Article Critique Assignment Homework

r Academy of Management Journal 2018, Vol. 61, No. 1, 348–369. https://doi.org/10.5465/amj.2015.1088

THE MIXED BLESSING OF LEADER SENSE OF HUMOR: EXAMINING COSTS AND BENEFITS

KAI CHI YAM National University of Singapore

MICHAEL S. CHRISTIAN University of North Carolina Chapel Hill

WU WEI Wuhan University

ZHENYU LIAO Washington University in St. Louis

JARED NAI Singapore Management University

Workplace humor is ubiquitous, yet scholars know little about how it affects employees’ behaviors in organizations. We draw on an emerging psychological theory of humor— benign violation theory—to suggest that a leader’s sense of humor often conveys counter- normative social information in organizations. We integrate this theory with social information processing theory to develop hypotheses about the effects of a leader’s sense of humor on follower behavior. We suggest that although a leader’s sense of humor is positively associated with leader2member exchange and ultimately work engagement, it can also signal to followers the acceptability of norm violation at work. These per- ceptions in turn are positively associated with followers’ deviance. Furthermore, we propose that these indirect effects are moderated by leader aggressive humor. Data from two three-wave field studies in China and the United States provide support for our hypotheses. Taken together, our results suggest that a leader’s sense of humor can be a mixed blessing and elicit unforeseen negative behaviors from their followers.

A sense of humor1 is often said to be a critical component of successful leadership. Broadly de- fined, a sense of humor refers to a trait-like individual

tendency to use or display behaviors, attitudes, and abilities relating to amusement during social interac- tions (Martin, 2001). Successful leaders anecdotally often use their sense of humor to garner support, mo- tivate their followers, and even create lasting memo- ries for the world. For example, former California governor Arnold Schwarzenegger was hit by an egg in a rally by an angry protester. In his response to mul- tiple media outlets, he said “this guy owes me bacon now.” His humor eased tension and even gathered support from protesters. In the corporateworld, many successful CEOs are likewise praised for their sense of humor. When asked to advise junior analysts in his organization, Warren Buffet said “I try to buy stock in businesses that are so wonderful that an idiot can run them. Because sooner or later, one will.” Practitioners thus often credit increased employee satisfaction and performance to leaders’ senses of humor (Katz, 1996). Given the perceived effectiveness of humor, some

The last three authors contributed equally and thus are listed in an arbitrary order. Correspondence should be addressed to Wu Wei ([email protected]). We thank Associate Editor Amy Colbert and three anonymous re- viewers for their constructive feedback throughout the re- viewprocess.We also thank Scott J. Reynolds, Christopher Barnes, and Ryan Fehr for helpful feedback on an earlier draft of this manuscript. This research was supported by NUS start up grant WBS R-317-000-117-133, the National Natural Science Foundation of China under Grant 71272231, and the Social Science Foundation of the Min- istry of Education of China under Grant 17YJA630105.

1 Following past research on humor (Mesmer-Magnus, Glew, & Viswesvaran, 2012), in this paper we considered the terms humor and sense of humor to be synonymous.

348

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express written permission. Users may print, download, or email articles for individual use only.

political and business leaders have even hired humor coaches to improve their leadership effectiveness (Dampier & Walton, 2013).

We draw on benign violation theory (BVT; McGraw & Warren, 2010) to explain when and why a leader’s sense of humor influences the workplace in positive and negative ways. Briefly stated, BVT suggests that the display of humor often necessitates a benign norm violation. We integrate BVT with social information processing theory (Salancik & Pfeffer, 1978) to propose that a leader’s sense of hu- mor signals the acceptability of norm violations in the organization, leading followers to engage in in- creased workplace deviance, or “voluntary behavior that violates significant organizational norms, and in so doing, threatens thewell-being of the organization and/or its members” (Robinson & Bennett, 1995: 556). Moreover, our framework also extends past research onpositive outcomes of humor (e.g.,Mesmer- Magnus, Glew, & Viswesvaran, 2012). We argue that by violating benign norms in their interactions with followers, leader humor creates interpersonal rap- port by signaling relational openness andpermissive dispositional qualities, thus positively relating to the quality of leader2member exchanges (LMX) and follower work engagement. In other words, we sug- gest that leader sense of humor can have indirect effects on followers’ deviance and work engagement via perceived acceptability of norm violation and LMX respectively.

Furthermore, we provide a more complete un- derstanding of the “mixed blessing” of a leader’s sense of humor by examining the moderating role of the style of humor a leader tends to use. We con- ceptualize sense of humor as a “broad bandwidth” trait construct, and humor styles as more specific

“narrow facets” that represent within-person con- sistency in tendencies to enact a sense of humorwith certain patterns of behavior (see Moon, 2001). While having a sense of humor is valence free, a leader may have a specific tendency to express his or her sense of humor in interpersonally negative or posi- tiveways.Wehelp advance the literature by focusing on the tendency to use an aggressive humor style (e.g., Martin, Puhlik-Doris, Larsen, Gray, & Weir, 2003; Mesmer-Magnus et al., 2012), thus providing a more nuanced and precise view of the effects of a leader’s sense of humor on followers. Specifically, we theorize that the indirect relation of leader humor with follower deviance will be strengthened, and the indirect relation of leader humor with follower work engagement will be weakened, when a leader tends to use aggressive humor, a negatively valenced style of humor that is carried out “at the expense and detriment of one’s relationships with others” (Martin et al., 2003: 52); often referred to as teasing with a humorous undertone (Mesmer-Magnus et al., 2012).

We theorize that the effects of a leader’s sense of humor on followers’ perceptions will be moderated by aggressive humor style, and ultimately be asso- ciated with follower outcomes. The more aggressive a leader’s style, themore a sense of humorwill signal acceptability of norm violations, which will be pos- itively associated with deviance; in contrast, the more aggressive a leader’s style, the less a sense of humor will signal high quality LMX, which will be negatively associated with work engagement (Figure 1). We test our hypotheses with two three- wave field studies of full-time employees in China and the United States. In the first study, we examine our basic proposition of the indirect effect of leader

FIGURE 1 Theoretical Model of the Current Research

Leader sense of humor

Perceived acceptability of norm violation

Leader aggressive humor

Follower deviance

Leader member exchange

Work engagement

2018 349Yam, Christian, Wei, Liao, and Nai

sense of humor on follower deviance through signals of norm violation acceptability. In the second study, we extend these results by testing our full theoretical framework with a moderated mediation model and both positive and negative follower outcomes.

Our research makes a number of theoretical and practical contributions. First, humor research in organizations has been characterized as “sporadic” (Duncan, Smeltzer, & Leap, 1990), partly because the literature is fragmented and lacks an overarch- ing theoretical framework. Although scholars have written about leader humor, most research is theo- retical in nature (Cooper, 2005, 2008; Duncan, 1982; Malone, 1980). As a result, there is very little em- pirical research examining the effects of a leader’s sense of humor. As Crawford (1994: 54) noted “of all the communicative strategies that leaders utilize, the use of humor is most promising, but least un- derstood.” Leadership scholars consequently lack a complete understanding of a prevalent phenome- non in organizations, highlighting the need for new theoretical and empirical investigations. By introducing BVT (McGraw & Warren, 2010), which has been largely supported by multiple empirical studies in social psychology (e.g.,McGraw&Warner, 2014; McGraw & Warren, 2010; McGraw, Williams, & Warren, 2014; McGraw, Warren, Williams, & Leonard, 2012), our research moves the organiza- tional literature forward by providing a framework for understanding humor in organizations. Second, in contrastwith past research,we argue that humor is not a panacea, suggesting that although a leader’s sense of humor may increase work engagement among followers, it may also increase follower de- viance. Third, we further contribute to the literature of leader humor by demonstrating why and when these effects occur, both critical components for building and testing theory (Colquitt & Zapata- Phelan, 2007). In terms of why, we hypothesize that follower perceived acceptability of norm viola- tions is the mechanism underlying the negative as- sociation of leader sense of humor with deviance whereas heightened LMX underlies the positive as- sociation of leader sense of humor with work en- gagement. In terms of when, we identify the style of humor used—humor that is aggressive in nature— for which these effects will be the strongest or will dissipate. Finally, our findings shed light on the pros and cons of leader humor in the workplace, pro- viding practitioners a more nuanced understanding of this mixed blessing and a guide to leverage the positive outcomes of leader humor while minimiz- ing its negative impacts.

THEORY AND HYPOTHESIS DEVELOPMENT

Benign Violation Theory

McGraw and colleagues (McGraw&Warner, 2014; McGraw & Warren, 2010; McGraw et al., 2012) pro- posed BVT to explain what is perceived to be hu- morous. BVT suggests that humor is comprised of three interrelated components. First, a norm viola- tion must occur. The norm violation can come in forms of physical or symbolic violations, and can range from social to moral norm violations. For example, we often laugh when being tickled (i.e., a violation of a physical norm), or when hearing stories or jokes that violate an expected social norm (e.g., “what do dinosaurs and decent lawyers have in common? They are both extinct”). Second, the norm violation must be perceived as benign—or nonthreatening—in nature. For example, people generally laugh when tickled by loved ones (i.e., perceived as benign), but not when tickled by strangers (i.e., perceived as threatening and hence not funny). In addition, the lawyer joke above would likely not be perceived as humorous if it was actu- ally accompanied by images of dead lawyers. In other words, excessive norm violations that offend or threaten the perceiver can dampen humor. Third, humor requires the first two conditions to be inter- preted simultaneously (McGraw & Warren, 2010). Like all other theories, a critical boundary condition of BVT is that it does not attempt to explain all types of humor generation. Rather, BVT argues that the broadest domain of humor is produced by leveraging benign violations.

Empirical research has largely supported the pre- dictions made by BVT. For example, when partici- pants are temporally, socially, hypothetically, or spatially distant from the violations (i.e., high psy- chological distance), they tend to view those viola- tions as benign in nature and hence funny (McGraw et al., 2012). In a longitudinal study, for example, participants found jokes regarding the destruction of Hurricane Sandy to be not funny during the crisis (i.e., when psychological distance is close and hence malign violations), but progressively funnier as the passageof time assuages the threateningnature of the disaster by expanding the psychological distance between the participants and the event (McGraw et al., 2014). Neuroscientific studies have also pro- vided some support for BVT. For example, Goel and Dolan (2001) found that the juxtapositions of two mental sets (i.e., norm violation and benignity) lead to neurological activity in the ventral medial prefrontal cortex (a region that is associated with

350 FebruaryAcademy of Management Journal

affective responses) and the experience of humor. Beyond empirical studies, BVT has also been sup- ported by numerous anecdotes. For example, co- medians David Letterman and John Stewart were heavily criticized for joking too soon after the 9/11 terrorist attack.

