer, HIV/AIDS, adolescent sex, managed health care). When the experimenter collected the forms, she told study participants that the information would be used to select a group member to act as the group coordinator and be used by that person to make decisions about the tasks that each person would perform. The participants would soon learn that the task was related to the health/medicine field.

After collecting the background forms, the computer indicated that the study participant was group member “B” and that others were “A, C, and D.” The computer then paused while the experimenter ostensibly reviewed the forms and selected a group coordinator. As described more fully in the following, the experimenter announced via the computer that group member “A” was selected to act as the group coordinator who would review the performance of other group members and distribute pay. The introduction of “A” as the group coordinator ensured the creation of a hierarchical group, which helps to instantiate one of our scope conditions. In addition, we stressed that the group coordinator would be paid separately and thus was a “nonrecipient” of the group pay, as required by another scope condition.

While the group coordinator ostensibly received instructions from the experimenter on her tasks (evaluating performances, devising a pay distribution, sending messages), the computer displayed information to other group members about their tasks. Instructions indicated that each group member would be assigned to read a different one-page article on stem cell research and analyze it in terms of its general position, its reference to factual information, and the biases inherent in its specific arguments.³ Additional instructions indicated how to detect factual and biased statements. We provided a hard copy of the coding sheet so that group members could complete the task first on paper while the article appeared on the screen and then enter their analysis into an electronic coding sheet to be sent to the group coordinator. Group members had 15 minutes to complete this task; the computer provided information about the elapse of time so that everyone had sufficient time to complete the task. In reality, all group members read the same stem cell article that had been “fictionalized” to simplify the coding task.

At the conclusion of the coding task, the group coordinator purportedly began the process by which she would make pay decisions. The group coordinator appeared to “interact” with other group members, soliciting their opinions as well as providing a rationale for her decision. Such “interaction” was preprogrammed and thus depended on each participant’s randomly assigned experimental condition. Although group members were told that their performance would be evaluated in terms of the quality of their analysis, the number of details reported, and the clarity of explanations, in reality, the group coordinator’s evaluations and pay distribution were preprogrammed. The coordinator’s pay distribution was as follows: B $7, C $14, and D $9. As the lowest paid group member, “B” was disadvantaged by the distribution, as indicated by one of our scope conditions. After displaying the pay distribution, study participants responded to questionnaire items, received debriefing, and were paid $15 as compensation for time spent in the one-hour experimental session.

Independent Variables

Our manipulations of legitimacy and procedural justice were created through information represented as instructions or messages from the experimenter or the group coordinator. This information appeared on subjects’ computer screens at appropriate times. We explicitly told study participants to write down some information on the sheet provided by their computers.

As noted previously, subjects completed a background experience form as the basis for manipulating the authorization of the group coordinator. To convey that the group coordinator had high authorization, study participants learned from the experimenter that: (1) based on her experiences and responsibilities in paid and nonpaid jobs, she selected Group Member A to act as the group coordinator and (2) she thought A was very well-suited for the job. In addition, while the group members are doing the task, the group coordinator sent them a message stating “I appreciate the support I have had from the experimenter.” To manipulate low authorization, the experimenter told participants “None of the group members had the type of experience I was looking for. So, I randomly picked one of you to be the coordinator. Group Member A will act as the group coordinator.” And, again while the group members are doing the task, the coordinator comments, “I know that the experimenter picked me by chance, without strong support.”

We built in the possibility for study participants to respond to the experimenter’s selection of a group coordinator to manipulate endorsement. All subjects were asked: “What concerns, if any, do you have about Group Member A acting as the group coordinator?” They could type in any concerns and send them to the experimenter. They were also asked: “How much do you support Group Member A acting as the group coordinator?” (Responses were recorded on a scale ranging from 1 = no support to 9 = a great deal of support). A concluding question was: “Would you nonetheless be willing to go along with Group Member A (yes/no)?” Participants believed that their “coworkers” also responded to these questions.

For high endorsement, the experimenter told the participants, “Based on the responses to the questions about A as the group coordinator, Group Member C strongly supports A and Group Member D strongly supports A. Neither C nor D expressed any concerns.” In addition, while the group members were doing the task and in conjunction with the authorization manipulations, the group coordinator also noted her appreciation for the support she received “generally from the group members.” To manipulate low endorsement, we provided feedback from the experimenter indicating that “Group Member C weakly supports A and Group Member D weakly supports A because they both have some concerns. But they indicated that they would go along with the selection.” And later, the group coordinator sends a message saying, “I know some group members had some concerns and did not really support me as the group coordinator.” Similar to authorization, in both high and low endorsement manipulations, participants did not receive specific numbers from the experimenter regarding support from their fellow “group members”; they were only given the described general responses.⁴

To manipulate procedural justice, the preprogrammed messages ostensibly from the group coordinator contained intentional misspellings to mimic what is likely to occur in everyday emails between students. We enlisted undergraduate lab assistants to help us to phrase messages from the group coordinator in a manner that would “sound like” a fellow undergraduate.

