Emotional Benefits of Leader Legitimacy

Abstract

Leaders with legitimacy attain substantial benefits in the workplace, like cooperation and minimal dissension from workers. We argue that legitimacy also benefits leaders by enhancing workers’ positive emotions and displays and reducing negative emotions and displays to their behaviors. Employed, adult study participants took the role of a worker and responded to a vignette description of a leader’s behaviors (use of fair/unfair decision- making procedures; high/low use of power benevolently) and status characteristics (Study 1, three months/three years of experience; Study 2, Black male/White male). Results confirm the expected patterns of effects of legitimacy on positive and negative emotions. Perceived legitimacy also largely mediates the effects of the leader’s behaviors and experience (but not race) on emotional responses. Legitimated leaders benefit from a flow of positive emotions and displays and are largely protected from negative emotions and displays that could jeopardize their legitimacy and ultimately the functioning of the work group.

Keywords

emotions group processes inequality leadership legitimacy processes

Critical to the smooth operation of an organization and the successful performance of its leaders, legitimacy affords substantial benefits to organizational authorities. Legitimacy is the process through which patterns of behavior or cultural/social objects gain social support and approval (Johnson, Dowd, and Ridgeway 2006; Zelditch and Walker 1984). People view an occupant of a leadership position who is legitimate as the right and proper person for that position who has earned others’ support. Without legitimacy, leaders risk workers’ noncompliance, which, in turn, often leads to their own ineffective behaviors and performance, creating a vicious circle that reinforces barriers to increasing their legitimacy (Vial, Napier, and Brescoll 2016).

In contrast, when workers perceive leaders as legitimate, they grant them more leeway in their directives and greater freedom of action and offer more compliance and cooperation without resistance, challenge, or scrutiny (Baldassarri and Grossman 2011; Johnson et al. 2006; Zelditch and Walker 1984). Workers are also unlikely to perceive legitimated leaders as “uppity,” ineffective, or a source of resentment (Vial et al. 2016). Gaining legitimacy is particularly crucial for the successful performance of women and people of color, who have the extra burden of “proving” they are competent and worthy to occupy leadership positions (Ridgeway and Nakagawa 2014; Vial et al. 2016).

Research delineates how legitimated leaders benefit in terms of their workers’ behaviors toward and cognitions about workplace authorities. Yet omitted from these investigated benefits are the affective consequences of leader legitimacy. Such consequences involve both the experience of emotions—“relatively short-term evaluative response[s], essentially positive or negative in nature, involving distinct somatic (and often cognitive) components” (Kemper 1978:47)—and their expression. Emotions and their display are often “contagious”: workers can be “infected” by others’ emotional expressions, both positive and negative (Barclay and Kiefer 2014; Hatfield, Cacioppo, and Rapson 1992).

At work, felt positive emotions, like satisfaction, pleasure, and happiness, and their outward expression encourage teamwork, cooperation, and collaboration (Barsade and Gibson 2007). The display of positive emotions transcends positions in work hierarchies, rewarding both those who express and receive them, thus benefiting the entire work group (Lovaglia and Houser 1996; Ridgeway and Johnson 1990; Turner and Stets 2005). Moreover, positive displays come at a low cost and foster enhanced performance and positive social interaction among leaders and workers. Thus, all else equal, positive displays are more common than negative ones (Barasde and Gibson 2007; Ridgeway and Johnson 1990).

Felt negative emotions, such as frustration, annoyance, or disappointment, and their display, in contrast, can be costly to the parties expressing and receiving them. Under certain conditions, such emotions may challenge the status quo or fortify a negotiator’s position, ultimately resulting in positive behaviors to redress the cause of negative emotions. More typically, however, they jeopardize the functioning of the work group (Ridgeway and Johnson 1990). Many contextual events—including the actions of authorities—that foment negative displays disrupt everyday interaction in groups within organizations, increasing discord, reducing cooperation, and stimulating withdrawal by workers (Sinaceur and Tiedens 2006; Turner and Stets 2005).¹

Given the impact of individual and contagious affective reactions in the workplace, their absence in studies about the benefits of legitimacy is notable. Here, we investigate how leader legitimacy affects workers’ affective responses. Specifically, we address the extent to which perceptions of the legitimacy of a workplace leader enhance or inhibit workers’ felt emotions and intended displays in view of uncertainty regarding a worker’s request to that authority. Uncertain situations with uncertain outcomes are typical in the workplace (Cullen et al. 2014). How workers react emotionally during such situations may stem, at least in part, from their leaders’ legitimacy. We predict that legitimacy benefits leaders by eliciting workers’ positive emotions and their display and dampening the experience and display of negative emotions toward the authority.

To assess our predictions regarding the relationships between legitimacy, emotions, and emotional displays, we draw on responses from employed adults in the United States who participated in one of two online vignette studies. In each study, the vignette manipulates an authority’s behaviors contributing to perceived legitimacy: use of procedures fairly (high/low) and of power benevolently to ensure workers have the resources to perform their jobs (high/low; e.g., Hegtvedt et al. 2022; Huq, Jackson, and Trinkner 2017). Also, recognizing that an authority’s status may impact perceived legitimacy, in Study 1, we conceptualize the experience of a leader in the work group as a specific status characteristic (Berger, Rosenholz, and Zelditch 1980), operationalized as three weeks versus three years. In Study 2, intended to replicate the effects of authority behaviors on perceived legitimacy and its impact on emotions, we manipulate a diffuse status characteristic, leader’s race, conveyed by the authority’s name (Jamal/Brad).

We draw on the premises of several theoretical frameworks to make our predictions. Figure 1 shows our conceptual model with antecedents (procedural justice, benevolent power use, status characteristics) to legitimacy and affective responses, illustrating the mediating effects of legitimacy. Results show that perceived legitimacy is positively related to positive emotional responses and negatively related to negative emotional responses. Additionally, perceived legitimacy mediates the effects of leaders’ behaviors and years of experience (but not race) on emotional responses.

Figure 1.

Theorized Pathways of Leader’s Behaviors and Status, Legitimacy, and Emotions

Background: Gaining Legitimacy

Recent research focuses on how leaders gain legitimacy in the workplace. Hegtvedt et al. (2022) theorize about and empirically examine three factors that positively impact gaining legitimacy for authorities. The first factor, from Tyler’s (e.g., Tyler 1997, 2001) identity-based approach to legitimacy, is procedural justice, which refers to the use of decision-making procedures that involve principles of representativeness (i.e., voice of workers in decisions affecting them), impartiality, consistency, accuracy, and correctability. Tyler argues that authorities who use procedurally just decision-making rules signal workers’ value to the group, thereby boosting their dignity. Such behaviors then enhance worker identification with and commitment to the group, increasing the likelihood that they will see their leader as legitimate (e.g., Huq et al. 2017; Tyler 1997).

A second factor, based on a resource-based approach, highlights how benevolent power use by authorities enhances legitimacy (Hegtvedt and Johnson 2009). Power is the ability of actors to impose their will on others despite resistance through control over valued resources (Cook and Emerson 1978; Weber [1924] 1978). Building on Blau’s (1964) classic work on authorities and legitimacy and developments in social exchange theory (e.g., Lawler 2006; Molm 2006), this approach focuses on the nature of power use by authorities in exchanges with workers. Authorities depend on workers’ performances to achieve the organization’s overarching goals. Subordinates depend on an authority for material and nonmaterial resources to get their jobs done. This mutual dependence creates opportunities to enhance the perceived legitimacy of an authority. Authorities use their power benevolently by providing workers with opportunities to grow professionally (e.g., enhanced training, development of a new project; Blau 1964) and with resources needed to succeed in their jobs. Benevolent power use highlights the less formal nature of exchanges, which fosters feelings of mutual obligation and reciprocity and increases the attachment of authorities to their subordinates (Dulebohn et al. 2012). Moreover, using power benevolently demonstrates a leader’s concern for the well-being of subordinates, which cultivates the leader’s identification with workers (Scholl et al. 2017) and enhances workers’ perceptions of leader legitimacy.