Given that humor often requires norm viola- tions, displays of humor will likely signal to others that norm violations are socially acceptable in in- terpersonal interactions. This is particularly rele- vant in organizational settings, which are highly social environments where norms are constantly communicated, signaled, and learned (Bettenhausen & Murnighan, 1985). In the following sections, we integrate BVT with social information processing theory (Salancik & Pfeffer, 1978) to suggest that a leader’s sense of humor sends two important im- plicit messages to followers. First, it signals the acceptability of norm violations, which will be positively associated with follower deviance. Sec- ond, it signals relational openness and permissive- ness, which will be positively associated with high quality LMX that drives work engagement. Finally, we suggest that these mediated relationships are moderated by a particular humor style—aggressive humor.

Integrating Benign Violation and Social Information Processing Theories

Benign violation theory suggests that humor in- volves violations of norms. However, in order to understand the implications and outcomes of humor in organizational settings, BVT must be integrated with theories that are specific to the dynamics inherent in the workplace (Heath & Sitkin, 2001). According to social information processing theory (Salancik & Pfeffer, 1978), employees do not operate in a vacuum in the workplace. Rather, employees actively seek to understand and behave in congru- ence with the norms and expectations within their organizationsbyprocessing surrounding social cues. From a social information processing perspective, leaders are seen as role models, or symbols of the “way things are done,” and guide the ways followers organize and make sense of their environment (Shamir, House, & Arthur, 1993). Thus, leaders’ ac- tions and interpersonal styles can send powerful messages and social cues to followers. Through processing these cues, followers learnwhat behavior is expected, rewarded, and punished in the organi- zation (Hogg, 2010). The processing of this in- formation also helps followers reduceuncertainty by

enabling them to make sense of the normative envi- ronment in the workplace (Van den Bos, 2001).

Social information processing suggests that fol- lowers create cognitive representations of leader behavior in specific interactions as indicators of the expectations and values of the organization. These cognitive representations—referred to as scripts (Gioia &Manz, 1985), schemas (James& Jones, 1976), or molar perceptions (Schneider, 1990)—are sym- bolic and general, rather than specific. As followers attempt to make meaning of their environment, they will look to leaders’ behaviors in specific situations as gestalt representations of basic principles to be applied across multiple situations (James, Hater, Gent, & Bruni, 1978). Thus, leader behavior sym- bolizes “the way things are around here” (Zohar & Luria, 2004: 322; see also Schneider & Reichers, 1983, Zohar, & Tenne-Gazit, 2008).

Thus, when integrated with benign violation the- ory, social information processing theory suggests that when leaders display a sense of humor in in- teractions with followers, the implications are be- yond simple mimicry (i.e., followers understand that humor is expected and rewarded and will be- have accordingly); a meaningful message is con- veyed about the values of the organization: behaving counter-normatively “is the way things are done.” We argue that leaders who display a sense of humor during interpersonal interactions with followers communicate two important signals. The first, that violating norms is acceptable, can have deleterious consequences in terms of follower deviance. The second, that the leader and followers are in a per- missive exchange relationship, can have positive effects on follower work engagement. We begin by discussing the path to deviance.

Implications for Norm Violation Acceptability and Workplace Deviance

In organizations, norms can be institutionalized as formal rules (e.g., organizational codes of conduct) or as informal perceived descriptive norms (e.g., be nice to your coworkers; Morris, Hong, Chiu, & Liu, 2015; Yam, Chen, & Reynolds, 2014). When leaders display humor and, as a result, violate norms, fol- lowers will likely perceive that it is socially accept- able to violate norms in the organization for two reasons. First, leaders’ formal position makes them strong sourcesof normative expectations. Leaders, as role models, are more likely to be observed by fol- lowers who are scanning the environment for in- formation on how to behave in the work context

2018 351Yam, Christian, Wei, Liao, and Nai

(Robinson & O’Leary-Kelly, 1998). In other words, leaders who make light of norm violations in order to produce humor are likely to imply to followers that mild norm violations in the organizations are generally acceptable. Second, when a leader acts in a humorous manner, others will likely react with laughter and amusement, an implicit signal of ap- proval (Billig, 2005). Followers will be likely to in- terpret this social information as signaling the acceptability of norm violations. When a norm vio- lation is enacted—and interpreted by others—in a playful, humorous way, it also signals to followers that violations need not be taken seriously or scru- tinized (Ford, 2000).

Extant empirical research provides some support for this hypothesis. For example, Ford,Wentzel, and Lorion (2001) found that men who were previously exposed to sexist humor later believe that there is an increased tolerance of violating sexism norms com- pared to men who were not exposed to such humor. In another study, participants who were exposed to jokes targeted at outgroups were more likely to en- dorse outgroup prejudice compared to participants in the control condition (Hodson, Rush, &MacInnis, 2010). Interestingly, both studies found that sexist or racist comments not conveyed in a humorous man- ner did not have the same effects, suggesting that humor expands the bound of appropriate behaviors by creating new norms that tolerate behaviors typi- cally viewed as deviant. We thus hypothesize:

Hypothesis 1. Leader sense of humor is posi- tively associated with followers’ perceived ac- ceptability of norm violations

Through its effects on perceived norm violation acceptability, leader humor is likely to increase fol- lower deviance. Workplace deviance, by definition, involves the violation of organizational norms that threaten the well-being of an organization and/or its members (Bennett & Robinson, 2000; Robinson & Bennett, 1995). Deviance can generally be classified along two dimensions—interpersonal and organi- zational deviance. Examples of interpersonal de- viance include making fun of someone at work and acting rudely toward coworkers, whereas examples of organizational deviance include taking property from work and coming in late to work without per- mission (Bennett & Robinson, 2000). Although the two dimensions of deviance focus on different targets, in this research we focus on overall work- place deviance. This approach is consistent with prior research (e.g., Christian & Ellis, 2011; Judge, Scott, & Ilies, 2006; Lee & Allen, 2002) as the two

dimensions are often highly correlated and have overlapping antecedents (Berry, Ones, & Sackett, 2007; Hershcovis et al., 2007).

As followers learn that it is acceptable to violate norms in an organization as a result of leader’s sense of humor, they are more likely to engage in deviant behavior. The key to this is the socially constructed perception that norm violations are acceptable. For example, although bribery violates the law in some Asian countries, it is a socially acceptedwayof doing business and people engage in this practice because theybelieve that suchbehavior iswidely tolerated by their peers, leaders, and organizations (Buffalo & Rodgers, 1971). Also consistent with this idea is that role models who display antisocial behavior have a stronger influence on antisocial behavior among other individuals in work groups than group mem- berswhoarenot considered rolemodels (Robinson& O’Leary-Kelly, 1998). Leaders are role models in or- ganizations given their formal position and status, and are thus likely to influence not only the per- ceptions (i.e., acceptability of norm violation), but also the behaviors (i.e., deviance) of their followers. In addition, to the extent that violating norms is perceived as socially acceptable, followers are more likely to engage in these behaviors because they be- lieve that thesebehaviorswill gounpunishedandare supported by their leaders. For example, employees are more likely to engage in unethical behaviors in organizations that have poorly conceived rules and community codes (Peterson, 2002; Yam, Reynolds, & Hirsh, 2014). We thus hypothesize:

Hypothesis 2. The perceived acceptability of norm violations mediates the relationship be- tween leader sense of humor and follower work- place deviance.

STUDY 1

Participants and Procedure

We contacted 340 part-time Master of Business Administration (MBA) students to participate in this research. All of our participants worked full-time andwere enrolled in aMBAprogrampart-time in the evenings or weekends at a large university in Central China. In our initial contactwith the participants, we provided a general overview of the research (e.g., a three-wave study, organizational behavior research) but didnot disclose any specific researchhypotheses to participants. A total of 215 participants (Mage 5 30.75, Mtenure with leader 5 3.44 years, 56.9% male) completed all three waves of the surveys, yielding

352 FebruaryAcademy of Management Journal

a response rate of 63.24%. This relatively high re- sponse rate was due to consistent communication between the author team and the participants, as well as monetary compensation. Our participants came from a variety of industries, most notably from the banking, finance, manufacture, and government sectors.

Each of the three waves was separated by roughly two weeks. At Time 1, participants completed a mea- sure of leader sense of humor. At Time 2, participants completed a measure of perceived acceptability of norm violation. At Time 3, participants self-reported their own interpersonal and organizational devi- ance. All survey items were translated from English to Mandarin, and then back translated following estab- lished best practices for survey translation (Brislin, 1970). Each participant was compensated with 20 RMB (3 USD) for their time per survey.

Measures

Leader sense of humor. We measured leader sense of humorwith a seven-item scale adapted from Thorson and Powell (1993). Since the original scale was designed formeasuring recognition of oneself as a humorous person, we reworded the items to reflect perceptions of others’ sense of humor for the purpose of this study. A sample item is “my leader uses hu- mor to entertain coworkers.”All of the items for this scale are listed in the Appendix (1 5 strongly dis- agree to 7 5 strongly agree; a 5 .96).2

Perceived acceptability of norm violations. We measured perceived acceptability of norm viola- tions with a five-item scale developed by Van Kleef, Homan, Finkenauer, Gündemir, and Stamkou (2011). Participants were asked to what extent they thought it acceptable for a person in the organiza- tions to be “asocial,” “immoral,” “improper,” “rude,” and “well-mannered (reverse-coded)” (1 5 not at all acceptable to 75highly acceptable;a5 .77).

The five adjectives were chosen because they repre- sent broad social norms observed in daily lives and across many different organizations.