For high procedural justice, we designed the group coordinator’s messages to show respect, interest in the members’ ideas as a means of representing “voice,” and lack of bias or neutrality (see Leventhal et al. 1980; Tyler and Lind 1992). For example, for the first message, the group coordinator said that she tried to make sure each of the members got an article on a topic that they mentioned on their background forms and noted regret when she could not do so (demonstrating respect). Once the members coded the article, the group coordinator asked for their ideas about what she should look for when she reviewed the articles (giving voice). And after reviewing their work, she mentioned that she found some articles more interesting than others, but that did not affect how she judged their work (instantiating lack of bias).

In contrast, for low procedural justice, we adjusted the messages so that they would sound disrespectful and dismissive, uninterested in the ideas of others, and biased. For example, for the first message, the group coordinator indicated that she did not use background information except in one case. And while study participants could give ideas about how to review the coding, the group coordinator said that she had her own ideas and plainly ignored ideas that were sent. Finally, to represent bias in her procedures, the coordinator stated that her favorite article was the one reviewed by Group Member C (who received the highest pay).

After learning of the pay distribution, to capture perceptions of distributive injustice, we asked study participants: “How unfair or fair (and how unjust or just) do you perceive the group coordinator’s final pay distribution to be?” The response scale ranges from 1 = very unfair or very unjust, to 9 = very fair or very just. The high bivariate correlation (.93) between these two items allows us to combine them into an additive, standardized scale that was reverse coded so that low numbers suggest that study participants view the situation as fair and high numbers represent unfairness.

Measures of Dependent Variables

After learning about the pay distribution, we tapped into self-reported emotions by asking, “How much did you feel each of the following when you saw the pay distribution?” The 9-point response scale indicated that the respondents felt each emotion not at all (1) to a great deal (9). As previously noted, disadvantageous injustice is likely to stimulate relatively active negative emotions. Consistent with previous research, we focus on self-reported negative emotions of anger, frustration, and resentment.⁵ Like other studies (e.g., Hegtvedt and Killian 1999; Murphy 2009; Stecher and Rosse 2005), we develop a composite measure. Principle component factor analysis indicates that anger, frustration, and resentment represent a single factor with an alpha scale reliability of .86. Thus, we created an additive, standardized scale ranging from 1, indicating the lack of the experience of negative emotion, to 9, representing the experience of active negative emotions.

To measure intended displays of negative emotions toward the experimenter, coworkers, and group coordinator, we asked three questions: “If you were to interact with coworkers C and D/the group coordinator/the experimenter again, how likely would you be to express each of the following emotions regarding your reactions to the pay distribution?” The 9-point response scale ranged from 1 = not at all likely to 9 = very likely. Factor analysis again produced a factor including anger, frustration, and resentment for intended displays toward each of the targeted others. The alpha reliabilities for the additive, standardized scales regarding intended displays of negative emotions were as follows: toward the experimenter = .92, toward coworkers = .87, and toward the group coordinator = .88.

We used questionnaire items representing manipulation checks to reflect our emphasis on the legitimacy of the group coordinator. We asked separate questions regarding how much support the experimenter and group members C and D provided for the selection of the group coordinator. Low numbers on the nine-point scale indicated lack of support while high numbers indicated support. Results reveal a main effect for authorization on analysis of the perceived support of the experimenter, F(1, 131) = 9.00, p < .01 (M = 5.10 in low conditions and M = 6.16 in high conditions). A main effect for endorsement also emerges in analysis of the items indicating the averaged support of group members C and D, F(1, 131) = 452.60, p < .001 (M = 2.97 in low endorsement condition and M = 8.29 in the high endorsement condition).

The manipulation check of procedural justice involved specific questions about what the group coordinator actually did in making the pay distribution. In the low procedural justice condition, respondents viewed the procedures as more arbitrary (M = 6.87 compared to 5.49 in the high procedural condition) and saw the coordinator as less interested in the opinions of group members (M = 4.04 vs. 5.03) and less consistent in how she applied criteria (M = 4.24 vs. 4.71) (but not measurably more biased, M = 5.08 vs. 4.79). Analysis of these three items (omitting the nonsignificant “biased” item) combined into an additive, standardized scale (with “arbitrary” reverse coded) produces a main effect for procedural justice, F(1, 131) = 11.7, p = .001 (M = 3.85 in the low procedural justice condition and M = 4.78 in the high procedural condition).