Expectation states theory (Berger et al. 1980) offers a third factor affecting legitimacy gains: status characteristics. The specific status characteristic, experience of the leader in a work group, represents skills specific to a given task. A higher level of experience counts as the more valued state of the specific status characteristic than less experience because an authority who has spent more time with a work group and workplace is likely to be more experienced at work-related tasks and have greater familiarity with the culture of the group and organization. Individuals presume possession of this more valued state to signal greater competence on tasks related to that characteristic than possession of the less valued state. In groups, those perceived as more competent receive more action opportunities, thereby exerting greater influence and receiving higher evaluations, which reinforces the worthiness of those with the more valued state of the specific status characteristic. And levels of a specific status characteristic shape other performance judgments, including those of legitimacy. Cohen and Zhou (1991) demonstrate that supervisory experience in a job signals the suitability—that is, the legitimacy—of the person for that position.

Similarly, diffuse status characteristics, such as race, suggest two or more states, each accorded different values of esteem and worthiness (Berger et al. 1980). Diffuse status characteristics activate general and specific expectations about performance. Due to widely held racial status beliefs, individuals generally assume that White authorities are more competent than people of color in similar positions. Several recent studies empirically confirm patterns of influence, group performance, and power use expected stemming from race as a status characteristic (e.g., Manago, Sell, and Goar 2019; Melamed et al. 2019; Thye and Harrell 2017). Beliefs that White individuals are more “typical” in authority positions add to the perception that White individuals are more legitimate in such positions than people of color, all else equal (Ridgeway 2019; Ridgeway and Nakagawa 2014).

Effects of Procedural Justice, Benevolent Power, and Status on Reported Emotions

We first consider the effects of factors affecting legitimacy on reported positive and negative emotions (Figure 1). Justice scholars have demonstrated that people respond with satisfaction to perceived justice and with generic distress or specific emotions, such as anger, resentment, and frustration, to perceived (disadvantageous) injustice (see Hegtvedt and Parris 2014). Specifically, drawing on the identity-based approach, when leaders ensure procedural justice in decision-making by allowing representation (or “voice”) of individuals affected by the decisions, eliminating bias, and ensuring accuracy, they convey to members their value to and high regard in the group. Feelings of value often enhance self-esteem and, in turn, generate positive emotions (e.g., satisfaction, contentment; e.g., DeCremer 2004; Tyler and Lind 1992). In contrast, violations of procedural justice threaten members’ perception of their standing or value to the group, potentially challenging feelings of value and esteem and thus stimulating negative emotions (e.g., Barkworth and Murphy 2015).

Likewise, use of fair procedures should affect intended displays of emotions, that is, outward expressions exhibited through facial expressions, language, tone of voice, and physical gestures (Rafaeli and Sutton 1989). Compared to positive emotions, expression of negative emotions is more costly generally, and contextual factors may increase those costs (Hochschild 1983; Turner and Stets 2005). In the absence of situational constraints and consistent with previous empirical patterns, we predict:

Hypothesis 1: Procedural justice in decision-making by a leader is positively related to the intensity of workers’ reported positive emotions and intended displays and negatively related to the intensity of workers’ reported negative emotions and intended displays.

Drawing on the resource-based approach, we argue that benevolent power use by leaders (ensuring that workers have resources and opportunities to do their jobs well) produces successful exchanges (Hegtvedt and Johnson 2009). Successful exchanges, in turn, foster feelings of mutual obligation and reciprocity and evoke positive emotions toward the relationship (Lawler 2006). Also, when workers have the resources and opportunities they need to succeed, they feel valued by the authority (Hegtvedt and Johnson 2009) and are thus likely to experience positive emotions. In contrast, when leaders fail to use power benevolently, inhibiting successful exchanges and feelings of reciprocity, workers are more likely to feel negative emotions. Given that felt emotions lead to displays in the absence of constraints, when authorities use power benevolently, the resulting experience of positive and negative emotions should also impact the likely display of these emotions. Thus, we expect:

Hypothesis 2: Benevolent power use by a leader is positively related to the intensity of workers’ reported positive emotions and intended displays and negatively related to the intensity of workers’ reported negative emotions and intended displays.

Leaders’ status conveys perceived competence (Berger et al. 1980), with implications for status hierarchies. Ridgeway and Berger’s (1986) extension of expectation states theory regarding valued status positions focuses on the legitimation process of a status hierarchy. Society has beliefs about the types of people that usually hold leadership positions in the surrounding society. When a person is advantaged by a high state of a diffuse status characteristic, such as race, in a hierarchy, it looks “right” (Ridgeway 2019; Ridgeway and Berger 1986) because it is consistent with broad cultural beliefs about who generally has higher status and holds leadership positions. Thus, White authorities are perceived as “looking the part” of the leader more so than Black authorities. Experienced leaders should also look more the part than inexperienced ones. A leader who has an advantage on a diffuse status characteristic may be treated as more legitimate than a leader disadvantaged on a diffuse status characteristic (even with ability) because cultural beliefs about who should be in leadership positions are distinct from beliefs about who is competent in the situation. Additionally, status structures are more likely to be legitimate when the high status member also has diffuse status advantages (e.g., race) than when the high status member has a diffuse status disadvantage (Ridgeway, Johnson, and Diekema 1996). This structural arrangement of high levels of additional status characteristics (both diffuse and specific), beyond holding the leader position, allows these leaders to receive more benefits and rewards, including intangible ones like emotions, than leaders without status advantages. Therefore, we argue that experience and race will have a similar impact on emotions, through legitimacy, even though experience is a specific and race is a diffuse status characteristic.

From a structural standpoint (Kemper 1978), high status leaders (i.e., experienced or White) may be more likely to elicit positive emotions and displays compared to authorities who possess the less valued state (i.e., less experienced or Black). Leaders with additional status advantage are also less likely to elicit negative emotions and be the recipients of negative displays than those who are status disadvantaged. In addition, workers may perceive high compared to low status leaders as doing a better job even when job performances are similar. Owing to this double standard (Foschi 2000), workers may feel more positive emotions and fewer negative emotions toward a status advantaged leader compared to a status disadvantaged leader.²

Thus, we predict:

Hypothesis 3: Leaders’ status (represented by levels of experience in Study 1 and race in Study 2) is positively related to the intensity of workers’ reported positive emotions and intended displays and negatively related to the intensity of workers’ reported negative emotions and intended displays.

Mediating Process of Legitimacy

The proposed effects of procedural justice, benevolent power use, and status on emotional responses ignore how the designation of a leader as legitimate may mediate those effects and profoundly impact workers’ emotions. Legitimated authorities have the advantage over those lacking legitimacy because they are perceived to deserve the right to exercise their authority (e.g., Ford and Johnson 1998; Zelditch and Walker 1984). Additionally, with legitimated authorities, work groups are perceived as more effective, and members are more committed to the team (Yoon and Thye 2011), which likely boosts satisfaction and worker morale. Moreover, when the leader is legitimate, workers perceive less uncertainty, their experiences grow more positive (owing to fewer disagreements; see Kalkhoff 2005), and their sense of security increases (Webster and Walker 2014). Beliefs that the leader is the right and proper person for the job and has coworker support should thus foster positive emotions.

Conversely, leaders who lack legitimacy are perceived as less deserving to exercise their authority and are more likely to engender negative emotions, particularly in uncertain situations. Perceived illegitimacy of a leader, moreover, fuels workers’ feelings of uncertainty, creating more distress (Kanter 1977), insecurity, and decreased commitment to the work group. Empirical evidence shows that lack of leader legitimacy enhances subordinates’ reported negative emotions (Johnson et al. 2016; Johnson, Ford, and Kaufman 2000).

Leader legitimacy should also shape workers’ intended displays of emotions toward the authority. Legitimacy should enhance display of positive emotions because workers will expect fewer disagreements and more certainty with legitimated leaders in future interactions. In addition, displaying positive emotions elicits few sanctions from coworkers and the leader. In contrast, workers may be more reluctant to display negative emotions to legitimated leaders to avoid incurring sanctions (Johnson et al. 2000). In the absence of leader legitimacy, workers may expect more disagreements and uncertainty and perceive fewer costs to expressing negative emotions. Thus, we expect:

Hypothesis 4: Perceived legitimacy of the leader is positively related to the intensity of workers’ reported positive emotions and intended display and negatively related to the intensity of workers’ reported negative emotions and intended display.