Deviance. We measured deviance with a well- established 19-item scale (Bennett & Robinson, 2000). We used this particular scale to measure de- viance because it has been used in prior research with Chinese participants (Fehr, Yam, He, Chiang, & Wei, 2017), ensuring high content validity. Partici- pants self-reported on both interpersonal (e.g., “made fun of someone at work”) and organization deviance (e.g.,“takenproperty fromworkwithoutpermission”) that theyhad committed on a frequency scale (15 not at all to 55 quite often,a5 .96).Weused a self-report rather than an other-report measure of deviance be- cause many workplace deviant behaviors are often done in private without the knowledge of coworkers or leaders. In addition, deviance is commonly self- reported in prior research (for a review, see Berry et al., 2007).3

RESULTS

Preliminary Analyses

Descriptive statistics for Study 1 are presented in Table 1.

Prior to hypothesis testing, we conducted a series of confirmatory factor analyses (CFAs) to ensure the distinct factor structure of our three key variables. The hypothesized three-factor model composed of leader sense of humor, perceived acceptability of norm violations, and deviance demonstrated a good fit to the data, x2 (116)5 310.85, RMSEA5 .08, CFI5 .94. This three-factor model was also superior to al- ternative models, including a two-factor model in which leader sense of humor and deviance were set to load on a single factor (Δx2(2) 5 765.78, p , .01, RMSEA 5 .19, CFI 5 .71); a two-factor model in which perceived acceptability of norm violations and deviance were set to load on a single factor (Δx2(2) 5 674.34, p , .01, RMSEA 5 .19, CFI 5 .74); a two-factor model in which leader sense of humor and perceived acceptability of norm violations were set to load on a single factor (Δx2(2)5 654.01, p, .01, RMSEA 5 .18, CFI 5 .75).

2 To provide empirical evidence for the valence-free nature of our leader sense of humor scale,we conducted an MTurk study with 100 participants. Specifically, we pre- sented all seven items to them in random order. After reading each item, participants were asked “Please recall the most recent incident in which someone does this at your workplace. To what extent would you consider the type of humor he/she used as: –2 5 negative, 0 5 neutral, 2 5 positive.” A one-sample t-test (comparing the mean to zero) suggested that these behaviors are neither negative nor positive (M5 .07, t [99]5 1.32, p5 .19). These results sug- gest that our leader sense of humor scale is valence-free.

3 We tested ourhypotheses by treating interpersonal and organizational deviance as a single construct for pre- sentational parsimony in both Studies 1 and 2. Results remained identical when the two types of deviance were treated as two separated dependent variables.

2018 353Yam, Christian, Wei, Liao, and Nai

FollowingBecker (2005),we ran all of our analyses with and without demographic controls and the re- sults were essentially identical with the inclusion of these variables. Moreover, past humor research has failed to find meaningful differences in terms of a sense of humor across gender or age (Moran, Rain, Page-Gould, & Mar, 2014; Thorson & Powell, 1993). For the purpose of presentational parsimony, we thus present the results without controls but provide their bivariate correlations with the study variables in Table 1.

Tests of Hypotheses

We used ordinary least squares regression to test Hypothesis 1. Leader sense of humor at Time 1 was positively associatedwith perceived acceptability of normviolations at Time 2 (adjustedR25 .05,b5 .23, p, .01, Table 2). To test Hypothesis 2,we conducted a bootstrapping-based mediation test using the PROCESSmacro (Hayes, 2013). This procedure is an extension of the Sobel test (Sobel, 1982) and is rec- ommended over alternative procedures (e.g., Baron & Kenny, 1986) because it does not assume a normal sampling distribution of indirect effects, and simu- lation studies have shown that it is more valid and

statistically powerful than traditional methods (MacKinnon, Lockwood, & Williams, 2004). As Preacher and Hayes (2008) recommended, we esti- mated the indirect effect of leader sense of humor on follower deviance via perceived acceptability of norm violation using unstandardized coefficients and a bootstrapping procedure with 5,000 resam- ples to produce a 95% confidence interval around the estimated indirect effects. The bootstrapped indirect effect is significant if the bias-corrected 95% confidence interval (CI) excludes zero. Results revealed that leader sense of humor was associated with increased follower deviance, mediated by perceived acceptability of norm violation in the organization (indirect effect5 .04, SE5 .02, 95%CI 5 .01 to .09; direct effect 5 .08, SE 5 .03, 95% CI 5 .01 to .15; total effect5 .12, SE5 .03, 95%CI5 .05 to .19). Together, these results provide support for Hypotheses 1 and 2.

Study 1 Discussion

In Study 1,we found support for Hypotheses 1 and 2 by conducting a three-wave field study in China. Leader sense of humor, perceived acceptability of norm violations, and deviance were all temporally

TABLE 1 Correlation Table for Study 1

Variables Means (SD) 1 2 3 4 5

1. Leader sense of humor (T1) 4.42 (1.57) .96 2. Acceptability of norm violations (T2) 1.89 (1.18) .23** .77 3. Deviance (T3) 1.57 (.84) .22** .33** .96 4. Follower age (T1) 30.75 (5.54) .01 .05 –.05 (–) 5. Follower gendera (T1) 1.57 (.50) .10 .15* .13 .16* (–)

*p , .05 **p , .01

a 15 Female, 2 5Male Note: Alphas are presented on the diagonal.

TABLE 2 Regression Analyses for Study 1

DV5 Perceived acceptability of norm violation DV5 Deviance

Variables B SE B b B SE B b

Leader sense of humor .17 .05 .23** .08 .04 .15* Perceived acceptability of norm violation .21 .05 .30** Adjusted R2 .05 .12

*p , .05 **p , .01

354 FebruaryAcademy of Management Journal

separated in order to reduce common method effects (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Nevertheless, Study 1 is incomplete in that it only tested the negative path of leader sense of hu- mor to follower outcomes. In the following sections, we argue that humor can be a “mixed blessing ,” suggesting that although leader sense of humor might be positively associated with follower de- viance, it can also lead to positive follower out- comes such as LMX and work engagement. We further argue that a leader’s humor style moderates each path. We then conducted a second study to examine the full model.

A Mixed Blessing: Implications for LMX and Work Engagement

Social information processing theory and BVT also suggest a positive path for leader humor for three reasons. First, humor reduces social distance between a leader and followers (Graham, 1995; Mesmer-Magnus et al., 2012). Leaders who use hu- mor implicitly send a signal that they deemphasize the hierarchical difference between themselves and followers because they are willing to violate the normal hierarchical system in organizations (Cooper, 2008). Second, when leaders interact in a manner that violates norms, this signals that the leader ispermissive—acceptingof counter-normative behavior—with their followers, thus signaling that their relationship with particular followers is un- complicated, playful and open. Indeed, research in behavioral ethics suggests that leaderswho aremoral, thus less likely to violate ethical norms, are perceived to be less warm and permissive (Wellman, Mayer, Ong, & DeRue, 2016). Third, a leader who displays humor is demonstrating a willingness to be vul- nerable because they are openly violating norms, and thus likely seem less guarded and more open during social exchanges with followers. By model- ing vulnerability, a leader implies that the norma- tive interaction between the leader and follower is one that is amicable and safe. Followers who see their leaders as de-emphasizing hierarchy, permis- sive, and willing to display vulnerability may per- ceive the leader to be more relationship-oriented (Cooper, 2008; Decker & Rotondo, 2001), leading to the sense that the leader2follower relationship is of high quality. As such, leader humor is perceived to have positive relational intent and serves to main- tain positive work relationships with followers (Gkorezis, Petridou, & Xanthiakos, 2014; Pundt & Venz, 2017).

More generally, extant research suggests that people who exhibit a sense of humor are more en- joyable to be around as they increase the positive affect of people around them. For example, humor- ous people enjoy more successes in forming roman- tic relationships (Bressler & Balshine, 2006) and friendships (Kalbfleischl, 2013). Specific to the workplace, humor has been theorized to be an ef- fective ingratiation tactic for gaining social capital and establishingpositive interpersonal relationships (Cooper, 2005). As a result, we suggest that leaders who display a sense of humor around their followers will create higher levels of LMX (Robert, Dunne, & Iun, 2016), which is an indication of the relationship quality between a leader and follower (Graen & Scandura, 1987; Graen & Uhl-Bien, 1995).

Hypothesis 3. Leader sense of humor is posi- tively associated with LMX.

Through its effects on LMX,we suggest that leader humor is likely to increase follower work engage- ment. Work engagement is defined as a relatively enduring state of mind referring to the simultaneous investment of personal energies in work perfor- mance (Christian, Garza, & Slaughter, 2011). En- gagement is a conceptually and empirically unique motivational construct that is evoked when an em- ployee is able to develop a feeling of self-investment of their personal resources at work (Kahn, 1990). Work engagement thus involves a holistic use of cognitive, emotional, and physical energy (Rich, Lepine, & Crawford, 2010), and empirical studies suggest that follower work engagement can be cul- tivated by leader behavior, including LMX (Matta, Scott, Koopman, & Conlon, 2015; for a meta-analytic review, see Christian et al., 2011).

In order to personally invest their full energy at work, an employeemust feel comfortable and safe to express themselves (Kahn, 1990; May, Gilson, & Harter, 2004). A high quality of LMX as a result of a leader’s sense of humor will ensure employees feel safe to be their true self, which in turn enables them to fully invest their personal energy in their work roles. Furthermore, when a leader displays humor, the resultant high quality LMX provides social sup- port and reduces the negative effects of stress. In- deed, humor is associated with an optimistic reinterpretation of negative events and can assist in coping via increased social support (Ziv, 1981). Al- though no studies have examined the role of leader humor on follower work engagement, there is ev- idence that humor might reduce burnout (Abel, 2002)—a close antipode to work engagement

2018 355Yam, Christian, Wei, Liao, and Nai

(e.g., Schaufeli, Salanova, González-Romá, & Bakker, 2002). In sum, we propose that leader humor will in- crease LMX quality and, through heightened LMX, fol- lowers would feel more engaged and invest their personal energy in their work.

Hypothesis 4. Leader sense of humor is posi- tively associated with follower work engage- ment, mediated by increased LMX.

The Moderating Role of Leader Aggressive Humor

Thus far, Hypotheses 124 provide an account of the effects of leader’s sense of humor on follower deviance and work engagement. Nonetheless, there are reasons to believe that these effects will partly depend on the specific style of humor that a leader displays. Integrating BVT and social information processing theorywith research ondifferent styles of humor (Chen & Martin, 2007; Martin et al., 2003; Vernon, Martin, Schermer, & Mackie, 2008), we ar- gue that leader aggressive humor will exacerbate the negative effects of a leader’s sense of humor on follower deviance and attenuate the positive ef- fects of a leader’s sense of humor on follower work engagement.