Results

Table 1 reports the means and standard deviations for self-reported negative emotions and intended displays of these emotions toward the superordinate authority (i.e., experimenter), coworkers, and the authority (i.e., group coordinator). Not surprisingly, the mean for self-reported negative emotions (M = 4.03) is significantly higher than the means for their intended displays toward the superordinate authority (M = 2.76; t = 7.65, p < .001), the coworkers (M = 3.12; t = 6.32, p < .001), and the authority (M = 3.21; t = 5.42, p < .001). Table 1 also provides the bivariate correlations.

Table 1.

Means (Standard Deviations)^a and Correlations of Variables

	Negative emotions	Intended displays to superordinate authority	Intended displays to coworkers	Intended displays to authority	Perceived distributive injustice	Authorization	Endorsement	Procedural justice
Negative emotions	4.03 (2.23)
Intended displays to superordinate authority	.59***	2.76 (2.17)
Intended displays to coworkers	.67***	.75***	3.12 (2.14)
Intended displays to authority	.66***	.79***	.81***	3.21 (2.21)
Perceived distributive injustice	.55***	.48***	.50***	.55***	6.19 (1.87)
Authorization	−.17*	−.12	−.14	−.13	−.06	.50 (.50)
Endorsement	−.26**	−.16^†	−.21*	−.26**	−.09	−.06	.51 (.50)
Procedural justice	−.06	−.07	−.01	.06	−.05	−.06	−.01	.48 (.50)

Valid N (listwise) = 136; means (in bold) and standard deviations are presented on the diagonal.

†

p ≤ .10 *p ≤ .05. **p ≤ .01. ***p ≤ .001 (two-tailed tests).

We used a series of ordinary least squares regressions to test our hypotheses, as our dependent variables reflect interval data and are normally distributed (Miles and Shevlin 2001). We report standardized coefficients. Model 1 in Table 2 assesses the effects of perceptions of distributive injustice, authorization, endorsement, and procedural justice on self-reported negative emotions and the intended displays of these emotions toward the superordinate authority, coworkers, and the authority. Model 2 in Table 2 assesses the mediation effect of self-reported negative emotions between sources of legitimacy and the intended displays toward the three targets. The highest variance inflation factor (VIF) in our models is 1.56, indicating that multicollinearity was not present (O’Brien 2007).

Table 2.

Ordinary Least Squares Standardized Regression Coefficients (Standard Errors) for Perceived Distributive Injustice, Authorization, Endorsement, and Procedural Justice on Active Emotional Responses

Independent variable	Active negative emotions	Intended displays to superordinate authority		Intended displays to coworkers		Intended displays to authority
	Model 1	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2
Constant	1.16^† (.63)	.010(.68)	−.51(.62)	.32(.65)	−.33(.57)	−.11(.64)	−.68(.57)
Perceived distributive injustice	.52*** (.08)	.46*** (.09)	.23** (.10)	.47*** (.09)	.20** (.09)	.52*** (.08)	.28*** (.09)
Authorization	−.16* (.31)	−.10(.33)	−.03(.30)	−.11(.32)	−.01(.28)	−.10(.31)	−.02(.28)
Endorsement	−.21** (.31)	−.12(.33)	−.03(.31)	−.17* (.32)	−.06(.28)	−.21** (.31)	−.11^† (.28)
Procedural justice	−.04(.31)	−.06(.33)	−.05(.30)	.00(.32)	.03(.28)	.07(.31)	.09(.27)
Negative Emotions	—	—	.45*** (.08)	—	.54*** (.08)	—	.49*** (.08)
N	139	137	137	135	135	137	137
R²	.37***	.26***	.39***	.29***	.47***	.36***	.51***

†

p ≤ .10 *p ≤ .05. **p ≤ .01. ***p ≤ .001 (two-tailed tests).

Self-Reported Negative Emotions

Our hypotheses predict that perceived distributive injustice will be positively related to self-reported negative emotions while the collective sources of legitimacy will be negatively related to these emotions. Results confirm these opposing forces.

As seen in Table 2, perceived distributive injustice of the authority’s distribution is positively related to the intensity of self-reported negative emotions (β = .52, p < .001), supporting Hypothesis 1. On the other hand, high levels of authorization and endorsement of the authority dampen self-reported negative emotions, (β = −.16, p < .05; β = −.21, p < .01) relative to low levels of authorization and endorsement, supporting Hypotheses 3a and 3b.⁶ Hypothesis 5 predicting that use of fair procedures by the authority will be negatively related to the intensity of self-reported negative emotions is not supported.⁷

Intended Displays of Negative Emotions

Consistent with results for self-reported emotions, perceived distributive injustice also positively and significantly affects the intended displays of negative emotions toward all three targets (Model 1, Table 2). Supporting Hypothesis 2, the more individuals perceive their outcome as unfair, the more likely they are to indicate their intention to display negative emotions toward the superordinate authority (β = .46, p < .001), coworkers (β = .47, p < .001), and the authority (β = .52, p < .001).