We argue that perceived leader legitimacy is a fundamental mediator in shaping positive and negative emotions in workgroups (Figure 1). Insofar as leaders’ behaviors (using fair decision-making procedures and power benevolently) and possession of highly valued states of status characteristics enhance legitimacy, uncertainty and doubt in the workplace decrease, generating confidence in the leader (Kanter 1977; Webster and Walker 2014). Holding leaders with such regard translates into emotional responses to their behaviors and status characteristics and, by extension, in the absence of constraints, to intended emotional displays. Thus, we posit:

Hypothesis 5: Perceived legitimacy of the leader mediates the positive effects of fairness in decision-making, use of power benevolently, and leader status on workers’ intensity of reported positive emotions and intended displays and mediates the negative effects of these factors on workers’ intensity of reported negative emotions and intended displays.

Study 1

Methods

Study design and data

Study participants read a vignette³ titled “Workplace Dynamics” (adapted from Johnson et al. 2000) of a hypothetical manager-employee situation. We developed eight versions of the vignette using a 2 × 2 × 2 factorial design: (1) use of fair decision-making procedures (high/low), (2) use of benevolent power (high/low), and (3) length of experience of the manager in the workplace (three weeks/three years). Participants assumed the role of a salesperson employed in a medium-size clothing manufacturing company, who is paid a salary and does not receive commissions from sales. The employee has seen his or her sales performance steadily increase and thus asks the manager for a moderate 5 percent raise. The manager indicates that the raise will be considered at the next performance review; therefore, the pay outcome is uncertain in the vignette. Each vignette alters information to create variation in our three theoretical factors. After reading the vignette, participants answered a questionnaire with items measuring their perceptions of the manager as legitimate, their felt emotions in their interaction with the manager, and the likelihood of expressing their emotions to the manager.

Following best practices for vignette experiments, we developed plausible and clear vignettes, excluding complex and distracting information (Aguinis and Bradley 2014). Manipulated factors are orthogonal to evaluate the independent effects of each (Karren and Barringer 2002). Respondents assessed the vignettes online in a single session. The between-subject design limits respondents’ ability to detect our experimental manipulations. Together, these practices ensure internal validity (Aguinis and Bradley 2014).

The Survey Research Lab at a major university, in partnership with Dynata, executed the study online in fall 2018. Dynata maintains a large, diverse sampling frame that mirrors national census data regarding distributions of gender, race, age, and geography. Dynata contacts potential respondents via email from their opt-in panels.⁴ Those participating (for a small monetary incentive) were randomly assigned to one of eight vignettes, with the goal of 85 to 100 respondents for each. We collected 807 responses from adults employed full-time, not self-employed, between the ages of 20 and 65, and residents of the United States for 10 years or more. Table 1 includes descriptive characteristics of study participants. The survey took 10 to 15 minutes (median = 10.88) to complete. Missing responses on questionnaire items were randomly distributed across all vignettes. Sample sizes across conditions ranged from 98 to 105.

Table 1.

Sample Characteristics and Means (Standard Deviations) for Focal Factors in Studies 1 and 2

	Study 1 (experience)N = 807		Study 2 (race)N = 673		Range
	M	SD	M	SD	Minimum	Maximum
% Female	.50	(.50)	.52	(.50)	0	1
% Non-White^a	.27	(.45)	.46	(.50)	0	1
Age	41.42	(12.34)	48.04	(11.18)	20	65
Education^b	3.68	(1.46)	3.74	(1.28)	1	7
Income^c	3.61	(1.71)	4.12	(2.02)	1	8
Legitimacy	4.99	(1.69)	4.67	(1.81)	1	7
Positive emotions	4.59	(1.93)	4.20	(1.97)	1	7
Positive display	4.57	(1.86)	4.20	(1.94)	1	7
Negative emotions	3.87	(2.02)	3.68	(2.05)	1	7
Negative display	3.42	(1.89)	3.24	(1.87)	1	7

In Study 1, non-White includes Asian, Black, Latinx, Native American, and multiracial respondents. In Study 2, non-White consists of Black respondents.

Education categories: (1) high school graduate/GED/less than high school, (2) technical/vocational, (3) some college or associate’s degree, (4) bachelor’s degree, (5) master’s degree, (6) professional degree, (7) doctoral degree.

Income categories: (1) less than $25,000; (2) $25,0001 to 50,000; (3) $50,001 to 75,000; (4) $75,000 to 100,000; (5) $100,001 to 150,000; (6) $150,001 to 200,000; (7) $200,001 to 250,000; (8) over $250,00.

Manipulations

To manipulate procedural justice, we provided information on the listening and encouraging style of the manager that reflects the procedural justice principle of voice. For high procedural justice (coded 1) the vignette read, “The manager listens to employees and encourages them to: 1) offer suggestions about how they handle their work; and 2) comment on policy development within the division.” For low procedural justice (coded 0) the vignette read, “The manager does not listen to employees or encourage them to . . . ” do the acts described in the fair condition. Additionally, the salesperson gets “an appointment with your manager very easily (given that willingness to listen to employees)” or “with great difficulty (given that unwillingness to listen to employees)” (high and low procedural justice, respectively).

Drawing on Blau’s (1964) work, we operationalize benevolent power use by how the manager provides opportunities (e.g., training and attending conferences) and resources to the salespeople to facilitate their work. Benevolent power use is high (coded 1) or low (coded 0) when: “The manager also often (rarely) finds ways to provide opportunities for sales employees to perform the work that they do well, for example by ensuring (denying) access to opportunities for further training or sending them to sales conferences to expand their networks.” The manager also either “strongly” or “weakly” “urged upper management to approve” the work group for “the much-desired new tablet computers to facilitate [their] work,” and they are consequently either at the top or the bottom of the company list to acquire the tablets.

For manager’s experience, the vignette indicates that the manager has supervised the employee’s division “for three years and is thus very familiar with the people and culture in your division as well as other divisions,” (coded 1) or “for three weeks and is thus not familiar with the people and culture in your division as well as other divisions” (coded 0).

Measures

To measure the latent construct, perceived legitimacy of the manager, respondents indicated “how much” or “to what extent” they support the manager, see the manager as the right and proper person for the position, believe that coworkers support the manager, and believe that coworkers see the manager as the right and proper person for the position (1 = very little, 7 = a great deal). The alpha reliability is .95.⁵

Our four dependent variables tap into positive and negative emotions characteristic of justice studies (Barclay and Keifer 2014). We asked: “How much did you feel each of the following in your interactions with your manager?” and “How likely would you be to express each of the following feelings when you next interact with your manager?” (1 = not at all, 7 = a great deal). We created four additive scales (each with three items, averaged by the number of items) based on theoretical reasoning, the vignette context, and results from principle components analysis: (1) reported positive emotions (satisfied, pleased, and happy; α = .97), (2) likelihood of intended display of positive emotions (satisfaction, pleasure, and happiness; α = .94), (3) reported negative emotions (frustrated, disappointed, and annoyed; α = .94), and (4) likelihood of intended display of negative emotions (frustration, disappointment, and annoyance; α = .93).

Analysis strategy

We used analysis of variance to assess our vignette manipulations. We then employed structural equation modeling (SEM) in Stata to test the direct and indirect effects of our manipulations in simultaneous equations and evaluate mediated paths (Acock 2013). The experimental vignette methodology allows for clear instantiation of exogenous and endogenous variables in SEM, and random assignment to the experimental conditions eliminates the need for controls.

Structural equation models begin with a measurement phase and then move to a structural phase (Acock 2013). Thus, we first created a measurement model to assess the fit and validity of our latent construct. We then included structural paths between the exogenous and endogenous variables to examine the indirect effects of our manipulations through legitimacy on emotional responses. We used maximum likelihood parameter estimation to account for missing values.