Whereas a sense of humor represents a general tendency to display any style of humor (Martin, 2001), traditional conceptualizations of aggres- sive humor refer to a more specific style of hu- mor that is aimed at teasing or ridiculing (Martin et al., 2003; also known as disparagement humor; Zillmann, 1983). In organizations, it is relatively rare to observe excessively aggressive humor aimed solely at ridiculing others (e.g., racist jokes aimed at a follower; Romero & Cruthirds, 2006). Rather, leaders often usemild aggressive humor in the workplace, which includes sarcasm, satire, or teasing to humorously convey disapproving in- formation to their followers (e.g., “John, you seem so busy” to sarcastically convey a message that John is not putting up enough effort). Similar to a leader’s sense of humor, in this research we do not focus on whether a leader uses aggressive humor toward a specific follower or not, but rather a leader’s general tendency in using this style of humor toward all followers of his/her workgroup.

We suggest that a leader’s sense of humor signals more severe norm violations when the leader’s hu- mor style is aggressive compared to other styles of humor. Displaying a sense of humor coupledwith an aggressive humor style effectively violates norms in two ways: it not only involves using humor—a

benign violation—but it also signals violation of so- cial norms of civility. Aggressive humor signals to followers that the accepted social norm of being re- spectful toward others is not important. Thus, a leader who has this kind of sense of humor and tends to be aggressive communicates (a) that violat- ing organizational norms is acceptable for reasons that we have outlined above, but added to this is (b) the signal that violating norms of “human decency” is acceptable. Theoretically, compared to other forms of humor, Martin et al. (2003) categorized ag- gressive humor as a form of hostile behavior. For example, sarcasm, a form of aggressive humor, is often associated with increased interpersonal con- flict in a work group because it conveys disapproval, contempt, and scorn humorously (Huang, Gino, & Galinsky, 2015; Zhang & Liao, 2015). Given its hos- tile interpersonal nature, we expect that the malign normviolation, signaled byaggressivehumor, builds on the benign violation of the nature of a leader’s sense of humor, exacerbating the perception of the severity of acceptability of violations of norms. Thus, we argue that a leader who has a strong sense of hu- mor and tends to use aggressive humor would signal to followers an even stronger perceived acceptability of norm violation in the workplace, which ulti- mately leads to more behavior that is deviant. We hypothesize:

Hypothesis 5. The indirect effect of leader sense of humor on follower deviance, via perceived acceptability of norm violation, is moderated by leader aggressive humor such that the indirect effect is stronger when leader aggressive humor is high, but weakens when leader aggressive humor is low.

Similarly,we suggest that a leaderwhohas a strong sense of humor and often uses aggressive humor is less likely to build effective leader2follower re- lationships, ultimately attenuating thepositive effect of a leader’s sense of humor on follower work en- gagement. Aggressive humor, even in its mild form, is often carried out at the expense of the leader’s re- lationship with his/her followers. At the dyadic level, even if an aggressive joke is perceived as hu- morous, it remains harmful to interpersonal re- lationships (Toplak & Katz, 2000). Research suggests that well-meaning remarks delivered in an aggres- sive form (e.g., saying “your work ethic is obviously poor” sarcastically to someone who clearly worked optimally) is often less well received compared to the same intent, but delivered with more sincere content (e.g., “yourwork ethic is obviously excellent”;

356 FebruaryAcademy of Management Journal

Pexman & Olineck, 2002). More generally, aggressive humor is perceived as more contemptuous (Dews & Winner, 1995) and is more likely to instigate conflict (Huang et al., 2015) than other forms of humor. As a result, the effects of leader sense of humor on LMX will be attenuated when a leader uses aggressive humor.

The attenuating effects of aggressive humor styles on the relationship between leader humor and fol- lower LMX may also reach beyond the targeted fol- lower, providing a social interpretation that affects behavior (Salancik & Pfeffer, 1978). Research has suggested that followers often react to their leaders negatively when leaders treat their peers poorly (Mitchell, Vogel, & Folger, 2015), even when the follower is well treated (Christian, Christian, Garza, &Ellis, 2012;He, Fehr,Yam,Long,&Hao, 2017). This streamof research suggests that the positive effects of leader sense of humor on interpersonal outcomes will be significantly reduced even by merely ob- serving leaders who use aggressive humor styles toward other followers, rather than experiencing leader aggressive humor first-hand. We suggest when a leader has a strong sense of humor, strong LMX is less likely to be formed if the leader tends to use aggressive humor. We thus hypothesize:

Hypothesis 6. The indirect effect of leader sense of humor on follower work engagement, via LMX, is moderated by leader aggressive humor such that the indirect effect is stronger when leader aggressive humor is low, but weakens when leader aggressive humor is high.

STUDY 2

Participants and Procedure

We contacted 700 full-time employees to partici- pate in this research through Qualtrics, a third-party online survey administration company in theUnited States (for recent examples of data collection using Qualtrics, see DeCelles, DeRue, Margolis, & Ceranic, 2012; Long, Bendersky, & Morrill, 2011). In our ini- tial contact with the participants, we provided a general overview of the research (e.g., a three-wave study, organizational behavior research) but did not disclose any specific research hypotheses to partic- ipants as in Study 1. We also informed participants that they must have daily interaction with their leaders in order to participate. A total of 288 partic- ipants agreed to participate in the study and 200 participants (Mage 5 43.26, Mtenure with leader 5 5.93 years, 40% male; 72.5% Caucasian, 10% Asian

American, 8.5% Hispanic American) completed all three waves of surveys, yielding a response rate of 28.57%. Most of these participants worked in sales (24.5%), banking and financial services (17.5%), and engineering (15%).

Each of the three waves was separated by roughly 10days.AtTime 1, participants completedmeasures of leader sense of humor and leader aggressive hu- mor. At Time 2, participants completed measures of perceived acceptability of norm violation and LMX. At Time 3, participants self-reported their own de- viance and work engagement. Each participant was compensated with $5 for their time per survey and received a $10 bonus if they completed all three surveys.

Measures

Leader sense of humor. We measured leader sense of humor as in Study 1 (15 strongly disagree to 7 5 strongly agree; a 5 .97).

Leader aggressive humor. We measured leader aggressive humor with an eight-item scale adapted fromMartin et al. (2003). Since the original scalewas designed for self-report rather than other-report, we reworded the items to reflect perceptions of the leaders’ aggressive humor for the purpose of this study. A sample item is “my leader’s sense of humor often offends others.” All of the items for this scale are listed in the Appendix (1 5 strongly disagree to 7 5 strongly agree; a 5 .76).

Perceived acceptability of norm violation. We measured perceived acceptability of norm violation as in Study 1 (1 5 strongly disagree to 7 5 strongly agree; a 5 .80).

Leader member exchange. We measured LMX with the well-established eight-item measure de- veloped by Bauer and Green (1996). Sample items include “I usually know how satisfied my leader is withme” and “I knowwhere I standwithmy leader” (15 strongly disagree to 75 strongly agree; a5 .96).

Deviance. We measured deviance as in Study 1 (15 strongly disagree to 75 strongly agree; a5 .91).

Work engagement. We used Rich et al. (2010)’s 18-item scale to measure work engagement. Work engagement is theorized to have three sub- dimensions—physical (e.g., “I work with intensity onmy job”), emotional (e.g., “I am enthusiastic inmy job”), and cognitive engagement (e.g., “Atwork, I pay a lot of attention to my job”). As our theorizing did not distinguish the three different types of engage- ment and because they were highly correlated (rs ranged from .72 to .87, ps, .01), we used the average

2018 357Yam, Christian, Wei, Liao, and Nai

score to form an overall composite of work engage- ment (1 5 strongly disagree to 7 5 strongly agree; a 5 .91).

RESULTS

Preliminary Analyses

Descriptive statistics for study variables and con- trols for Study 2 are presented in Table 3.

We again conducted a series of CFAs to ensure the distinct factor structure of our key variables. The hypothesized six-factor model (i.e., leader sense of humor, leader aggressive humor, perceived accept- ability of norm violation, LMX, deviance, and work engagement) demonstrated a good fit to the data, x2(309) 5 699.95, RMSEA 5 .08, CFI 5 .91, and was also superior to three alternative five-factor models, including a five-factormodel inwhich leader sense of humor and leader aggressive humor were set to load on a single factor (Δx2(5)5 202.21, p, .01, RMSEA5 .10, CFI5 .87); a five-factormodel inwhich LMX and perceived acceptability of norm violation were set to load on a single factor (Δx2(5) 5 763.03, p , .01, RMSEA 5 .14, CFI 5 .74); and a five-factor model in which deviance and work engagement were set to load on a single factor (Δx2(5) 5 1353.64, p , .01, RMSEA 5 .17, CFI5 .61). Finally, to ensure that our temporal separationdidnot affect the factor structure, we further compared our six-factor model to a three- factor model in which the variables were grouped by time (i.e., leader sense of humor and aggressive hu- mor; LMX and perceived acceptability of norm vio- lation; work engagement and deviance). Results

suggested that the six-factor model demonstrated su- perior fit to this model (Δx2(12) 5 2311.12, p , .01, RMSEA5 .21, CFI5 .40)

As with Study 1, we ran all of our analyses with and without demographic controls and the results were essentially identical with the inclusion of these variables (Becker, 2005). For the purpose of pre- sentational parsimony, we present the results with- out any controls.

Test of Hypotheses

As in Study 1, we conducted ordinary least squares regression to test Hypothesis 1. Leader sense of humor at Time 1 was positively associated with perceived acceptability of norm violations in the organization at Time 2 (b 5 .27, p , .01). To test Hypothesis 2, we conducted a bootstrapping-based mediation test using the PROCESS macro (Hayes, 2013). Results revealed that leader sense of humor was associated with increased follower deviance, mediated by perceived acceptability of norm viola- tion in the organization (indirect effect 5 .02, SE 5 .01, 95%CI5 .002 to .05;direct effect5 .03,SE5 .02, 95% CI 5 2.02 to .08; total effect 5 .05, SE 5 .02, 95% CI 5 .01 to .10). These results provide addi- tional support for Hypotheses 1 and 2.