No support emerges for Hypothesis 4a, predicting a highly authorized authority will decrease the likelihood of intended displays of negative emotions toward the superordinate authority. In contrast, Hypothesis 4b regarding endorsement is confirmed: when people believe that their coworkers strongly support the authority, they are less likely to indicate that they would express negative emotions toward their coworkers (β = −.17, p < .05).

Hypotheses 4c-1 and 4c-2 predict that high levels of authorization and endorsement, respectively, will be negatively related to the likelihood of intended displays of negative emotions toward the authority. Hypothesis 4c-1 is not supported; authorization does not significantly affect intended displays of emotions. In support of Hypothesis 4c-2, a high level of endorsement (β = −.21, p < .01) is related to intended displays toward the authority. When coworkers strongly support the authority, actors are less likely to indicate intended displays of negative emotions toward the authority than when coworkers weakly support the authority.⁸

Hypothesis 6 predicts that the authority’s use of fair procedures is negatively related to the intended displays of negative emotions toward any of the targets. Yet, no support emerged for this hypothesis.⁹

Test of Mediation Effects

In addition to examining the opposing effects of perceptions of distributive injustice and sources of legitimacy on self-reported felt emotions and intended displays, we examined two types of potential mediation effects. First, we included self-reported emotions as a mediator of the relationship between legitimacy factors and emotional displays because legitimacy may indirectly affect the intended displays through self-reported negative emotions. Model 2 (Table 2) shows that the impact of self-reported negative emotions is positive on displays of emotions toward the superordinate authority (β = .45, p < .001), coworkers (β = .54, p < .001), and the authority (β = .49, p < .001), as expected. Including self-reported negative emotions also reduces somewhat the significant effect of perceived distributive injustice on intended displays toward the superordinate authority (β = .23, p < .01) and toward coworkers (β = .20, p < .01). In regard to legitimacy, the previously significant effects of endorsement grow insignificant on displays toward coworkers and marginally significant on displays toward the authority.¹⁰ These results show that endorsement indirectly affects intended displays through self-reported negative emotions.

Second, we recognize that justice approaches generally suggest that perceptions of distributive justice should mediate between situational factors and emotional responses to injustice (see Hegtvedt 2006). In our situation, however, we do not expect that these perceptions will mediate the effects of legitimacy on the emotion variables. Perceptions of distributive injustice in our study concern the specific act of the authority—an unjust outcome (i.e., pay) relative to other group members. In contrast, the collective sources of legitimacy (the situational variables of authorization and endorsement) concern the authority and not the particular act of the authority. The social object of legitimation, then, is the authority and not the authority’s act.

To check our expectation, we examined the effects of the levels of authorization and endorsement on perceptions of distributive injustice. As expected, authorization and endorsement did not significantly affect perceptions of distributive injustice regarding the outcome. As such, distributive injustice perceptions of the act fail to mediate the effects of legitimacy of the authority on emotional responses.

Discussion

In our study, self-reported negative emotions by actors in an unjust situation are significantly affected by actors’ perceptions of distributive injustice and the collective sources of legitimacy of the individual in the authority position in charge of distributing a resource to the group members. Perceptions of injustice regarding the outcome heighten self-reported negative emotions. Yet, high levels of authorization and endorsement of the authority dampen them, demonstrating the opposing forces of injustice perceptions and legitimacy on felt emotions.

In addition, the more unjust an actor perceives an outcome to be, the more likely he or she intends to display these emotions to all three targets (the superordinate authority, coworkers, and the authority). Strong endorsement of the authority, in contrast, attenuates intended displays to coworkers. Yet, as our mediation analysis indicates, endorsement effects are no longer significant when self-reported felt emotions are included in the models, indicating that endorsement indirectly affects intended displays through self-reported negative emotions. Strong authorization and endorsement negatively impact self-reported negative emotions, indicating that both sources of legitimacy have indirect effects on the intended displays through self-reported negative emotions. Thus, without the experience of high authorization and endorsement, actors would feel even stronger negative emotions, and these negative emotions increase the likelihood that they would display them toward superordinates, coworkers, and those in authority.

Use of fair or unfair procedures in the decision-making process did not significantly influence any of the emotional responses. In our study, group members had full distribution information, which undoubtedly reduced uncertainty and the impact of procedural justice (see van den Bos 2005). Nonetheless, the lack of a procedural justice effect may also stem from the possibility that our procedural justice manipulation was weaker than that of endorsement or authorization. Future research should discover under what conditions procedural justice influences emotional responses, similar to legitimacy.