We test Hypotheses 1 to 3 through assessing the total effects of our manipulations on positive and negative emotions. To examine Hypothesis 4, we examine direct effects of legitimacy on emotions. To assess the mediation predicted in Hypothesis 5, we examine the presence of significant total effects and significant indirect effects of our manipulations on emotions through legitimacy and nonsignificant direct effects of antecedents to emotions (Kenny, Kashy, and Bolger 1998).⁶

Results

Univariate Analyses and Vignette Manipulation Checks

Table 1 shows means (standard deviations) for the manipulated variables, legitimacy, and emotions. As expected, means for reported positive emotions and intended displays are higher than the means for reported negative emotions and intended displays.⁷

Manipulation checks show that participants in the high procedural justice condition rated the manager as more encouraging of suggestions (M = 5.95) and comments on policies (M = 5.73) than those in the low condition (M = 2.86 and M = 2.96, respectively), F(1, 797) = 569.47, p < .001 and F(1, 794) = 457.75, p < .001, respectively. Checks on benevolent power use show differences in assessments of the extent to which the manager found ways to provide opportunities to enhance job performance (M = 5.53 vs. M = 3.86 for high and low benevolent power use, respectively), F(1, 797) = 181.91, p < .001, and urged upper management to provide tablet computers (M = 5.82 vs. M = 4.07), F(1, 789) = 200.20, p < .001. Participants in the “experienced” condition viewed their manager as more familiar with the people and culture of the division (M = 6.03) than those in the “unexperienced” condition (M = 3.18), F(1, 799) = 523.21, p < .001.

Additionally, respondents indicated how involved they felt in the situation and how realistic they believed their responses to be (1 = not very involved/realistic, 7 = very involved/realistic). Respondents indicated that they were largely involved (M = 5.19) and believed their responses were realistic (M = 5.87).

Measurement Model

We assessed the fit of our one-factor measurement model for legitimacy using the following indices: model chi-square, Bentler comparative fit index (CFI; Bentler 1990), Tucker-Lewis index (TLI; Tucker and Lewis 1973), and Steiger-Lind root mean square error of approximation (RMSEA; Steiger 1990). A model chi-square with a p value greater than .05, CFI greater than .90, TLI greater than .95, and RMSEA less than .05 indicates good model fit. Fit indices for the measurement model indicate a good fit between the model and the sample data: χ²(20) = 31.775, p = .046, CFI = .997, TLI = .994, RMSEA = .032.

Structural Model

Fit statistics for the structural model indicate a good fit: χ²(21) = 29.437, p = .104; CFI = .998; TLI = .996; RMSEA = .026. Table 2 provides the total, direct, and indirect effects represented in our model. The total effects of procedural justice on the intensity of reported positive emotions (β = .25, p < .001), intended positive display (β = .23, p < .001), intensity of reported negative emotions (β = −.25, p < .001), and intended negative display (β = −.17, p < .001) are significant and in the predicted direction, confirming Hypothesis 1. Procedural justice positively affects the intensity of reported positive emotions and likelihood of their display and negatively affects the intensity of reported negative emotions and likelihood of their display. Total effects of benevolent power on intensity of reported positive emotions (β = .10, p < .05), intended positive display (β = .08, p < .05), and intensity of reported negative emotions (β = −.08, p < .05) are significant and in the predicted direction. Total effects of benevolent power on the likelihood to display negative emotions are not significant. Results largely confirm Hypothesis 2, that benevolent power significantly affects emotional responses. Total effects of the authority’s experience on reported positive emotions (β = .14, p < .001) and intended positive display (β = .15, p < .001) are significant but are not significant for negative emotions and displays. Thus, Hypothesis 3 is partially supported.

Table 2.

Study 1: Structural Equation Model Results of Procedural Justice, Benevolent Power, and Manager Experience on Positive and Negative Reported Emotion Experience and Intended Display

Structural paths	Legitimacy	Reported positive emotions	Intended positive display	Reported negative emotions	Intended negative display
Direct effects
Manipulations
Procedural justice	.26*** (.11)	.02 (.08)	.01 (.10)	–.14*** (.15)	–.12*** (.16)
Benevolent power	.12** (.11)	–.01 (.08)	–.02 (.09)	–.03 (.15)	–.03 (.15)
Status characteristic	.13** (.11)	.03 (.08)	.04 (.09)	.00 (.15)	–.03 (.16)
Mediator
Legitimacy		.88*** (.04)	.83*** (.04)	–.45*** (.06)	–.18*** (.06)
Indirect effects
Procedural justice		.22*** (.12)	.22*** (.13)	–.12*** (.09)	–.05* (.05)
Benevolent power		.10** (.11)	.10** (.12)	–.05** (.08)	–.02* (.03)
Status characteristic		.09** (.11)	.11** (.12)	–.06** (.08)	–.02* (.03)
Total effects
Procedural justice		.25*** (.15)	.23*** (.14)	–.25*** (.16)	–.17*** (.16)
Benevolent power		.10* (.15)	.08* (.15)	–.08** (.17)	–.05(.16)
Status characteristic		.14*** (.15)	.15*** (.15)	–.06(.17)	–.06(.16)

Note: Standardized coefficients (standard errors) are reported. Error variances and factor variances and covariances are available on request.

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

Hypothesis 4 predicts that perceived legitimacy of the authority will enhance positive and reduce negative reported emotions and intended displays. Confirming Hypothesis 4, legitimacy enhances the intensity of workers’ reported positive emotions (β = .88, p < .001) and likelihood of intended positive displays (β = .83, p < .001) and dampens the intensity of workers’ reported negative emotions (β = −.45, p < .001) and likelihood of intended negative displays (β = −.18, p < .001).

Hypothesis 5 predicts that perceived leader legitimacy mediates the effects of the manipulated factors on the intensity of reported emotions and likelihood of intended displays, both positive and negative. The indirect effects of each treatment on emotions through legitimacy indicate support for the mediation hypothesis. Procedural justice and benevolent power have positive indirect effects on the intensity of workers’ reported positive emotions (β = .22, p < .001 and β = .10, p < .01, respectively) and negative indirect effects on the intensity of workers’ reported negative emotions (β = −.12, p < .001 and β = −.05, p < .01, respectively). The authority’s experience indirectly influences workers’ intensity of reported positive (β = .09, p < .01) and negative emotions (β = −.06, p < .01). Although results regarding positive emotions confirm mediation, in the absence of a total effect of experience on negative emotions, we cannot conclude mediation.

Similar patterns emerge for intended emotion displays. Fair decision-making and benevolent power use by the leader positively affect the likelihood of workers’ intended positive displays indirectly through legitimacy (β = .22, p < .001 and β = .10, p < .01, respectively). Procedural justice negatively affects the likelihood of workers’ intended negative displays through legitimacy (β = −.05, p < .01). Although benevolent power also has an indirect effect (β = −.02, p < .05), in absence of the total effect, mediation is not observed. The leader’s length of experience exerts the hypothesized indirect effect on intended positive displays (β = .11, p < .01). Results regarding negative displays show a significant indirect effect (β = −.02, p < .05), yet with no total effect of experience on negative displays, mediation is inconclusive.

In assessing Hypothesis 5, the effects of procedural justice on reported positive emotions and intended positive display are fully mediated by perceived legitimacy. Owing to the significant direct effect of procedural justice on reported negative emotion and intended negative emotion display, its effects are partially mediated by perceived legitimacy. Perceived legitimacy fully mediates effects of benevolent power on reported positive and negative emotions and intended positive displays. Regarding intended negative displays, mediation could not be assessed because there is no total effect of benevolent power on negative display. Finally, manager experience is fully mediated by perceived legitimacy for reported positive emotions and intended positive displays. Given lack of total effects of manager experience on negative emotions, mediation could not be assessed. These results mostly confirm Hypothesis 5, that perceived legitimacy mediates the effects of procedural justice, benevolent power, and experience on emotions.

Study 2

Methods

Study design, manipulations and data

To replicate Study 1, Study 2 vignettes include the same scenario and manipulations for procedural justice and benevolent power. But here we manipulated status by the race of the manager (White/Black) using the name “Brad” (coded 0) or “Jamal” (coded 1).⁸ Gaddis (2017) shows that more than 95 percent of his study respondents typically perceive Brad as a “White” name and Jamal as a “Black” name. Other studies (e.g., Pedulla 2014) used White-sounding and Black-sounding names to manipulate race. We repeat the name of the manager five times in the vignette.

In contrast to the managerial experience manipulation in Study 1, in Study 2, we hold manager experience constant at two years across vignettes. We are not comparing experience across the two studies; therefore, we are not limited to using three weeks or three years (as in Study 1). We chose two years to ensure equivalence of experience across Black and White managers and avoid the assumption that Black managers have less experience (and therefore less competence) than White managers. We reasoned that holding constant experience at two years would isolate the race effect. However, it is possible that providing this information may, in fact, suppress the effect of race in this study because experience may enhance the legitimacy of Black managers. We return to this design issue in the discussion.