To examine the positive pathwayof leader sense of humor and Hypotheses 324, we again conducted a bootstrapping-based mediation test using the PROCESS macro (Hayes, 2013). First, we found that leader sense of humor at Time 1 was positively as- sociated with LMX at Time 2 (b 5 .16, p , .05). Re- sults further revealed that leader sense of humorwas

TABLE 3 Correlation Table for Study 2

Variables Means (SD) 1 2 3 4 5 6 7 8 9

1. Leader sense of humor (T1) 3.88 (1.60) .97 2. Leader aggressive humor (T1) 3.40 (.98) .06 .76 3. Accept. of norm violations (T2) 2.22 (1.04) .28** .15* .80 4. LMX (T2) 5.14 (1.41) .15* –.12† –.04 .96 5. Deviance (T3) 1.48 (.55) .15* .05 .21** –.131 .91 6. Work engagement (T3) 5.49 (1.19) .23** –.04 .01 .19** –.14* .98 7. Follower age (T1) 43.26 (13.13) –.33** –.13† –.13† –.07 –.07 –.17* (–) 8. Follower gendera (T1) 1.60 (.49) –.03 .04 .12† –.11 –.12† –.04 –.07 (–) 9. Follower raceb (T1) 1.28 (.45) .07 .09 –.09 .00 –.09 .11 –.08 –.09 (–)

†p , .10 *p , .05

**p , .01 a 15 Male, 25 Female b 15 White, 25 Non-white

Note: Alphas are presented on the diagonal.

358 FebruaryAcademy of Management Journal

associated with increased work engagement, medi- ated by increased LMX (indirect effect 5 .02, SE 5 .02, 95%CI5 .003 to .06;direct effect5 .15,SE5 .05, 95% CI5 .05 to .26; total effect5 .17, SE5 .06, 95% CI 5 .07 to .28), providing support for Hypotheses 3 and 4.

To test Hypothesis 5, we began by examining the interactive effect of leader aggressive humor and leader sense of humor on perceived acceptability of normviolation. In Step 1, both leader sense of humor (b 5 .27, p , .01) and leader aggressive humor (b 5 .14, p , .05) were positively associated with per- ceived acceptability of norm violation (Table 4). In Step 2, results suggested that after the inclusion of the interaction term the model explained signifi- cantly more variance (adjusted R2 5 .13; ΔR2 5 .04, p, .01) and that the interaction termwas significant (b 5 .20, p , .01). To aid interpretation, we plotted the interaction effect in Figure 2 and it was in the expected direction. We then utilized the methods of Hayes (2013) to test Hypothesis 5 in an integrative fashion at one standard deviation above and below the mean of the moderator (i.e., leader aggressive humor).When leader aggressivehumorwashigh, the mediated model was significant (conditional in- direct effect 5 .03, SE 5 .02, 95% CI 5 .002 to .07). When leader aggressive humor was low, however, the mediated model was not significant (conditional indirect effect 5 .01, SE 5 .01, 95% CI 5 2.003 to .03). The index ofmoderatedmediationwas likewise significant (Index 5 .02, SE 5 .01, 95% CI 5 .001 to .04), providing full support forHypothesis 5. In other words, although leader sense of humor is generally associated with increased deviance, this effect is stronger when leader aggressive humor is high but dissipates when leader aggressive humor is low.

We followed the same procedure in testing Hy- pothesis 6. In Step 1, leader sense of humor (b5 .16, p , .05) was positively associated with LMX, whereas leader aggressive humor (b52.13, p, .10) was negatively (albeit marginally) associated with LMX (Table 5). In Step 2, results suggested that after the inclusion of the interaction term the model explained significantlymore variance (adjustedR25 .05; ΔR2 5 .02, p, .05) and that the interaction term was significant (b 5 .14, p , .05). To aid in- terpretation, we plotted the interaction effect in Figure 3 and it was in the expected direction. We then utilized the methods of Hayes (2013) to test Hypothesis 6 in an integrative fashion at one stan- dard deviation above and below the mean of the moderator (i.e., leader aggressive humor). When leader aggressive humor was high, the mediated

model was not significant (conditional indirect ef- fect52.00, SE5 .02, 95% CI52.03 to .04). When leader aggressive humor was low, however, the mediated model was significant (conditional in- direct effect 5 .04, SE 5 .02, 95% CI 5 .02 to .09). The index of moderated mediation was likewise sig- nificant (Index 5 2.02, SE 5 .01, 95% CI 5 2.06 to2.003), providing full support forHypothesis 6 (see Table 6). In other words, although leader sense of humor is generally associated with increased work engagement, this effect is stronger when leader ag- gressive humor is low but dissipates when leader aggressive humor is high. Together, our results sug- gested that leader sense of humor is indeed a mixed blessing and revealed the important moderating role of leader aggressive humor.4 For purposes of pre- sentational parsimony, we have also presented all results in a path model (see Figure 4).

GENERAL DISCUSSION

Across two three-wave field studies in China and the United States, we found consistent support for our hypotheses that leader sense of humor is amixed blessing. On the one hand, it signals to followers the acceptability of normviolations in the organizations, which in turn is positively associated with follower deviance. On the other hand, it increases LMX, which in turn is positively associated with follower work engagement. We further demonstrated that these mediated effects were moderated by leaders’ humor styles. Specifically, humorous leaders who tend to use aggressive humor were most likely to promote follower deviance and least likely to en- courage follower work engagement. Below, we dis- cuss the theoretical and practical contributions of our work and suggest future research directions.

Theoretical Implications

Our research makes a number of important theo- retical contributions to the literature on leadership andhumor.Humor is pervasive in theworkplace, yet research on humor in organizations has almost been non-existent for several decades. In fact, the same could be said in other disciplines such as psychology

4 We controlled for LMX and work engagement when testing Hypothesis 5 and controlled for perceived accept- ability of norm violation and deviance when testing Hypothesis 6. Adding these controls to the model did not affect the findings and hence we presented the uncontrolled analyses for presentational parsimony.

2018 359Yam, Christian, Wei, Liao, and Nai

and sociology. For example, sociologist Murray Davis (1995) once characterized the study of humor as a “stillborn field.”We suspect that these sporadic research attemptswere partly due to a lack of a coherent theoretical framework. By introducing benign viola- tion theory (McGraw & Warren, 2010) to the organiza- tional literature, we believe that future researchers can systematically examine the effects of humor in work settings. Perhaps more importantly, whereas benign violation theory was originally developed to explain “what makes things funny,” we integrate this theory with social information processing theory and thus ex- tend BVT by applying it to understand the negative con sequences of observing and processing leader humor.

Our study also contributes to research on the relational nature of leadership. In particularly, our “mixed blessing” approach demonstrated that the effects of leader humor are more nuanced than previously assumed. By taking an initial step to- ward examining the negative outcomes of leader humor to their followers, we found that leader humor is associated with greater follower de- viance. By drawing from past work on the re- lational benefits of humor (e.g., Romero & Cruthirds, 2006), we found that leader humor is associated with increased follower work en- gagement. Thus, our results indicate that al- though leaders may improve the quality of their

TABLE 4 Regression Analyses for the Negative Path of Humor (Study 2)

DV5 Perceived acceptability of norm violation

Model 1 Model 2

Variables B SE B b B SE B b

LMX –.05 .05 –.07 –.03 .05 –.04 Leader sense of humor .18 .04 .27** .19 .04 .29** Aggressive humor .14 .07 .14* .13 .07 .12†

Leader sense of humor .13 .04 .20** X Aggressive humor adjusted R2 .09 13 DR2 .04**

†p , .10 *p , .05

**p , .01

FIGURE 2 The Interactive Effect of Leader Sense of Humor and Aggressive Humor on Acceptability

of Norm Violation (Study 2)

1

2

3

4

5

6

7

A cc

ep ta

bi li

ty o

f N

or m

V io

la ti

on

Low Aggressive Humor

High Aggressive Humor

High Leader Sense of HumorLow Leader Sense of Humor

360 FebruaryAcademy of Management Journal

relationships with followers and followers’ work engagement using humor, they may also un- intentionally increase perceptions of the accept- ability of norm violations and deviance. Our research thus provides a more complete un- derstanding of the effects of leader humor and highlights the need for a dialectical perspective on the consequences of humor usage in organizations.

Third, our study extends the extant literature on workplace humor by examining style. Whereas past research on leader humor often aggregated various types of humor (Avolio, Howell, & Sosik, 1999), we suggest that sense of humor is too broad to capture the nuances of outcomes, and that style of humor is

a more specific factor that should be considered in understanding the consequences of leader humor. As our results suggest, whereas sense of humor was associated positively with deviance and work en- gagement (i.e., the two mediated effects), aggressive humoramplifies ornullifies sucheffects. In addition, past research on specific humor styles tends to focus on positive forms of humor (for a review, see Mesmer-Magnus et al., 2012) and our work is one of the few that establishes the importance of a negative form of humor, aggressive humor, as a moderator between the links of leader sense of humor and fol- lower outcomes. Therefore, our work contributes to thehumor literaturebroadlybyexploring tendencies

TABLE 5 Regression Analyses for the Positive Path of Humor (Study 2)

DV 5 LMX

Model 1 Model 2

Variables B SE B b B SE B b

Perceived acceptability of norm violation –.10 .10 –.07 –.05 .10 –.04 Leader sense of humor .14 .06 .16* .13 .06 .15* Aggressive humor –.18 .10 –.13† –.17 .10 –.12†

Leader sense of humor .14 .06 .16* X Aggressive humor adjusted R2 .03 .05 DR2 .02*

†p , .10 *p , .05

FIGURE 3 The Interactive Effect of Leader Sense of Humor and Aggressive Humor on LMX (Study 2)

1

2

3

4

5

6

7

Low Leader Sense of Humor

L M

X Low Aggressive Humor

High Aggressive Humor

High Leader Sense of Humor

2018 361Yam, Christian, Wei, Liao, and Nai

to use humor in tandem with aggressive humor. We believe this approach can best inform scholars and practitioners when leader humor is most and least beneficial to their followers.