Overall, our findings suggest two key implications about the relationship between legitimacy, justice, and emotions in work settings. First, what people report feeling and what they intend to display are shaped by what they think others think, even in seemingly unfair situations. This is so even under weak group conditions, with little actual threat of sanctions, negligible group history, and an absence of any strong identification with the group. Future research might examine variation in these group conditions. For example, how does variation in the strength of group identity affect the impact of the legitimacy of the authority on emotional responses? It may be that endorsement has a stronger effect on felt emotions when group identity is strong (and thus threat of sanctions is also strong) than when it is weak, whereas authorization may have a stronger effect when it is weak than strong. Finally, we cannot definitely determine whether uncertainty or threat of sanctions, or both, act as mechanisms to explain the effect of legitimacy on emotional responses. Given that the effects of endorsement are reduced on intentional displays when we include self-reported emotions, threat of sanctions may not be strong enough in our study to act as a mechanism to inhibit intentional displays. Future work is needed to tease out these mechanisms.

Second, our pattern of findings complements the literature on reciprocal emotional management and the role of coworkers in how emotions and their management operate in the workplace. Recent literature suggests that disadvantaged actors sometimes manage their negative emotions by venting with fellow coworkers as a result of perceived distress at work and in workplace interactions (e.g., Lively 2000, 2008). Although represented as a means of social support, this process also serves as a mechanism to maintain the existing stratified order (see also Lively and Weed 2014). Similarly, Schwalbe and Shay (2014) argue that the “rules” actors often use in interaction to minimize the risk to express emotions may inadvertently reproduce the social order. They note, “One price we pay for emotional safety in face-to-face interaction is thus the reproduction of larger-scale inequality” (Schwalbe and Shay 2014:160). In our study, the force of legitimacy of the authority represents another mechanism by which emotions affect stratified orders. Legitimacy processes affect self-reported active negative emotions, which in turn affect intentional expressions of these emotions toward the authority and coworkers. If workers are less likely to express their anger to unjust situations, they stymie potential challenges to the status quo. Similar to reciprocal emotion management, legitimacy processes play a role in unintentionally maintaining inequality by dampening emotional responses that could motivate social action (Walker 2014; Zelditch and Walker 1984).

Although our pattern of findings is robust, our study involves only female participants. Future legitimacy and emotions studies should involve male participants as well. Given previous research on justice and on legitimacy and emotions, we anticipate that the effects of legitimacy would be similar, but this is an empirical question. An additional limitation is our use of a one-shot assessment of justice and reactions. An interesting pathway for future research is to capture ongoing dynamics where the authority continues to distribute resources unfairly to see if the attenuating effect of legitimacy processes on negative emotions dissipates over time. Walker’s (2014) example of a judge who issues unfair rulings over time and ultimately loses legitimacy illustrates this important concern. This possibility suggests a reciprocal relationship and calls for an examination of how justice and legitimacy unfold over time. The point at which legitimacy ceases to be an opposing force attenuating felt and expressed emotional responses to injustice awaits empirical scrutiny.

Footnotes

Acknowledgements

The authors gratefully acknowledge the support of the National Science Foundation, the helpful comments of the SPQ editors and anonymous reviewers, and the research assistance of Leslie Brody, Lynn Chen, Miek Debeukelaer, Angela Choi, Maria Town, Kyle Valenti, Krysia Waldron, and Monica Yau.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article. This research was supported by a grant from the National Science Foundation (#SES-0136747) to the first two authors.

Supplemental Material

Additional supporting information may be found at .

1

We adopt a frequently used definition of emotions as “relatively short term evaluative response[s] essentially positive or negative in nature involving distinct somatic (and often cognitive) components” (:47).

2

We focus on externally directed, active negative emotions (e.g., anger, resentment) that are presumed to stimulate behavioral responses designed to redress perceived disadvantageous injustice. More passive and internally focused negative emotions (e.g., sadness, depression) are less likely to mediate between injustice perceptions and behavioral responses than is anger (Barclay, Skarlicki, and Pugh 2005; Khan, Quratulain, and Crawshaw 2013).

3

We chose a topic for the coding task based on the responses of 40 students in an undergraduate social psychology class. Several semesters prior to the experiment, we asked students to rate various topics of interest to them. Stem cell research received the highest rating, owing to the “hotness” of the topic at the time.

4

Only two participants reported that they would not go along with the coordinator. We ran all analyses without these two participants and found similar results, with no substantive differences with those reported in the results section.

5

We also measured more passive, internally directed negative emotions of depression, worry, and helplessness. Factor analyses produced a single factor regarding the self-reported passive negative emotions. The additive, standardized scale for the self-reported passive negative emotions carries a reliability of .806 and a mean of 2.92, lower than that for active negative emotions. Because the meaning of these passive emotions is distinct from the relatively active negative emotions and their role in responses to injustice, it is conceptually misleading to combine both sets of negative emotions.