The same research lab executed Study 2 online in fall 2019. We strove for 50 percent Black respondents and 50 percent White respondents, with a goal of 25 percent of each of the following categories: Black females, White females, Black males, and White males. Dynata used stratified (by race and gender) random assignment of participants until enough were in each experimental condition. To check recall of the race-related name of their manager, we asked respondents to choose their vignette manager’s name out of four names (Brad, Dwayne, Greg, and Jamal). Seventy-two percent of respondents correctly named Jamal or Brad for their assigned condition, out of 938 responses. The final sample was 673 respondents (170 Black females, 140 Black males, 181 White females, and 182 White males). Sample sizes in the eight conditions ranged from 72 to 90.⁹ Table 1 includes descriptive characteristics on Study 2 participants. Missing responses were randomly distributed with no skip or missing patterns emerging. Participants completed the survey in 10 to 15 minutes (median = 13.80).

Measures and analysis strategy

We replicated measurement of our mediating factor, perceived manager legitimacy, and emotion outcomes in Study 2 (α reliabilities: legitimacy = .95, reported positive emotions = .97, reported negative emotions = .94, likelihood of intended display of positive emotions = .94, likelihood of intended display of negative = .93). We used analysis of variance to evaluate the manipulation checks and SEM to fit measurement and structural models.

Results

Univariate Analyses and Vignette Manipulation Checks

Table 1 shows means (standard deviations) for the manipulated factors, legitimacy, and emotion responses. As in Study 1, means for reported positive emotions and intended displays are higher than means for reported negative emotions and intended displays.

Participants in Study 2 perceived managers in the high procedural justice condition as more encouraging of suggestions (M = 6.14 vs. M = 1.97), F(1, 661) = 1,173.55, p < .001, and comments on policies (M = 5.69 vs. M = 2.12), F(1, 657) = 708.97, p < .001, than those in the low condition. Respondents in the high compared to the low benevolent power condition rated managers as more likely to give opportunities to enhance job performance (M = 5.78 vs. M = 2.28), F(1, 661) = 678.23, p < .001, and urge upper management to provide tablet computers (M = 6.20 vs. M = 2.43), F(1, 659) = 815.82, p < .001. Participants reported that they were largely involved (M = 5.04) and believed their answers were realistic (M = 5.90).

Measurement and Structural Model

The measurement model indicates a good fit with the data, χ²(21) = 29.446, p = .104; CFI = .998; TLI = .996; RMSEA = .024. The structural model also has good fit: χ²(22) = 29.603, p = .128; CFI = .999; TLI = .997; RMSEA = .023.

Table 3 provides the direct, indirect, and total effects. Results are similar to Study 1 in terms of coefficient size, direction, and significance, with a few notable differences. The total effects of procedural justice on the intensity of reported positive emotions (β = .19, p < .001), intended positive display (β = .18, p < .001), intensity of reported negative emotions (β = −.14, p < .001), and negative display (β = −.09, p = .016) are significant and in the predicted direction, confirming Hypothesis 1. Total effects of benevolent power on reported intensity of positive emotions (β = .27, p < .001), intended positive display (β = .25, p < .001), intensity of reported negative emotions (β = −.20, p < .001), and intended negative display (β = −.16, p < .001) are significant and in the predicted direction, confirming Hypothesis 2. The authority’s race does not produce significant total effects on the intensity of workers’ positive or negative emotions or intended displays, disconfirming Hypothesis 3 and therefore also mediation of race effects proposed in Hypothesis 5. Like Study 1, procedural justice and benevolent power affect emotions, but unlike managerial experience in Study 1, the status characteristic of race does not significantly affect emotions.

Table 3.

Study 2: Structural Equation Model Results of Procedural Justice, Benevolent Power, and Race of Manager on Positive and Negative Reported Emotion Experience and Intended Display

Structural paths	Legitimacy	Reported positive emotions	Reported positive display	Reported negative emotions	Reported negative display
Direct effects
Manipulations
Procedural justice	.18*** (.11)	.03(.08)	.05(.10)	–.02(.12)	–.02(.14)
Benevolent power	.27*** (.12)	.04(.08)	.05(.10)	–.02(.13)	–.05(.13)
Status characteristic	–.05(.11)	.00(.08)	.02(.10)	.01(.12)	–.07(.13)
Mediator
Legitimacy		.87*** (.04)	.75*** (.04)	–.66*** (.05)	–.39*** (.05)
Indirect effects
Procedural justice		.15*** (.13)	.13*** (.11)	–.12*** (.07)	–.07*** (.06)
Benevolent power		.24*** (.13)	.21*** (.11)	–.18*** (.07)	–.11*** (.07)
Status characteristic		–.04(.13)	–.04(.11)	.03(.06)	.02(.05)
Total effects
Procedural justice		.19*** (.14)	.18*** (.14)	–.14*** (.16)	–.09* (.14)
Benevolent power		.27*** (.14)	.25*** (.14)	–.20*** (.16)	–.16*** (.14)
Status characteristic		–.04(.14)	–.02(.14)	.04(.16)	–.05(.14)

Note: Standardized coefficients (standard errors) are reported. Error variances and factor variances and covariances are available on request.

p < .05. ***p < .001 (two-tailed tests).

Perceived legitimacy is positively related to the intensity of workers’ reported positive emotions (β = .87, p < .001) and likelihood of intended positive displays (β = .75, p < .001) and negatively related to the intensity of workers’ reported negative emotions (β = −.66, p < .001) and likelihood of intended negative displays (β = −.39, p < .001), confirming Hypothesis 4.

For Hypothesis 5, the authority’s use of fair procedures and power benevolently have a positive indirect effect on the intensity of workers’ reported positive emotions (β = .15, p < .001 and β = .24, p < .001, respectively) and a negative indirect effect on the intensity of workers’ reported negative emotions (β = −.12, p < .001 and β = −.18, p < .001, respectively). Similar patterns emerge for intended emotion displays. The leader’s use of procedural justice and benevolent power positively affect the likelihood of workers’ intended positive displays through legitimacy (β = .13, p < .001 and β = .21, p < .001, respectively) and negatively affect the likelihood of intended negative displays (β = −.07, p < .001 and β = −.11, p < .001, respectively).

Perceived legitimacy of the manager fully mediates the effects of procedural justice and benevolent power on reported positive and negative emotions and intended positive and negative displays (Hypothesis 5). The significant indirect effects of procedural justice and benevolent power and the absence of significant direct effects lend strong support to Hypothesis 5. This pattern of results replicates Study 1 findings, supporting the key role of perceived legitimacy regarding workers’ emotions.

Although random assignment to the experimental conditions eliminates the need for controls, we conducted ordinary least squares regression analyses to determine if respondents’ race or gender had any independent (or interactive) effects altering our findings. We found that neither gender, in Study 1, nor race or gender, in Study 2, significantly altered the patterns of results.

Discussion

Although previous research demonstrates that legitimated leaders benefit in terms of their workers’ behaviors and cognitions, little work considers the effects of perceived legitimacy on workers’ affective responses. We address this issue by examining emotional benefits of leader legitimacy in two vignette studies, each involving the use of fair decision-making procedures and benevolent power as antecedents to perceived leader legitimacy. We also assessed the impact of the authority’s status on emotional outcomes, focusing on the specific status of workplace experience in Study 1 and on the diffuse status of the authority’s race in Study 2. Both studies clearly demonstrate that legitimated leaders enjoy greater emotional benefits in the workplace than their counterparts who lack legitimacy.

Results generally support our proposed conceptual model. First, findings are consistent with our theoretical arguments derived from the identity-based (Hegtvedt and Parris 2014; Tyler and Lind 1992) and resource-based approaches (Hegtvedt and Johnson 2009; Lawler 2006). Procedural justice and benevolent power are positively related to reported positive emotions and intended displays and negatively related to reported negative emotions and negative displays. Second, results are mixed regarding our theoretical arguments derived from expectation states theory (Ridgeway and Berger 1986). Managerial experience is positively related to positive emotion outcomes but not significantly related to negative emotions outcomes; race is not significantly related to positive or negative emotions outcomes. We further discuss possible reasons for these results and implications for theory.

Third, regarding our theoretical arguments derived from the legitimacy literature, findings are consistent with the thrust of our model (Johnson et al. 2000; Zelditch and Walker 1984). Perceived legitimacy of the leader strongly enhances the intensity of workers’ reported positive emotions and likelihood of their display and mediates the effects of the authority’s behaviors and experience on these positive emotions. Also, perceived legitimacy of the leader reduces the likelihood of eliciting workers’ negative emotions and their outward display and largely mediates the effects of the authority’s behavior on negative emotion outcomes.

Our findings suggest that legitimated leaders benefit from a flow of positive emotions, which should foster workplace cooperation, collaboration, and teamwork (Lovaglia and Houser 1996; Ridgeway and Johnson 1990; Turner and Stets 2005). A positive work environment, in turn, reinforces legitimacy of the leader. Legitimacy also acts as a buffer that protects leaders from negative emotions and displays in the workplace that could jeopardize their legitimacy. Legitimated authorities benefit because negative expressions can lead to discord, disruptions, scrutiny, demoralization, and struggles between leaders and workers (Barclay and Kiefer 2014; Barsade and Gibson 2007; Sinaceur and Tiedens 2006).

As noted earlier, leader experience is significantly related to both legitimacy and positive emotions outcomes. Authority’s race, however, was not significantly related to legitimacy or emotions, contrary to what we predicted. Certainly, evidence indicates that race is a status characteristic linked to competence (e.g., Manago et al. 2019; Melamed et al 2019; Thye and Harrell 2017) and that women and people of color in leadership positions are more likely to struggle in gaining legitimacy than White men (Ridgeway and Nakagawa 2014; Vial et al. 2016). The lack of a race effect may stem from the prominence in the vignettes of the leader’s behaviors compared to the information provided about the leader’s race. Procedural justice and benevolent power use were manipulated by describing actions of and information about the authority, whereas race was manipulated only by information (i.e., a name). Although information alone worked for the Study 1 specific status characteristic, the difference in how behaviors versus statuses are introduced could have diminished the effects of diffuse status (i.e., racialized) biases on assessments of the situation. Whether actions trump the effect of an authority’s race in actual (rather than imagined) work situations remains to be investigated. We are also mindful that perceptions are murkier and more multifaceted in workplaces (compared to vignette descriptions) because they stem from everyday interactions that accumulate and involve both verbal and nonverbal cues.

Additionally, authority’s race may not have had significant effects due to choices we made in the vignette design. First, we assigned two years of experience for both Black and White authorities. We made this decision to avoid the assumption that Black managers have less experience and therefore have less competence than White managers. In doing so, however, we may have legitimated Black authorities via sufficient years of experience, suppressing any race effect, rather than isolating the race effect. Experience cultivates credibility with workers because more experienced leaders have more knowledge of company culture and structure (Kanter 1977) and conceivably greater opportunity for support from above and below. In addition, Ridgeway (2019) and Manago et al. (2019) note that specific status characteristics (in our case, experience) often decrease the effect of a diffuse status characteristic. As a result, the design of Study 2 may compromise the test of expectation states theory’s predictions about race effects on legitimacy and emotions. Future research might address this design limitation by using a short length of experience, such as two weeks instead of two years. White managers with little experience may be given the benefit of the doubt and perceived as more competent than Black managers with little experience.

Second, we chose a salesperson job in a manufacturing company. Future research could vary the type of occupation, such as investment banker or engineer, to see if occupation type affects the impact of race of the authority.

Although vignette studies are useful in testing theoretical ideas, there are limitations. Vignette methodology fails to capture workers’ actual feelings and emotional expressions toward their leaders within their own work setting. Emotional reactions to participants’ own workplace experiences may also reflect their knowledge of feeling and display rules within their work contexts (Hochschild 1983; Wingfield 2010). Our study captures what participants report they are likely to feel and likely to express in a hypothetical situation, not what they actually feel or express in response to their encounters with their own managers. Relatedly, Collett and Childs (2011) compared emotional responses of participants using a vignette study and a laboratory study. They found that although participants reported similar experience of emotions across the studies, participants reported greater intensity of emotions in the in-person lab study. Kawakami et al. (2009) also found discrepancies between people’s imagined and actual feelings and behavior in response to a racist act: respondents reacted more strongly when imagining how they would feel than when confronted with an actual situation. Our use of U.S. working adult samples, at least, ensures familiarity with the scenario described in the vignette that Collett and Childs (2011) argue enhances consistency between imagined and actual behavior.

Despite limitations of vignette studies, the data fit our proposed model hypothesizing that perceptions of leader legitimacy mediate the effects of leader behaviors on emotions outcomes well. Nonetheless, we can imagine a different proposed ordering—one that suggests that self-reported emotions (not necessarily their expression) predict perceived legitimacy. In such a model, leader behaviors could elicit positive feelings, which, in turn, affect perceptions of legitimacy. To assess which model fits the data better, we used the mathematical method, Akaike information criterion (AIC), where lower values reveal a better fit (Burnham and Anderson 2004). In both studies, our model fit the data better than the alternative (Study 1 hypothesized AIC = 16,426.843 vs. alternative AIC = 22,754.310; Study 2 hypothesized AIC = 19,369.662 vs. alternative AIC = 26,180.743).

Although the AIC allows for a statistical assessment of fit of the models representing different conceptual ordering, we recognize the interplay between cognitions (about legitimacy) and affect. Thus, the relationship between emotions and perceived legitimacy may be reciprocal. Legitimacy begets positive emotions; positive emotions beget legitimacy. Lack of legitimacy begets negative emotions; negative emotions negatively impact legitimacy. Although our results illustrate that legitimated leaders benefit from workers’ emotion, future research should scrutinize the potential reciprocal relationship between perceived legitimacy and affect in the workplace. Additionally, we examined these relationships under a condition of uncertainty between the leader and workers. Future work might consider whether these relationships hold under a condition of certainty (e.g., when the authority makes a final decision).

To capitalize on the complexity of the workplace, future research should also involve a survey study of U.S. adults that investigates the effects of perceptions of their own authorities’ behaviors, status, and legitimacy on their emotional responses in their workplaces. Such a study would allow assessment of our conceptual model in the context of repeated interactions and consider variation in the type, size, and composition of the work setting and organization. Questions might also tap into emotion management techniques that workers use when they feel but do not express negative emotions for fear of sanctions by others, creating emotional dissonance that may be impacted by leader legitimacy. Survey data would additionally allow for assessment of whether women and people of color in leadership positions, who use behaviors such as procedural justice or benevolent power, increase their legitimacy and, in turn, reap emotional benefits along with cognitive and behavioral benefits.

Our study highlights the important role of leader legitimacy on cultivating positive emotions and inhibiting negative emotions in the workplace. In doing so, we extend previous theorizing and empirical work on the positive consequences of leader legitimacy on workers’ cognitions and behaviors.

Supplemental Material

sj-docx-1-spq-10.1177_01902725231199720 – Supplemental material for Emotional Benefits of Leader Legitimacy

Supplemental material, sj-docx-1-spq-10.1177_01902725231199720 for Emotional Benefits of Leader Legitimacy by Cathryn Johnson, Ryan Gibson, Kate Hawks and Karen A. Hegtvedt in Social Psychology Quarterly

Supplemental Material

sj-pptx-2-spq-10.1177_01902725231199720 – Supplemental material for Emotional Benefits of Leader Legitimacy

Supplemental material, sj-pptx-2-spq-10.1177_01902725231199720 for Emotional Benefits of Leader Legitimacy by Cathryn Johnson, Ryan Gibson, Kate Hawks and Karen A. Hegtvedt in Social Psychology Quarterly

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research supported with a grant from the National Science Foundation SES-1756853 to Cathryn Johnson and Karen A. Hegtvedt.

ORCID iD

Cathryn Johnson

Supplemental Material

Supplemental material for this article is available online.

1

We focus on positive and negative actions of authorities that are likely to evoke emotions. We appreciate a reviewer’s comment pointing out that under certain circumstances, negative emotional responses may spur positive workplace changes, and constant displays of positive emotions may be counterproductive (particularly if workers manage emotions without acknowledging negative ones).

2

Although expectation states theory and Ridgeway and Berger’s (1986) extension regarding the legitimation of status hierarchies applies to groups (where group members are task and collectively oriented), research (e.g., ) demonstrates the anticipated effects of status characteristics even outside those scope conditions.

3

See , available with the online version of the article.