Finally, our study contributes to work on organi- zationalnorms (Morris et al., 2015) by suggesting that norm violations may be seen as domain general and therefore have far-reaching effects on behavior. In both of our studies, we found that leader humor was associated with a range of different deviant

behaviors that might not conceptually map on to the specific forms of leader humor. Although past work on humor and normative beliefs suggests that hu- morous norm violations may transfer within do- main, such that sexist jokes lead others to perceive that sexism is acceptable (e.g., Ford et al., 2001), our study extends this idea by suggesting that norm vi- olations may be interpreted more broadly than the specific norm being violated by a joke or a funny story. Leader humor may lead employees to form

TABLE 6 Summary of Indirect Effects and Conditional Indirect Effects

Paths & effects Estimates SE 95% confidence intervals

Leaders’displays of humor→Perceivedacceptability of norm violation→ Follower deviance

Indirect effects .02 .01 [.002, .051] Moderated mediation High aggressive humor .03 .02 [.003, .082] Low aggressive humor .01 .01 [–.003, .035]

Indirect difference .03 .02 [.002, .082] Leaders’ displays of humor→ LMX→ Work

engagement Indirect effects .02 .01 [.001, .060] Moderated mediation High aggressive humor –.00 .02 [–.032, .032] Low aggressive humor .04 .02 [.006, .103]

Indirect difference –.04 .03 [–.121,2.004]

FIGURE 4 Path Model of the Results (Study 2)

Leader sense of humor

Perceived acceptability of norm violation

Leader aggressive humor

Follower deviance

Leader member exchange

Work engagement .15*

.10†

.03 .01

.01 .19**

.13*

.13*

–.09

.10

.18**

–.14*

Unstandardized path estimates are reported. Solid lines depict the hypothesized relationships and dashed lines indicate relationships that are not hypothesized.

† p , .10 *p , .05

**p , .01

362 FebruaryAcademy of Management Journal

a gestalt perception of norm violation acceptability that overarches the normative landscape of the organization.

Practical Implications

Although successful leaders are often able to use humor to motivate their followers to achieve greater performance (e.g., Avolio et al., 1999; Vecchio, Justin, & Pearce, 2009), our findings call attention to a potential risk that humorous leaders should be aware of. To be clear, we are not suggesting that leaders should stop displaying humor in the work- place. Instead, as our mixed blessing model dem- onstrates, leaders must strive to embody the values they possess while simultaneously creating an en- vironment in which norm adherence is encouraged. We suggest that one caveat lies in the type of humor being displayed—aggressive humor harms relation- ship quality and elicits more norm violating behav- iors from followers. Therefore, we encourage leaders to continue expressing humor, but to minimize the usage of aggressive humor as much as possible. Ad- mittedly, it can be difficult to dictate what kind of humor a leader should display. Organizations can educate leaders on the potentially negative conse- quences of aggressive humor and encourage the usage of more positive forms of humor through training (Prerost, 1993). Through appropriate train- ing, we believe that humorous leaders can reap the benefits (i.e., more engaged followers) while mini- mizing the negative consequences (i.e., deviant followers). Indeed, compared to other structural policies (e.g., employee retreats designed to increase engagement), leader sense of humor could be a cost- less strategy in creating a more relaxed and engaged workplace.

Although leaders can be trained to reduce their usage of aggressive humor (Prerost, 1993) and hence reduce follower deviance, our research revealed both main and indirect effects of leader sense of hu- mor on follower deviance even when aggressive humor is not taken into account. Therefore, we sug- gest that it is equally important to socialize em- ployees to espouse the normative values in an organization and refrain from engaging in deviant behavior. One way to achieve this is by having a for- mal code of conduct for workplace interactions among colleagues. Another way to achieve this is by reinforcing identification with the organization. When employees identify themselves as an integral part of the organization, they will be less likely to engage in behaviors that would potentially harm the

organization or its members (Rousseau, 1998). With strong socialization programs for newcomers and a culture that promotes identification with the or- ganization, the effects of leaders’ displays of humor on follower deviance may be mitigated.

The broader implication of our research is that leaders need to be mindful of their status as role models. Leaders need to be aware that their actions serve as social cues for their followers and can cause both positive and negative consequences. Therefore, leaders should be careful in how they portray themselves to their followers in their interactions. Leaders should seek to increase their self-monitoring skills, such that they will be more aware of what types of humor are appropriate in different situations.

Limitations and Future Research Directions

Although there are several strengths of the current work, such as our effort to replicate the findings across two distinct cultures using multi-wave de- signs, several limitations warrant further discussion and future research. First and foremost, although in both studies we adapted a validated measure to as- sess leader sense of humor (Thorson&Powell, 1993), thismeasuremight not havebeenvalence-free in that a few items appear to capture not just a leader’s predisposition of using humor, but also her/his pre- disposition of using positive humor. Although the null correlation between leader sense of humor and aggressive humor (r5 .06,p5 .39) in Study 2 andour supplemental study suggest that our measure of leader sense of humor indeed is valence-free, we strongly recommend future research to develop new, rigorous humor assessments, specifically in the context of organizational studies.

Second, we have only examined one boundary condition of the link between leader sense of humor and follower outcomes, namely leader aggressive humor. Future research can examine whether other styles of leader humor might moderate the effects of leader sense of humor on follower outcomes. For example, self-deprecating humor might be the best style of humor indevelopinghigh-qualityLMXgiven its non-threatening nature (Martin et al., 2003). Given that a complete view of the leadership process involves not just leader characteristics, but also follower characteristics (Howell & Shamir, 2005; Kellerman, 2008), we encourage future research to examine additional follower characteristics as po- tential moderators. For example, followers who are high in perspective taking (Galinsky & Moskowitz,

2018 363Yam, Christian, Wei, Liao, and Nai

2000) might understand the higher-order messages that humorous leaders are signaling (e.g., a relaxing workplace rather than mere norm violations) and hence be less likely to engage in deviance.

Third, humor research in organizational behavior is rather sporadic (Duncan et al., 1990). Althoughwe examined follower deviance and work engagement, we encourage future research to continue to explore the additional consequences of leader humor and their underlying mechanisms. On the positive side, given the norm violating nature of most humor, a leader’s sense of humormight facilitate followers to think “outside of the box” and achieve greater crea- tivity (Cooper, 2008). On the negative side, followers mightmimic leaders’displays of aggressivehumor to their peers, thus leading to increased group re- lational conflict and decreased group cohesion. For these negative behavioral consequences, it would be useful to examine both other- and self-reported sources because self-reported negative behaviors (e.g., deviance) might suffer from biases associated with social desirability (Berry et al., 2007) or moti- vated forgetting (Kouchaki & Gino, 2016), a limita- tion of the current set of studies.

Fourth, given the correlational nature of our field studies, we were not able to definitively establish causal inferences. To partially offset concerns of re- verse causality,we followed recentwork (e.g.,Matta, Scott, Colquitt, Koopman, & Passantino, 2016) to compare the values of Akaike’s Information Crite- rion (AIC) and Bayesian Information Criterion (BIC) for our hypothesized model and alternative, reserve causal models. According to Kline, the model with the smallest AIC and BIC fits the data best and is “the one most likely to replicate” (Kline, 2011: 220). Re- sults suggested that thehypothesizedmodel inStudy 1 (Humor → Perceived acceptability of norm viola- tion → Deviance) had lower AIC and BIC values (AIC 5 1129.83, BIC 5 1153.43) compared to the reverse causal model (Deviance → Perceived ac- ceptability of norm violation → Humor: AIC 5 1402.28, BIC 5 1425.86). Similarly, the two hy- pothesized models (positive and negative paths of leader sense of humor) in Study 2 had lower AIC and BIC values compared to the reverse causal models (Hypothesized positive path: AIC 5 1336.16, BIC 5 1359.25 vs. reverse causal positive path: AIC 5 1452.85, BIC 5 1475.91; Hypothesized negative path: AIC 5 894.78, BIC 5 917.87 vs. reverse causal negative path: AIC 5 1323.02, BIC 5 1346.11). Al- though these results reinforce the validity in the di- rectionality of our findings, we recommend future research to conduct experimental investigations of

our proposed mechanisms. For example, future research could create simulated leader2follower interactions and manipulate, rather than measure, leader humor.

Relatedly, becausewewere not able to definitively establish causal inferences, there might be other al- ternative mediating factors that could explain our findings. As Zhao, Lynch, and Chen (2010) sug- gested, complementary mediation is likely when both the direct and indirect effects are statistically significant and in the same direction (i.e., the path from leader sense of humor to deviance in Study 1 and the path from leader sense of humor to work engagement in Study 2). We therefore suggest that there might be other complementary mediating mechanisms derived from other theories that are driving our results and recommend future research to examine these mechanisms more closely.

Fifth, our measure of deviance in both studies is self-reported in nature. While many deviant behav- iors are done in private and hence an other-report format would not have been applicable, we suggest future research control for factors such as social de- sirability biases that might likely create a floor effect for this measure.

Finally, although norm violation and deviance are often perceived as negative in many organizations, some organizations may explicitly encourage such behaviors. In industries such as entertainment or tourism, employees are often required to display humor. In such industries, a leader’s sense of humor may not signal norm violation, but rather norm ad- herence. In other words, organizational or industrial and societal norms sometimes might be in conflict, highlighting the important moderating role of in- dustry. Future research should examine the impli- cations for leader’s sense of humor in industries where humor is explicitly encouraged.

CONCLUSION

In this research, we integrated benign violation and social information processing theories and found that leader’s sense of humor is a mixed bless- ing. It leads to increased follower deviance and in- creased work engagement. Although the current research sheds some light on the link between leader’s sense of humor and follower outcomes, we recognize that we have only taken one step toward a more complete understanding of the effects of leader humor in organizations; many questions re- main for future studies to tackle. Nonetheless, we hope the introduction of benign violation theory and

364 FebruaryAcademy of Management Journal

the current work can spark additional research on humor in organizational behavior.

REFERENCES

Abel, M. H. 2002. Humor, stress, and coping strategies. Humor: International Journal of Humor Research, 15: 365–381.

Avolio, B. J., Howell, J.M., & Sosik, J. J. 1999. A funny thing happened on the way to the bottom line: Humor as a moderator of leadership style effects. Academy of Management Journal, 42: 219–227.

Baron, R., & Kenny, D. 1986. The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51: 1173–1182.

Bauer, T. N., & Green, S. G. 1996. Development of leader- member exchange: A longitudinal test. Academy of Management Journal, 39: 1538–1567.

Becker, T. E. 2005. Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organi- zational Research Methods, 8: 274–289.

Bennett, R. J., & Robinson, S. L. 2000. Development of a measure of workplace deviance. The Journal of Applied Psychology, 85: 349–360.

Berry, C. M., Ones, D. S., & Sackett, P. R. 2007. Inter personal deviance, organizational deviance, and their common correlates: A review and meta-analysis. The Journal of Applied Psychology, 92: 410–424.

Bettenhausen, K., & Murnighan, J. 1985. The emergence of norms in competitive decision making groups. Administrative Science Quarterly, 30: 350–372.

Billig, M. 2005. Laughter and ridicule: Towards a social critique of humor. London: Sage Publications.

Bressler, E. R., & Balshine, S. 2006. The influence of humor on desirability. Evolution and Human Behavior, 27: 29–39.

Brislin, R. W. 1970. Back-translation for cross-cultural re- search. Journal of Cross-Cultural Psychology, 1: 185–216.

Buffalo, M., & Rodgers, J. W. 1971. Behavioral norms, moral norms, and attachment: Problems of deviance and conformity. Social Problems, 19: 101–113.

Chen, G. H., & Martin, R. A. 2007. A comparison of humor styles, coping humor, and mental health between Chinese and Canadian university students. Humor: International Journal of Humor Research, 20: 215–234.

Christian, J. S., Christian, M. S., Garza, A. S., & Ellis, A. P. 2012. Examining retaliatory responses to justice

violations and recovery attempts in teams. The Jour- nal of Applied Psychology, 97: 1218–1232.

Christian,M. S., &Ellis,A. P. 2011. Examining the effects of sleep deprivation on workplace deviance: A self- regulatory perspective. Academy of Management Journal, 54: 913–934.

Christian, M. S., Garza, A. S., & Slaughter, J. E. 2011. Work engagement: A quantitative review and test of its re- lations with task and contextual performance. Per- sonnel Psychology, 64: 89–136.

Colquitt, J. A., & Zapata-Phelan, C. P. 2007. Trends in the- ory building and theory testing: A five-decade study of the Academy of Management Journal. Academy of Management Journal, 50: 1281–1303.

Cooper, C. 2008. Elucidating the bonds of workplace hu- mor: A relational process model. Human Relations, 61: 1087–1115.

Cooper, C. D. 2005. Just joking around? Employee humor expression as an ingratiatory behavior. Academy of Management Review, 30: 765–776.

Crawford, C. B. 1994. Theory and implications regarding the utilization of strategic humor by leaders. The Journal of Leadership Studies, 1: 53–67.

Dampier, P., &Walton,A. 2013.WhiteHousewit,wisdom, and wisecracks. London: Barzipan Publishing.

Davis, M. S. 1995. The sociology of humor: A stillborn field? Sociological Forum, 10: 327–339.

Decker, W. H., & Rotondo, D. M. 2001. Relationships among gender, type of humor, and perceived leader effectiveness. Journal of Managerial Issues, 13: 450–465.

DeCelles, K. A., DeRue, D. S., Margolis, J. D., & Ceranic, T. L. 2012. Does power corrupt or enable? When and why power facilitates self-interested behavior. The Journal of Applied Psychology, 97: 681–689.

Dews, S., &Winner, E. 1995. Muting themeaning: A social function of irony.Metaphor and Symbol, 10: 3–19.

Duncan, W. J. 1982. Humor in management: Prospects for administrative practice and research. Academy of Management Review, 7: 136–142.

Duncan, W. J., Smeltzer, L. R., & Leap, T. L. 1990. Humor and work: Applications of joking behavior to man- agement. Journal of Management, 16: 255–278.

Fehr, R., Yam, K. C., He, W., Chiang, J., & Wei, W. 2017. Polluted work: A self-control perspective on air pol- lution appraisals, organizational citizenship, and counterproductive work behavior. Organizational Behavior andHumanDecision Processes. Published online ahead of print.

Ford, T. E. 2000. Effects of sexist humor on tolerance of sexist events. Personality and Social Psychology Bulletin, 26: 1094–1107.

2018 365Yam, Christian, Wei, Liao, and Nai

Ford, T. E., Wentzel, E. R., & Lorion, J. 2001. Effects of exposure to sexist humor on perceptions of normative tolerance of sexism. European Journal of Social Psychology, 31: 677–691.

Galinsky, A. D., & Moskowitz, G. B. 2000. Perspective- taking: Decreasing stereotype expression, stereo- type accessibility, and in-group favoritism. Journal of Personality and Social Psychology, 78: 708–724.

Gioia, D. A., & Manz, C. C. 1985. Linking cognition and behavior: A script processing interpretation of vicari- ous learning. Academy of Management Review, 10: 527–539.

Gkorezis, P., Petridou, E., & Xanthiakos, P. 2014. Leader positive humor and organizational cynicism: LMX as a mediator. Leadership and Organization Develop- ment Journal, 35: 305–315.

Goel, V., & Dolan, R. J. 2001. The functional anatomy of humor: Segregating cognitive and affective compo- nents. Nature Neuroscience, 4: 237–238.

Graen, G. B., & Scandura, T. A. 1987. Toward a psychology of dyadic organizing. Research in –Organizational Behavior, 9: 175–208.

Graen, G. B., & Uhl-Bien, M. 1995. Relationship-based approach to leadership: Development of leader- member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain per- spective. The Leadership Quarterly, 6: 219–247.

Graham, E. E. 1995. The involvement of sense of humor in the development of social relationships. Communi- cation Reports, 8: 158–169.

Hayes,A. F. 2013. Introduction tomediation,moderation, and conditional process analysis: A regression- based approach. New York: Guilford Press.

He, W., Fehr, R., Yam, K. C., Long, L. R., & Hao, P. 2017. Interactional justice, leader2member exchange, and employee performance: Examining the moderating role of justice differentiation. Journal of Organiza- tional Behavior, 38:537–557.Published online ahead of print.

Heath, C., & Sitkin, S. B. 2001. Big‐B versus Big‐O:What is organizational about organizational behavior? Jour- nal of Organizational Behavior, 22: 43–58.

Hershcovis, M. S., Turner, N., Barling, J., Arnold, K. A., Dupré, K. E., Inness,M., LeBlanc,M.M.,&Sivanathan, N. 2007. Predicting workplace aggression: A meta- analysis. The Journal of Applied Psychology, 92: 228–238.

Hodson, G., Rush, J., & MacInnis, C. C. 2010. A joke is just a joke (except when it isn’t): Cavalier humor beliefs facilitate the expression of group dominance motives. Journal of Personality and Social Psychology, 99: 660–682.

Hogg, M. A. 2010. Influence and leadership. In S. T. Fiske, D. T. Gilbert, & G. Lindzey (Eds.),Handbook of social psychology: 1166–1207. Hoboken, New Jersey: John Wiley & Sons.

Howell, J. M., & Shamir, B. 2005. The role of followers in the charismatic leadership process: Relationships and their consequences. Academy of Management Re- view, 30: 96–112.

Huang, L., Gino, F., & Galinsky, A. D. 2015. The highest form of intelligence: Sarcasm increases creativity for both expressers and recipients. Organizational Be- havior and Human Decision Processes, 131: 162– 177.

James, L. R., Hater, J. J., Gent, M. J., & Bruni, J. R. 1978. Psychological climate: Implications from cognitive social learning theory and interactional psychology. Personnel Psychology, 31: 783–813.

James, L. R., & Jones, A. P. 1976. Organizational structure: A review of structural dimensions and their con- ceptual relationships with individual attitudes and behavior. Organizational Behavior and Human Performance, 16: 74–113.

Judge, T. A., Scott, B. A., & Ilies, R. 2006. Hostility, job attitudes, and workplace deviance: Test of a multi- level model. The Journal of Applied Psychology, 91: 126–138.

Kahn, W. A. 1990. Psychological conditions of personal engagement and disengagement at work. Academy of Management Journal, 33: 692–724.

Kalbfleischl, P. I. 2013. Interpersonal communication: Evolving interpersonal relationships. Oxford: Psy- chology Press.

Katz, J. 1996. Families and funny mirrors: A study of the social construction and personal embodiment of humor. American Journal of Sociology, 101: 1194–1237.

Kellerman, B. 2008. Followership: How followers are creating change and changing leaders. Boston, MA: Harvard Business School Press.

Kline, R. B. 2011. Principles and practice of structural equation modeling. New York: Guilford.

Kouchaki, M., & Gino, F. 2016. Memories of unethical ac- tions become obfuscated over time. Proceedings of the National Academy of Sciences of the United States of America, 113: 6166–6171.

Lee, K., & Allen, N. J. 2002. Organizational citizenship behavior and workplace deviance: The role of affect and cognitions. The Journal of Applied Psychology, 87: 131–142.

Long, C., Bendersky, C., & Morrill, C. 2011. Fairness monitoring: Linking managerial controls and fairness judgments in organizations. Academy of Manage- ment Journal, 54: 1045–1068.

366 FebruaryAcademy of Management Journal

MacKinnon, D. P., Lockwood, C. M., & Williams, J. 2004. Confidence limits for the indirect effect: distribution of the product and resampling methods.Multivariate Behavioral Research, 39: 99–128.

Matta, F. K., Scott, B., Colquitt, J., Koopman, J., & Passan- tino, L. 2016. Is consistently unfair better than spo- radically fair? An investigation of justice variability and stress. Academy of Management Journal. Pub- lished online ahead of print.

Matta, F. K., Scott, B.A., Koopman, J., &Conlon, D. E. 2015. Does seeing “eye to eye” affect work engagement and organizational citizenship behavior? A role theory perspective on LMX agreement. Academy of Man- agement Journal, 58: 1686–1708.

Malone, P.B. 1980.Humor:Adouble-edged tool for today’s managers? Academy of Management Review, 5: 357–360.

Martin, R. A. 2001. Humor, laughter, and physical health: Methodological issues and research findings. Psy- chological Bulletin, 127: 504–519.

Martin, R. A., Puhlik-Doris, P., Larsen, G., Gray, J., & Weir, K. 2003. Individual differences in uses of humor and their relation to psychological well-being: Develop- ment of the Humor Styles Questionnaire. Journal of Research in Personality, 37: 48–75.