6

We also asked participants: “Regardless of the pay distribution, how well do you think you performed on the task?” (1= not very to 9 = very well). In additional analyses, we ran our models controlling for the focal actor’s perception of her own performance. Perceived performance did not significantly affect self-reported active negative emotions directly, and the pattern of effects of legitimacy and perceived distributive injustice remained as those reported in .

7

Similar to the active negative emotions, distributive injustice positively affects self-reported passive negative emotions. Endorsement has a significant negative effect (parallel Model 1). Unlike the results involving active emotions, authorization is not significant for the passive emotions (see Table 3 in the online supplemental materials available at ).

8

We ran additional analyses controlling for the perception of the focal actor’s own performance on the task in the intended displays models of active negative emotions. Perceived performance did not significantly affect the intended display of active negative emotions toward any of the three targets, and the pattern of effects for legitimacy and perceived distributive injustice remained the same as those reported in .

9

Factor analyses also produced a single factor for the intended displays of the passive negative emotions to each target. The additive, standardized scales for intended display of these emotions were reliable: toward the superordinate authority = .78, toward coworkers = .81, and toward the authority = .77. The means for each scale were also lower than those for active negative emotions toward the superordinate authority (M = 2.24), the coworkers (M = 2.38), and the authority (M = 2.33). The regression analysis (parallel Model 1) produced some similarities and some differences compared to the analysis with active negative emotions. Like with active negative emotions, distributive injustice has a positive significant effect on intended displays of passive negative emotions to all of the targets. Endorsement exerts a significant negative effect; the emergence of an effect on the intended display to the superordinate authority is distinct from the findings regarding active negative displays (see Table 3 in the online supplemental materials available at ).

10

The pattern for mediation of self-reported passive negative emotions on intended displays is similar to that of active negative emotions except that when self-reported emotions are included (parallel Model 2), the effects on intended display to the authority remain significant at the .05 level (see Table 3 in the online supplemental materials available at ).

Bios

Cathryn Johnson is a professor of sociology and a senior associate dean of the Laney Graduate School at Emory University. Her work focuses on legitimacy processes within groups and organizations, identity processes, and emotions. She also examines the impact of legitimacy of institutional sustainability policies and programming on environmental justice perceptions, environmental identities, and behaviors. She is a previous coeditor of Social Psychology Quarterly.

Karen A. Hegtvedt is a professor of sociology at Emory University. Her work focuses on perceptions of and responses to injustice. Recent empirical and theoretical work has examined the interrelationship of collective sources of legitimacy, justice, and trust. In addition, she has examined antecedents to perceptions of environmental injustice and to environmentally responsible behavior in a series of papers. She is a previous coeditor of Social Psychology Quarterly.

Nikki Khanna is an associate professor of sociology at the University of Vermont. Her work looks at identity processes (with a focus on racial identity among multiracial Americans) and more recently, the role of race in adoption. She is the author of Biracial in America, and her work has been published in outlets such as Social Psychology Quarterly, Ethnic and Racial Studies, Sociological Perspectives, The Sociological Quarterly, and Teaching Sociology.

Heather L. Scheuerman is an assistant professor in the Department of Justice Studies at James Madison University. Her research centers on how social psychological concepts and processes structure behavior, including crime. She focuses on issues of justice broadly and in terms of how conceptualizations of justice and injustice affect behavioral outcomes.

References

Adams

John S.

1965. “Inequity in Social Exchange.” Advances in Experimental Social Psychology 2:267–99.

Barclay

Laurie J.

Skarlicki

Daniel P.

Pugh

S. Douglas

. 2005. “Exploring the Role of Emotions in Injustice Perceptions and Retaliation.” Journal of Applied Psychology 90:629–43.

Clay-Warner

Jody

. 2006. “Procedural Justice and Legitimacy: Predicting Negative Emotional Reactions to Workplace Injustice.” Advances in Group Processes 23:207–27.

Cropanzano

Russell

Stein

Jordan H.

Nadisic

Thierry

. 2011. Justice and the Experience of Emotion. New York: Routledge.

De Cremer

David

van Hiel

Alain

. 2010, “‘Becoming Angry When Another Is Treated Fairly’: On Understanding When Own and Other’s Fair Treatment Influences Negative Reactions.” British Journal of Management 21:280–98.

Dornbusch

Sanford M.

Scott

W. Richard

. 1975. Evaluation and the Exercise of Authority. San Francisco, CA: Jossey-Bass.

Erickson

Rebecca J.

Ritter

Christian

. 2001. “Emotional Labor, Burnout, and Inauthenticity: Does Gender Matter?” Social Psychology Quarterly 64:146–63.

Fox

Suzy

Spector

Paul E.

Miles

Don

. 2001. “Counterproductive Work Behavior (CWB) in Response to Job Stressors and Organizational Justice: Some Mediator and Moderator Tests for Autonomy and Emotions.” Journal of Vocational Behavior 59:291–309.