4

Dynata maintains panels of survey respondents recruited online. Recruitment is intended to approximate the U.S. population in terms of age, income, gender, education, and geography. Amengual and Bartley (2022) note that opt-in samples perform as well as population-based samples in survey experiments. And Ahler, Roush, and Sood (2021) show that Dynata’s sample has fewer issues regarding misrepresentation than samples drawn from Amazon MTurk.

5

Leader legitimacy involves perceptions of who is the right and proper person for the position and perceived support (Zelditch and Walker 1984). Support, however, may also be conceived as an outcome of legitimacy. Therefore, we used only right and proper items and found no difference in patterns of results in both studies (analyses available on request). In addition, propriety refers to an individual’s beliefs about what is proper and appropriate; endorsement refers to beliefs that peers and subordinates support the leader (; Zeldtich and Walker 1984). We ran our analyses separately for propriety and endorsement and found consistent substantive patterns in both studies (analyses available on request). This is not surprising given that respondent’s own perceptions should influence what they believe other employees think about the manager in a vignette study.

6

Although SEM is well suited to our analyses, our approach to mediation nonetheless relies on a series of correlational tests and thus does not represent causality. Although our models provide empirical support for our causal theorizing, our analysis falls short as a formal causal assessment of the impact of perceived legitimacy on emotions. The effects of the manipulated behaviors on both legitimacy and emotions can be described in causal terms, but the mediating sequence remains correlational in both studies (e.g., Spencer, Zanna, and Fong 2005).

7

For a correlation matrix for each study, see Supplemental File, Part B, available with the online version of the article.

8

See Supplemental File, Part A, available with the online version of the article.

9

When we designed Study 2, the Gaddis (2017) study led us to reason that if participants correctly recalled the name of their manager, they likely also noticed the manager’s race. In addition, some audit studies (i.e., Gaddis 2015), relied on “applicant” names to signal race (although the nature of the study prevented them from asking respondents to recall the race of the “applicant”). Pedulla (2014), however, in his vignette study, asked about the race of applicant at the end of a survey as a manipulation check. We thought that asking about the manager’s race directly would raise suspicion, but we acknowledge that not doing so is a study limitation. In the absence of such a manipulation check, we also ran the analysis with all respondents, that is, those who correctly and incorrectly identified the manager’s name. The patterns are substantively the same (see Supplemental File, Part C, available with the online version of the article).

Bios

Cathryn Johnson is a professor of sociology at Emory University. Her work on legitimacy, identity, and emotions processes has been published in American Sociological Review, American Journal of Sociology, Annual Review of Sociology, Social Forces, Social Psychology Quarterly, and other journals. Collaborative work with Karen Hegtvedt examines how leaders gain legitimacy and antecedents to perceptions of environmental justice. She previously coedited Social Psychology Quarterly and is the coauthor of Social Psychology: Individuals, Interaction, and Inequality (2018).

Ryan Gibson is an assistant professor of sociology at the University of New Hampshire and a faculty fellow at the Carsey School of Public Policy. Dr. Gibson’s research examines how humans perceive fairness, focusing on how racial and ethnic differences influence perceptions of justice, equity, and inclusion. Dr. Gibson’s work in this area has been published in journals such as Ethnic and Racial Studies, Social Psychology Quarterly, and Social Forces. Ongoing projects include applications of justice research across various domains, including business, health, the environment, and criminal justice.

Kate Hawks is a postdoctoral associate at New York University Abu Dhabi. Her work falls at the intersection of social psychology, health, inequality, and culture. Current projects examine the antecedents and consequences of the perceived legitimacy of public health authorities in the context of the COVID-19 pandemic and individual authorities in the workplace. Her work has appeared in Social Forces, Social Psychology Quarterly, and Advances in Group Processes.

Karen A. Hegtvedt is a professor of sociology at Emory University. Her work, focusing on perceptions of and responses (behavioral and emotional) to injustice, has appeared in Social Forces, Social Psychology Quarterly, Annual Review of Sociology, and other journals. Collaborative work with Cathryn Johnson examines the antecedents to legitimacy gains by authorities in the workplace and factors affecting perceptions of environmental justice. She previously coedited Social Psychology Quarterly and is the coauthor of Social Psychology: Individuals, Interaction, and Inequality (2018).

References

Acock

Alan C.

2013. Discovering Structural Equation Modeling Using Stata. College Station, TX: Stata Press.

Aguinis

Herman

Bradley

Kyle J.

2014. “Best Practice Recommendations for Designing and Implementing Experimental Vignette Methodology Studies.” Organizational Research Methods 17(4):351–71.

Ahler

Douglas J.

Roush

Carolyn E.

Sood

Gaurav

. 2021. “The Micro-task Market for Lemons: Data Quality on Amazon’s Mechanical Turk.” Political Science Research and Methods. doi:10.1017/psrm.2021.57.

Amengual

Matthew

Bartley

Tim

. 2022. “Global Markets, Corporate Assurances, and the Legitimacy of State Intervention: Perceptions of Distant Labor and Environmental Problems.” American Sociological Review 87(3):383–414.

Baldassarri

Delia

Grossman

Guy

. 2011. “Centralized Sanctioning and Legitimate Authority Promote Cooperation in Humans.” Proceedings of the National Academy of Sciences 108(27):11023–27.

Barclay

Laurie J.

Kiefer

Tina

. 2014. “Approach or Avoid? Exploring Overall Justice and the Differential Effects of Positive and Negative Emotions.” Journal of Management 40(7):1857–98.

Barkworth

Julie M.

Murphy

Kristina

. 2015. “Procedural Justice Policing and Citizen Compliance Behaviour: The Importance of Emotion.” Psychology, Crime & Law 21(3):254–73.

Barsade

Sigal G.

Gibson

Donald E.

2007. “Why Does Affect Matter in Organizations?” Academy of Management Perspectives 21(1):36–59.

Bentler

Peter M.

1990. “Comparative Fit Indexes in Structural Models.” Psychological Bulletin 107(2):238–46.

10.

Berger

Joseph M.

Rosenholtz

Susan J.

Zelditch

Morris

Jr.

1980. “Status Organizing Processes.” Annual Review of Sociology 6:479–508.

11.

Blau

Peter M.

1964. Exchange and Power in Social Life. New York, NY: John Wiley & Sons.

12.

Burnham

Kenneth P.

Anderson

David R.

2004. “Multimodel Inference.” Sociological Methods & Research 33(2):261–304.

13.

Cohen

Bernard P.

Zhou

Xueguang

. 1991. “Status Processes in Enduring Work Groups.” American Sociological Review 56:179–88.

14.

Collett

Jessica L.

Childs

Ellen

. 2011. “Minding the Gap: Meaning, Affect, and the Potential Shortcomings of Vignettes.” Social Science Research 40:513–22.

15.

Cook

Karen S.

Emerson

Richard M.

1978. “Power, Equity and Commitment in Exchange Networks.” American Sociological Review 43(5):721–39.

16.

Correll

Shelley J.

2004. “Constraints into Preferences: Gender, Status, and Emerging Career Aspirations.” American Sociological Review 69:93–113.

17.

Cullen

Kristen L.

Edwards

Bryan D.

Camron Casper

Wm.

Gue

Kevin R.

2014. “Employee’s Adaptability and Perceptions of Change-Related Uncertainty: Implications for Perceived Organizational Support, Job Satisfaction, and Performance.” Journal of Business Psychology 29:269–80.

18.

De Cremer

David

. 2004. “The Influence of Accuracy as a Function of Leader’s Bias: The Role of Trustworthiness in the Psychology of Procedural Justice.” Personality and Social Psychology Bulletin 30:293–304.

19.

Dulebohn

James H.

Boomer

William H.

Liden

Robert C.

Brouer

Robyn

Ferris

Gerald R.

2012. “A Meta-analysis of the Antecedents and Consequences of Leader-Member Exchange: Integrating the Past with an Eye toward the Future.” Journal of Management 38:1715–59.

20.

Ford

Rebecca

Johnson

Cathryn

. 1998. “The Perception of Power: Dependence and Legitimacy in Conflict.” Social Psychology Quarterly 61:16–32.

21.

Foschi

Martha

. 2000. “Double Standards for Competence: Theory and Research.” Annual Review of Sociology 26:21–42.

22.

Gaddis

S. Michael

. 2015. “Discrimination in the Credential Society: An Audit Study of Race and College Selectivity in the Labor Market.” Social Forces 93:1451–79.

23.

Gaddis

S. Michael

. 2017. “How Black Are Lakisha and Jamal? Racial Perceptions from Names Used in Correspondence Audit Studies.” Sociological Science 4:469–89.

24.

Hatfield

Elaine

Cacioppo

John T.

Rapson

Richard L.

1992. “Primitive Emotional Contagion.” Pp. 151–77 in Emotion and Social Behavior, edited by Clark

M. S.

Newbury Park, CA: Sage.

25.

Hegtvedt

Karen A.

Johnson

Cathryn

. 2009. “Power and Justice: Toward an Understanding of Legitimacy.” American Behavioral Scientist 53(3):376–99.

26.

Hegtvedt

Karen A.

Johnson

Cathryn

Gibson

Ryan

Hawks

Kate

Hayward

Jennifer L.

2022. “Power and Procedure: Gaining Legitimacy in the Workplace.” Social Forces 101(1):176–201.

27.

Hegtvedt

Karen A.

Parris

Christie

. 2014. “Emotions in Justice Processes.” Pp. 103–26 in Handbook of Sociology of Emotions. Vol. 2, edited by Stets

Turner

New York, NY: Springer.

28.

Hochschild

Arlie Russell

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

29.

Huq

Aziz Z.

Jackson

Jonathan

Trinkner

Rick

. 2017. “Legitimating Practices: Revisiting the Predicates of Police Legitimacy.” British Journal of Criminology 57:1101–22.

30.

Johnson

Cathryn

Dowd

Timothy J.

Ridgeway

Cecilia L.

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

31.

Johnson

Cathryn

Ford

Rebecca

Kaufman

Joanne M.

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

32.

Johnson

Cathryn

Hegtvedt

Karen A.

Khanna

Nikki

Scheuerman

Heather L.

2016. “Legitimacy Processes and Emotional Responses to Injustice.” Social Psychology Quarterly 79:95–114.

33.

Kalkhoff

Will

. 2005. “Collective Validation in Multi-actor Task Groups: The Effects of Status Differentiation.” Social Psychology Quarterly 68:57–74.

34.

Kanter

Rosabeth Moss

. 1977. Men and Women of the Corporation. New York, NY: Basic Books.

35.

Karren

Ronald J.

Woodard Barringer

Melissa

. 2002. “A Review and Analysis of the Policy-Capturing Methodology in Organizational Research: Guidelines for Research and Practice.” Organizational Research Methods 5(4):337–61.

36.

Kawakami

Kerry

Dunn

Elizabeth

Karmali

Francine

Dovidio

John F.

2009. “Mispredicting Affect and Behavioral Responses to Racism.” Science 323:276–78.

37.

Kemper

Theodore

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

38.

Kenny

David A.

Kashy

Deborah A.

Bolger

Niall

. 1998. “Data Analysis in Social Psychology.” Pp. 233–65 in The Handbook of Social Psychology. Vol. 1, edited by Gilbert

Fiske

Lindzey

Boston, MA: McGraw-Hill.

39.

Lawler

Edward J.

2006. “The Affect Theory of Social Exchange.” Pp. 244–67 in Contemporary Social Psychological Theories, edited by Burke

P. J.

Stanford, CA: Stanford University Press.

40.

Lovaglia

Michael J.

Houser

Jeffrey A.

1996. “Emotional Reactions and Status in Groups.” American Sociological Review 61:867–83.

41.

Manago

Bianca

Sell

Jane

Goar

Carla

. 2019. “Groups, Inequality, and Synergy.” Social Forces 97:1365–88.

42.

Melamed

David

Munn

Christopher W.

Barry

Leanne

Montgomery

Bradley

Okuwobi

Oneya F.

2019. “Status Characteristics, Implicit Bias, and the Production of Racial Inequality.” American Sociological Review 84:1013–36.

43.

Molm

Linda D.

2006. “The Social Exchange Framework.” Pp. 24–45 in Contemporary Social Psychological Theories, edited by Burke

P. J.

Stanford, CA: Stanford University Press.

44.

Pedulla

David S.

2014. “The Positive Consequences of Negative Stereotypes: Race, Sexual Orientation, and the Job Application Process.” Social Psychology Quarterly 77(1):75–94.

45.

Rafaeli

Anat

Sutton

Robert I.

1989. “The Expression of Emotion in Organizational Life.” Research in Organizational Behavior 11:1–42.

46.

Ridgeway

Cecilia L.

2019. Status: Why Is It Everywhere? Why Does It Matter? New York, NY: Russell Sage Foundation.

47.

Ridgeway

Cecilia L.

Berger

Joseph

. 1986. “Expectations, Legitimation, and Dominance Behavior in Task Groups.” American Sociological Review 51:603–17.

48.

Ridgeway

Cecilia L.

Johnson

Cathryn

. 1990. “What Is the Relationship between Socioemotional Behavior and Status in Task Groups?” American Journal of Sociology 95:1189–212.

49.

Ridgeway

Cecilia L.

Johnson

Cathryn

Diekema

David

. 1996. “External Status, Legitimacy, and Compliance in Male and Female Groups.” Social Forces 72:1051–77.

50.

Ridgeway

Cecilia L.

Nakagawa

Sandra

. 2014. “Status.” Pp. 3–25 in Handbook of the Social Psychology of Inequality, edited by McLeod

Lawler

Schwalbe

Dordrecht, the Netherlands: Springer.

51.

Scholl

Annika

Sassenberg

Kai

Scheepers

Daan

Ellemers

Naomi

de Wit

Frank

. 2017. “A Matter of Focus: Power-Holders Feel More Responsible after Adopting a Cognitive Other-Focus, Rather Than a Self-Focus.” British Journal of Social Psychology 56:89–102.

52.

Sinaceur

Marwan

Tiedens

Larissa Z.

2006. “Get Mad and Get More Than Even: When and Why Anger Expression Is Effective in Negotiations.” Journal of Experimental Social Psychology 42:314–22.

53.

Spencer

Steven J.

Zanna

Mark P.

Fong

Geoffrey T.

2005. “Establishing a Causal Chain: Why Experiments Are Often More Effective Than Mediational Analyses in Examining Psychological Processes.” Journal of Personality and Social Psychology 89:845–51.

54.

Steiger

James H.

1990. “Structural Model Evaluation and Modification.” Multivariate Behavioral Research 25:173–80.

55.

Thye

Shane R.

Harrell

Ashley

. 2017. “The Status Value Theory of Power and Mechanisms of Micro Stratification: Theory and New Experimental Evidence.” Social Science Research 63:54–66.

56.

Tucker

Ledyard R.

Lewis

Charles

. 1973. “A Reliability Coefficient for Maximum Likelihood Factor Analysis.” Psychometrika 38(1):1–10.

57.

Turner

Jonathan H.

Stets

Jan E.

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

58.

Tyler

Tom R.

1997. “The Psychology of Legitimacy: A Relational Perspective on Voluntary Deference to Authorities.” Personality and Social Psychology Review 1:3223–45.

59.

Tyler

Tom R.

2001. “A Psychological Perspective on the Legitimacy of Institutions and Authorities.” Pp. 416–36 in The Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, and Intergroup Relations, edited by Jost

Major

Cambridge, UK: Cambridge University Press.

60.

Tyler

Tom R.

Allan Lind

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

61.

Vial

Andrea C.

Napier

Jaime L.

Brescoll

Victoria L.

2016. “A Bed of Thorns: Female Leaders and the Self-Reinforcing Cycle of Illegitimacy.” The Leadership Quarterly 27:400–14.

62.

Weber

Max

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

Wittich

Berkeley: University of California Press.

63.

Webster

Murray

Jr. Slattery Walker

Lisa

. 2014. “Emotions in Expectation States Theory.” Pp. 127–53 in Handbook of the Sociology of Emotions. Vol. 2, edited by Stets

J. E.

Turner

J. H.

New York, NY: Springer.

64.

Wingfield

Adia Harvey

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

65.

Yoon

Jeongkoo

Thye

Shane

. 2011. “A Theoretical Model and New Test of Managerial Legitimacy in Work Teams.” Social Forces 90:639–59.

66.

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

0.02 MB

0.36 MB