May, D. R., Gilson, R. L., & Harter, L. M. 2004. The psy- chological conditions of meaningfulness, safety and availability and the engagement of the human spirit at work. Journal of Occupational and Organizational Psychology, 77: 11–37.

McGraw, A. P., & Warren, C. 2010. Benign violations making immoral behavior funny. Psychological Sci- ence, 21: 1141–1149.

McGraw, P., &Warner, J. 2014. The humor code: A global search for what makes things funny. New York: Simon & Schuster.

McGraw, A. P., Warren, C., Williams, L. E., & Leonard, B. 2012. Too close for comfort, or too far to care? Finding humor in distant tragedies and close mishaps. Psy- chological Science, 23: 1215–1223.

McGraw, A. P.,Williams, L. E., &Warren, C. 2014. The rise and fall of humor: Psychological distance modulates humorous responses to tragedy.Social Psychological & Personality Science, 5: 566–572.

Mesmer-Magnus, J., Glew, D. J., & Viswesvaran, C. 2012. A meta-analysis of positive humor in the workplace. Journal of Managerial Psychology, 27: 155–190.

Mitchell, M., Vogel, R., & Folger, R. 2015. Third parties’ reactions to abusive supervision of coworkers. The Journal of Applied Psychology, 100: 1040–1055.

Moran, J. M., Rain, M., Page-Gould, E., & Mar, R. A. 2014. Do I amuse you? Asymmetric predictors for humor

appreciation and humor production. Journal of Research in Personality, 49: 8–13.

Morris, M. W., Hong, Y., Chiu, C., & Liu, Z. 2015. Nor- mology: Integrating insights about social norms to understand cultural dynamics. Organizational Behavior and Human Decision Processes, 129: 1–13.

Moon, H. 2001. The two faces of conscientiousness: Duty andachievement striving in escalationof commitment dilemmas. The Journal of Applied Psychology, 86: 533–540.

Peterson, D. K. 2002. The relationship between unethical behavior and the dimensions of the ethical climate questionnaire. Journal of Business Ethics, 41: 313– 326.

Pexman, P. M., & Olineck, K. M. 2002. Understanding irony:Howdo stereotypes cue speaker intent? Journal of Language and Social Psychology, 21: 245–274.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. 2003. Common method biases in behavioral research: A critical review of the literature and rec- ommended remedies. The Journal of Applied Psy- chology, 88: 879–903.

Preacher, K. J., & Hayes, A. F. 2008. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav- ior Research Methods, 40: 879–891.

Prerost, F. J. 1993.A strategy to enhance humor production among elderly persons: Assisting in the management of stress. Activities, Adaptation and Aging, 17: 17–24.

Pundt, A., & Venz, L. 2017. Personal need for structure as a boundary condition for humor in leadership. Jour- nal of Organizational Behavior, 38: 872107.

Rich, B. L., Lepine, J. A., & Crawford, E. R. 2010. Job en- gagement: Antecedents and effects on job perfor- mance. Academy of Management Journal, 53: 617–635.

Robert, C., Dunne, T., & Iun, J. 2016. The impact of leader humor on subordinate job satisfaction: The crucial role of leader2subordinate relationship quality. Group & Organization Management, 41: 375–406.

Robinson, S. L., & Bennett, R. J. 1995. A typology of deviant workplace behaviors: A multidimensional scaling study. Academy of Management Journal, 38: 555– 572.

Robinson, S. L., & O’Leary-Kelly, A. M. 1998. Monkey see, monkey do: The influence of work groups on the an- tisocial behavior of employees. Academy of Man- agement Journal, 41: 658–672.

Romero, E. J., & Cruthirds, K.W. 2006. The use of humor in the workplace. The Academy of Management Per- spectives, 20: 58–69.

2018 367Yam, Christian, Wei, Liao, and Nai

Rousseau, D. M. 1998. Why workers still identify with or- ganizations. Journal of Organizational Behavior, 19: 217–233.

Salancik, G. R., & Pfeffer, J. 1978. A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 23: 224–253.

Schaufeli, W. B., Salanova, M., González-Romá, V., & Bakker, A. B. 2002. The measurement of engagement and burnout: A two sample confirmatory factor ana- lytic approach. Journal of Happiness Studies, 3: 71–92.

Schneider, B. 1990. Organizational climate and culture. San Francisco, CA: Jossey-Bass.

Schneider, B., & Reichers, A. E. 1983. On the etiology of climates. Personnel Psychology, 36: 19–39.

Shamir, B., House, R. J., & Arthur, M. B. 1993. The moti- vational effects of charismatic leadership: A self- concept based theory. Organization Science, 4: 577–594.

Sobel, M. E. 1982. Asymptotic confidence intervals for indirect effects in structural equation models. Socio- logical Methodology, 13: 290–312.

Thorson, J. A., & Powell, F. C. 1993. Relationships of death anxiety and sense of humor. Psychological Reports, 72: 1364–1366.

Toplak, Maggie, & Katz, Albert N. 2000. On the uses of sarcastic irony. Journal of Pragmatics, 32: 1467–1488.

Van den Bos, K. 2001. Uncertainty management: The in- fluence of uncertainty salience on reactions to per- ceived procedural fairness. Journal of Personality and Social Psychology, 80: 931–941.

Van Kleef, G. A., Homan, A. C., Finkenauer, C., Gündemir, S., & Stamkou, E. 2011. Breaking the rules to rise to power: How norm violators gain power in the eyes of others. Social Psychological & Personality Science, 2: 500–507.

Vecchio, R. P., Justin, J. E., & Pearce, C. L. 2009. The in- fluence of leader humor on relationships between leader behavior and follower outcomes. Journal of Managerial Issues, 21: 171–194.

Vernon, P. A., Martin, R. A., Schermer, J. A., & Mackie, A. 2008.A behavioral genetic investigation of humor styles and their correlations with the big-5 personality di- mensions. Personality and Individual. Differences: A Journal of Feminist Cultural Studies, 44: 1116–1125.

Wellman, N., Mayer, D. M., Ong, M., & DeRue, D. S. 2016. When are do-gooders treated badly? Legitimate power, role expectations, and reactions to moral ob- jection in organizations. The Journal of Applied Psychology, 101: 7832814.

Yam, K. C., Chen, X. P., & Reynolds, S. J. 2014. Ego de- pletion and its paradoxical effects on ethical decision

making. Organizational Behavior and Human De- cision Processes, 124: 204–214.

Yam, K. C., Reynolds, S. J., & Hirsh, J. B. 2014. The hungry thief: Physiological deprivation and its effect on un- ethical conduct. Organizational Behavior and Hu- man Decision Processes, 125: 123–133.

Zhang, Y., & Liao, Z. 2015. Consequences of abusive su- pervision: A meta-analytic review. Asia Pacific Journal of Management, 32: 959–987.

Zhao,X., Lynch, J.G., &Chen,Q.2010.ReconsideringBaron andKenny:Myths and truths aboutmediation analysis. The Journal of Consumer Research, 37: 197–206.

Zillmann, D. 1983. Disparagement humor. In P. E.McGhee & J. H. Goldstein (Eds.), Handbook of humor re- search: 85–107. New York: Springer.

Ziv, A. 1981. Psychology of humor (Hebrew). Tel-Aviv: Yahdav.

Zohar, D., & Luria, G. 2004. Climate as a social-cognitive construction of supervisory safety practices: Scripts as proxy of behavior patterns. The Journal of Applied Psychology, 89: 322–333.

Zohar, D., & Tenne-Gazit, O. 2008. Transformational leadership and group interaction as climate anteced- ents: A social network analysis. The Journal of Ap- plied Psychology, 93: 744–757.

KaiChiYam ([email protected]) is an assistant professor of management at the National University of Singapore. He received his PhD from the University of Washington. His research focuses on behavioral ethics, leadership, and humor.

Michael S. Christian ([email protected]) is the Sarah Graham Kenan Associate Professor of Organiza- tional Behavior at the Kenan–Flagler Business School at the University of North Carolina at Chapel Hill. He com- pleted his PhD in management at the University of Ari- zona. Dr. Christian’s current research is focused on human energy, engagement, and self-regulation as well as on the motivational drivers of deviant and unethical behavior.

WuWei ([email protected]) is a professor at the School of Economics andManagement atWuhanUniversity,China. He received his PhD from the School of Management at Huazhong University of Science and Technology. His re- search interests include team leadership, employee deviant behavior, impression management, and nonmarket action.

ZhenyuLiao ([email protected]) is a post-doctoral research fellow in the Olin Business School at the Washington University in St. Louis. His research interests include leadership complexities and dynamics, episodic leader– member interactions, and behavioral ethics.

368 FebruaryAcademy of Management Journal

Jared Nai ([email protected]) is an Assistant Professor of Organizational Behavior and Human Resources at the Lee Kong Chian School of Business at Singapore Manage- ment University. He completed his PhD in Management and Organization at the National University of Singapore. His current research interests include prosocial behavior, deviance, and advice seeking.

APPENDIX

Leader sense of humor (Studies 1 and 2)

1. Say things in such away as tomakepeople laugh 2. Is regarded as someone of a wit by coworkers 3. Uses humor to entertain coworkers 4. Cracks people up with the things he/she says 5. Can ease a tense situation by saying something

funny 6. Can exert control over a group by uses of humor 7. Say clever things that amuse others

Leader aggressive humor (Study 2)

1. If someone makes a mistake, my leader will often tease them about it.

2. My leader’s sense of humor often offends others. 3. When telling jokes or saying funny things, my

leader usually is not very concerned about how other people are taking it.

4. My leader does not like it when people use humor as a way of criticizing or putting some- one down (reverse-coded).

5. Sometimes my leader thinks of something that is so funny that he/she can’t stop himself/ herself from saying it, even if it is not appro- priate for the situation.

6. My leader never participates in laughing at others even if all others are doing it (reverse- coded).

7. If my leader doesn’t like someone, he/she often uses humor or teasing to put them down.

8. Even if something is really funny to my leader, he/shewill not laugh or joke about it if someone will be offended (reverse-coded).

2018 369Yam, Christian, Wei, Liao, and Nai

Copyright of Academy of Management Journal is the property of Academy of Management and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.