Foy

Steven

Freeland

Robert

Miles

Andrew

Rogers

Kimberly B.

Smith-Lovin

Lynn

. 2014. “Emotions and Affect as Source, Outcome and Resistance to Inequality.” Pp. 295–324 in Handbook of the Social Psychology of Inequality, edited by McLeod

J. D.

Lawler

E. J.

Schwalbe

New York: Springer.

10.

Hegtvedt

Karen A.

1990. “The Effects of Relationship Structure on Emotional Responses to Inequity.” Social Psychology Quarterly 53:214-28.

11.

Hegtvedt

Karen A.

2006. “Justice Frameworks.” Pp. 46–69 in Contemporary Social Psychological Theories, edited by Burke

P. J.

Stanford, CA: Stanford University Press.

12.

Hegtvedt

Karen A.

Johnson

Cathryn

. 2000. “Justice Beyond the Individual: A Future with Legitimation.” Social Psychology Quarterly 63:298–311.

13.

Hegtvedt

Karen

Killian

Caitlin

. 1999. “Fairness and Emotions: Reactions to the Process and Outcomes of Negotiations.” Social Forces 78:269–303.

14.

Hegtvedt

Karen A.

Parris

Christie

. 2014. “Emotions in Justice Processes.” Pp. 103–26 in Handbook of Sociology of Emotions, edited by Stets

Turner

New York: Springer.

15.

Hochschild

Arlie Russell

. 1983. The Managed Heart. Berkeley: University of California Press.

16.

Homans

George C.

1974. Social Behavior: Its Elementary Forms. New York: Harcourt Brace.

17.

Jasper

James M.

1998. “The Emotions of Protest: Affective and Reactive Emotions in and around Social Movements.” Sociological Forum 13:397–424.

18.

Jasso

Guillermina

. 2006. “Emotion in Justice Processes.” Pp. 321–46 in Handbook of the Sociology of Emotions, edited by Stets

J. E.

Turner

J. H.

New York: Springer.

19.

Johnson

Cathryn

Dowd

Timothy J.

Ridgeway

Cecilia L.

2006. “Legitimacy as a Social Process.” Annual Review of Sociology 32:53–78.

20.

Johnson

Cathryn

Ford

Rebecca

Kaufman

Joanne M.

2000. “Emotional Reactions to Conflict: Do Dependence and Legitimacy Matter?” Social Forces 79:107–37.

21.

Jost

John T.

Kay

Aaron C.

2010. “Social Justice: History, Theory, and Research.” Pp. 1122–65 in Handbook of Social Psychology. 5th ed., edited by Fiske

S. T.

Gilbert

D.T.

Lindzey

Hoboken, NJ: Wiley.

22.

Kahn

Arnold

Nelson

Robin E.

Gaeddert

William P.

Hearn

June L.

1982. “The Justice Process: Deciding upon Equity or Equality. Social Psychology Quarterly 45:3–8.

23.

Kemper

Theodore

. 1978. A Social Interactional Theory of Emotion. New York: Wiley.

24.

Khan

Abdul Karim

Quratulain

Samina

Crawshaw

Jonathan R.

2013. “The Mediating Role of Discrete Emotions in the Relationship between Injustice and Counterproductive Work Behaviors: A Study in Pakistan.” Journal of Business and Psychology 28:49–61.

25.

Krehbiel

Patricia J.

Cropanzano

Russell

. 2000. “Procedural Justice, Outcome Favorability, and Emotion.” Social Justice Research 13:339–60.

26.

Leventhal

Gerald S.

Karuza

Jurgis

Jr. Fry

William Rick

. 1980. “Beyond Fairness: A Theory of Allocation Preferences.” Pp. 167–218 in Justice and Social Interaction, edited by Mikula

New York: Springer-Verlag.

27.

Lively

Kathryn J.

2000. “Reciprocal Emotion Management: Working Together to Maintain Stratification in Private Law Firms.” Work and Occupations 27:32–63.

28.

Lively

Kathryn J.

2008. “Status and Emotional Expression: The Influence of ‘Others’ in Hierarchical Work Settings.” Pp. 287–300 in Social Structure and Emotion, edited by Clay-Warner

Robinson

D. T.

Burlington, MA: Elsevier.

29.

Lively

Kathryn J.

Powell

Brian

. 2006. “Emotional Expression at Work and at Home: Domain, Status, or Individual Characteristics?” Social Psychology Quarterly 69:17–38.

30.

Lively

Kathryn J.

Weed

Emi A.

2014. “Emotion Management: Sociological Insight into What, How, Why, and to What End?” Emotion Review 6:202–207.

31.

Markovsky

Barry

. 1985. “Toward a Multilevel Distributive Justice Theory.” American Sociological Review 50:822–39.

32.

Miles

Jeremy

Shevlin

Mark

. 2001. Applying Regression & Correlation: A Guide for Students and Researchers. Thousand Oaks, CA: Sage.

33.

Murphy

Kristina

. 2009. “Procedural Justice and Affect Intensity: Understanding Reactions to Regulatory Authorities.” Social Justice Research 22:1–30.

34.

O’Brien

Robert M.

2007. “A Caution Regarding Rules of Thumb for Variance Inflation Factors.” Quality and Quantity 41:673–90.

35.

Schieman

Scott

. 2006. “Anger.” Pp. 493–515 in Handbook of the Sociology of Emotions, edited by Stets

J. E.

Turner

J. H.

New York: Springer.

36.

Schwalbe

Michael

Shay

Heather

. 2014. “Dramaturgy and Dominance.” Pp. 155–81 in Handbook of the Social Psychology of Inequality, edited by McLeod

J. D.

Lawler

E. J.

Schwalbe

New York: Springer.

37.

Skarlicki

Daniel P.

Kulik

Carol T.

2005. “Third-Party Reactions to Employee (Mis)treatment: A Justice Perspective.” Research in Organizational Behavior 26:183–229.

38.

Stecher

Mary D.

Rosse

Joseph G.

2005. “The Distributive Side of Interactional Justice: The Effects of Interpersonal Treatment on Emotional Arousal.” Journal of Managerial Issues 17:229–46.

39.

Thoits

Peggy A.

1989. “The Sociology of Emotions.” Annual Review of Sociology 61:837–57.

40.

Turner

Jonathan H.

Stets

Jan E.

2005. The Sociology of Emotions. New York: Cambridge University Press.

41.

Tyler

Tom R.

2001. “A Psychological Perspective on the Legitimacy of Institutions and Authorities.” Pp. 416–36 in The Psychology of Legitimacy, edited by Jost

J. T.

Major

New York: Cambridge University Press.

42.

Tyler

Tom R.

2006. “Psychological Perspectives on Legitimacy and Legitimation.” Annual Review of Psychology 57:375–400.

43.

Tyler

Tom R.

Boeckmann

Robert J.

Smith

Heather J.

Huo

Yuen J.

1997. Social Justice in a Diverse Society. Boulder, CO: Westview.

44.

Tyler

Tom R.

Lind

E. Allan

. 1992. “A Relational Model of Authority in Groups.” Advances in Experimental Social Psychology 25:115–91.

45.

van den Bos

Kees

. 2005. “What Is Responsible for the Fair Process Effect?” Pp. 273–300 in Handbook of Organizational Justice, edited by Greenberg

Colquitt

J. A.

Mahwah, NJ: Lawrence Erlbaum Associates.

46.

van den Bos

Kees

Lind

E. Allan

Vermunt

Riel

Wilke

Henk A. M.

1997. “How Do I Judge My Outcome When I Do Not Know the Outcome of Others? The Psychology of the Fair Process Effect.” Journal of Personality and Social Psychology 72:1034–46.

47.

Walker

Henry A.

2014. “Legitimacy and Inequality.” Pp. 353–77 in Handbook of the Social Psychology of Inequality, edited by McLeod

J. D.

Lawler

E. J.

Schwalbe

New York: Springer.

48.

Walker

Henry A.

Zelditch

Morris

Jr.

1993. “Power, Legitimacy, and the Stability of Authority: A Theoretical Research Program.” Pp. 364–81 in Theoretical Research Programs: Studies in the Growth of Theory, edited by Berger

Zelditch

Jr. Stanford University Press.

49.

Walster

Elaine

Walster

G. William

Berscheid

Ellen

. 1978. Equity: Theory and Research. Boston, MA: Allyn & Bacon.

50.

Weber

Max.

[1924] 1978. Economy and Society. Vols. I and II. Edited by Roth

Wittich

CA: University of California Press, Berkeley.

51.

Wharton

Amy S.

2014. “Work and Emotions.” Pp. 335–58 in Handbook of Sociology of Emotions, edited by Stets

Turner

New York: Springer.

52.

Wingfield

Adia Harvey

. 2010. “Are Some Emotions Marked ‘White Only’? Racialized Feeling Rules in Professional Workplaces.” Social Problems 57:251–68.

53.

Zelditch

Morris

Jr.

2001. “Processes of Legitimation: Recent Developments and New Directions.” Social Psychology Quarterly 64:4–17.

54.

Zelditch

Morris

Jr.

2006. “Legitimacy Theory.” Pp. 324–52 in Contemporary Social Psychological Theories, edited by Burke

P. J.

Stanford, CA: Stanford University Press.

55.

Zelditch

Morris

Jr. Walker

Henry A.

1984. “Legitimacy and the Stability of Authority.” Advances in Group Processes 1:1–25.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB