36 Academy of Management Perspectives February ARTICLES

Why Does Matter in Organizations? by Sigal G. Barsade and Donald E. Gibson1

Executive Overview in and research about affect in organizations have expanded dramatically in recent years. This article reviews what we know about affect in organizations, focusing on how employees’ moods, , and dispositional affect influence critical organizational outcomes such as job performance, decision making, , turnover, prosocial behavior, teamwork, , and . This review highlights pervasive and consistent effects, showing the importance of affect in shaping a wide variety of organizational behaviors, the knowledge of which is critical for researchers, managers, and employees.

Why Does Affect Matter in Organizations? ffect permeates organizations. It is present in the interdependent relationships we hold with An organizational vignette... Abosses, team members, and subordinates. It is I had just mentioned how small a raise I was going to give present in deadlines, in group projects, in human to Jerry, my top salesperson this year. I could see a subtle resource processes like performance appraisals and wave of and wash over his usually calm features. I had been afraid this was going to happen. But selection interviews. Affective processes (more what could I do? I was caught in the middle—the CEO commonly known as emotions) create and sustain wanted to cut our budget by 6%! Jerry’s voice had an work motivation. They lurk behind political be- edge to it, and I could tell that my explanations about the havior; they animate our decisions; they are es- budget were not going to solve this one. Would he ex- sential to leadership. Strong affective are plode? Would he blame me? Worse, would he threaten to present at any time we confront work issues that quit? I could feel the good I had started with this morning rapidly disappearing. The insistent brittleness in matter to us and our organizational performance. his voice made me feel defensive and I was starting to get In the last 30 years, an “affective revolution” angry myself. I needed to decide what to do next, but I has taken place, in which academics and managers was having trouble remembering the rationale for the alike have begun to appreciate how an organiza- raise. I felt like yelling at him. That, I told myself, tional lens that integrates employee affect pro- cannot happen. I need to keep it under control. . .I’m the boss here, remember? He’s watching how I act. I vides a perspective missing from earlier views need to figure out how I want to deal with his (Barsade, Brief, & Spataro, 2003). While much anger—and mine. . . about affect remains difficult to explain, dramatic progress has been made in understanding individ- uals’ affective lives in organizations. In this article, we examine why affect is important to organiza- 1 Order of authorship is alphabetical; both authors contributed equally. tional life. We do so by drawing on a range of We would like to thank Peter Cappelli, Yochi Cohen Charash, Chia-Jung Tsay, Marina Milonova, Amanda O’Neill and our anonymous reviewers for studies that help identify critical organizational their help and insights.

* Sigal G. Barsade ([email protected]) is an Associate Professor of Management at the Wharton School, The University of Pennsylvania. Donald E. Gibson ([email protected]field.edu) is an Associate Professor of Management at the Dolan School of Business, Fairfield University.

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. 2007 Barsade and Gibson 37 outcomes driven by affect and show how under- to be assessed and examined differently than standing feelings can help researchers, managers, moods and dispositional affect. Because emotions and employees themselves explain and predict are focused on a specific target or cause, they have attitudes and behavior in organizations, from turn- come to be regarded as discrete, and are linked to over to decision making to leadership. We address specific tendencies to act (such as the to the question, “How does research seen through approach objects in anger and to avoid them in the lens of affect cause us to think differently ; Frijda, 1986). The discrete emotions ap- about the assumptions we make about how em- proach has identified “basic” or primary emotions, ployees work?” including , , anger, fear, , , and , each with a unique set of prototypi- Defining Affect in Organizations cal antecedents and consequences—though the e begin by defining a range of terms often precise number and identity of discrete emotions used in research on affect in organizations are subjects of much debate (see Ekman, 1992; (see Table 1). These terms describe phe- Ortony & Turner, 1990). Moods and dispositional W affect, in contrast, tend to be examined through nomena ranging from discrete emotions (fear, an- ger, or disgust), to moods ( cheerful versus an approach that summarizes the wide variety of feeling melancholy), to dispositional traits (“He’s possible human affective experiences into a few such a negative person”; “She’s always so up- critical underlying dimensions. Dimensional ap- beat!”), to meta-emotional abilities, such as emo- proaches often arrange affective experience labels tional intelligence (“My boss is very good at un- (such as “astonished,” “enthusiastic,” or derstanding how the people on our team are “grouchy”) in a circular graph called an affective feeling”). Affect can be thought of as an umbrella circumplex, and represent the dimensions as axes term encompassing a broad range of feelings that on that circumplex (see Figure 1). individuals experience, including feeling states, The first factor of the circumplex, on the x axis, which are in-the-moment, short-term affective is “pleasantness,” a dimension ranging from high experiences, and feeling traits, which are more sta- pleasantness to low pleasantness (or unpleasant). ble tendencies to feel and act in certain ways The second dimension, on the y axis, is an “acti- (Watson & Clark, 1984).2 Within feeling states vation/energy” dimension, ranging from high to there are two established categories: emotions and low energy (Russell, 1980). Moods are usually moods. Emotions are elicited by a particular target examined based on their hedonic tone—that is, or cause, often include physiological reactions and how pleasant (toward happy) or unpleasant (to- action sequences, and are relatively intense and ward sad) the mood is. Dispositional affect can be short-lived (Frijda, 1986; Lazarus, 1991). In con- examined as trait pleasantness (e.g. Staw & Bar- trast, moods are more diffuse, take the form of a sade, 1993) or as the combination of the pleasant- general positive (pleasant) or negative (unpleas- ness and energy dimensions, creating the two in- ant) feeling, and tend not to be focused on a dependent constructs of (PA) specific cause (Frijda, 1986; Tellegen, 1985). and (NA) (Watson, Clark, & There is only one category of feeling trait: dispo- Tellegen, 1988; see the dotted lines in Figure 1). sitional affect. This is a personality trait referring to For example, individuals characterized by high a person’s relatively stable, underlying tendency to dispositional NA tend to be distressed, upset, and experience positive and negative moods and emo- have a negative view of self over time and across tions (Watson & Clark, 1984). situations, as opposed to the more serene, calm, In terms of research approaches, emotions tend shown by people who are low in NA; people high in dispositional PA tend to be cheer- ful and energetic, and experience positive moods, 2 These affective states and traits differ from sentiments or even attitudes such as or well-being, across a variety of (e.g., job satisfaction) in that the latter reflect an evaluation of a particular object, and whether that object is evaluated as something that is liked or situations, as compared to those low in PA who disliked (Kelly & Barsade, 2001; Weiss, 2002). experience more sadness, melancholy, dullness, or 38 Academy of Management Perspectives February

Table 1 Translating Affective Terms Terms Used in Research Formal Definition Colloquial Terms Affect Umbrella term encompassing a broad range of feelings “I feel . . .” “She seems to be feeling . . . “ that individuals experience, including feeling states, such “He is usually unemotional . . .” as moods and discrete emotions, and traits, such as trait positive and negative affectivity (all defined below). Discrete Emotions Emotions are focused on a specific target or cause – For example, love, anger, hate, fear, generally realized by the perceiver of the ; , , sadness, , , relatively intense and very short-lived. After initial aggravation, , , joy, , intensity, can sometimes transform into a mood. fright, etc. Moods Generally take the form of a global positive (pleasant) or Feeling good, bad, negative, positive, negative (unpleasant) feeling; tend to be diffuse—not cheerful, down, pleasant, irritable, etc. focused on a specific cause—and often not realized by the perceiver of the mood; medium duration (from a few moments to as long as a few weeks or more). Dispositional (Trait) Affect Overall personality tendency to respond to situations in “No matter what, he’s always ____.” “She stable, predictable ways. A person’s “affective lens” on the tends to be in a ____ mood all the time.” world. “He is always so negative.” a) (Trait) Positive Individuals who tend to be cheerful and energetic, and who “She’s always so energetic and upbeat!” Affectivity experience positive moods, such as pleasure or well-being, “He’s such a downer all the time!” across a variety of situations as compared to people who tend to be low energy and sluggish or melancholy. b) (Trait) Negative Individuals who tend to be distressed and upset, and have a “She is always so hostile in her approach.” Affectivity negative view of self over time and across situations, as “Why is he always so anxious/nervous?” “I compared to people who are more calm, serene and admire his steady and serenity.” relaxed. “The ability to monitor one’s own and others’ feelings and “My manager is terrible at expressing his emotions, to discriminate among them, and to use this emotions.” “My teammate is great at information to guide one’s thinking and actions” (Salovey knowing how everyone else on the team is & Mayer, 1990: 189). feeling.” “The CEO is brilliant at dealing with her employees’ emotions—a real motivator!” Emotional Regulation Individuals’ attempts to “influence which emotions they “He handles his emotions really well, even have, when they have them, and how they experience and under high pressure situations.” express these emotions” (Gross, 1998a: 275). Requires an employee to “induce or suppress feeling in She has to put on a smile when dealing with order to sustain the outward countenance that produces the customers, because it’s part of the job. proper state of mind in others” (Hochschild, 1983: 7). Processes that allow the sharing or transferring of emotions “And when we feel good, it’s contagious.” from one individual to other group members; the tendency (Advertising slogan from Southwest Airlines) to mimic the nonverbal behavior of others, to “synchronize “I don’t know why, but every time I talk to facial expressions, vocalizations, postures, and him I feel really anxious afterwards.” movements” with others, and in turn, to “converge “Infectious .” emotionally” (Hatfield, Cacioppo, & Rapson, 1994). Collective Affect A “bottom-up” approach to collective affect emphasizes the “Our group has a _____ feel to it.” “What a affective composition of the various affective attributes of negative group!” “In our group showing the group’s members. That is, the degree to which positivity is very important.” individual level affective characteristics combine, often through emotional contagion, to form group level emotion or mood. A “top-down” approach to collective affect emphasizes the degree to which groups are characterized by emotion norms for feeling and expression. 2007 Barsade and Gibson 39 lethargy (Watson & Tellegen, 1985). Other affec- of basic constructs. This research progress now tive traits that can influence work behavior in- makes it possible to examine affective influences clude people’s propensities to feeling emotions on organizational outcomes with greater precision strongly (affective intensity; Larsen & Diener, and specificity. While this increasing consistency 1987); being prone to catching other people’s in definitions has helped, the range of approaches emotions (emotional contagion; Hatfield, Ca- to studying affect in organizations remains broad. cioppo, & Rapson, 1994; Doherty, 1997); and We identify three emerging trends in conceiving how emotionally expressive people tend to be affect in organizations that take the definitions we (emotional expressivity; Kring, Smith, & Neale, have outlined in new directions and will continue 1994). to shape research language and focus.

New Ways of Approaching Affect in Emotional Intelligence Organizations Recently, there has been significant popular and he delineation of affective terms outlined academic interest in the phenomenon of emo- above represents significant progress in a field tional intelligence (EI), an “ability to monitor Tthat has traditionally been characterized by one’s own and others’ feelings and emotions, to little agreement over the meaning and boundaries discriminate among them, and to use this infor-

Figure 1 The Circumplex Model of Affect.

Adapted from Feldman Barrett, L., & Russell, J.A. (1998). Independence and bipolarity in the structure of current affect. Journal of Personality and Social , 74(4), 967–984 and Larsen, R. J., & Diener, E. (1992). Promises and problems with the circumplex model of emotion. In M. S. Clark (Ed.), Emotion. Review of personality and (Vol. 13, pp. 25–59). Thousand Oaks, CA: Sage Publications, Inc. 40 Academy of Management Perspectives February mation to guide one’s thinking and actions” own and others’ emotions to reach goals (see (Salovey & Mayer, 1990: 189). The idea behind Salovey & Grewal, 2005 for review). emotional intelligence in the workplace is that it It is important to note, however, that there is a is a skill through which employees treat emotions debate in the emotional intelligence field as to the as valuable data in navigating a situation. Let’s say exact nature of emotional intelligence (see Brack- a sales manager has come up with an amazing idea ett, Rivers, Shiffman, Lerner, & Salovey, 2006). that will increase corporate revenue by up to One group of researchers uses an “abilities” ap- 200%, but knows that his boss tends to be irritable proach based on the four-factor model described and short-tempered in the morning. Having emo- above (Mayer & Salovey, 1997), and measures tional intelligence means that this manager will emotional intelligence through performance tests first recognize and consider this emotional fact (e.g., the MSCEIT, a computer-based EI test, about his boss, and despite the stunning nature of Mayer, Salovey, Caruso, & Sitarenios, 2003). This his idea—and his own excitement—he will regu- approach differs from that of other researchers late his own emotions, curb his enthusiasm, and who take a “mixed model” approach using self- wait until the afternoon to approach his boss. It report measures of emotional intelligence.3 Self- also means understanding how one’s own emo- report measures of EI ask respondents for percep- tions and those of others can facilitate thinking. tions of their own emotional abilities through For example, the head of a product development ratings on items such as, “I am generally very good team who is about to embark on a large-scale at calming someone down when he or she is development effort senses that the team is gener- upset,” or “I can tell how people are feeling even ally feeling down and disheartened because some if they never tell me.” While these self-report key members of the team have left for a different measures may indicate respondents’ perceptions of firm. He knows that he must get his team back emotional self-efficacy (Tett, Fox, & Wang, into a positive, upbeat mood for the team mem- 2005), there is serious question as to whether bers to be productively creative in the new respondents can be unbiased about their own project. He arranges to take them on a “brain- emotional skills (Matthews, Roberts, & Zeidner, storming retreat” of white-water river rafting 2004). One might compare this approach to as- where they can connect as the newly shaped team, sessing mathematical skills by asking respondents, and most importantly, raise their positivity as they “How good are you at solving algebraic equa- embark on this new project. In doing so, this tions?” rather than asking the person to actually project leader is actively managing his own emo- solve an algebraic equation. There is also serious tions and those of his team to help meet their concern that “mixed model” self-report-based ap- goals; he is “using emotions to think intelli- proaches have substantial overlap with other per- gently.” sonality measures such as the “Big Five” personal- Mayer and Salovey’s (1997) emotional intelli- ity factors leading to issues of construct validity gence model elaborates on this premise and is (Brackett, Mayer, & Warner, 2004).4 based on the following four factors: (1) Perceiving This issue highlights the degree to which emo- Emotions: the degree to which people are capable tional intelligence is still a nascent field, both of attending to their emotions, expressing those emotions, and reading the emotions of others; (2) Using Emotions: the process of knowing which 3 This includes researchers who base their self-report assessments on the four-factor EI model, and researchers who expanded the construct to emotions facilitate cognition effectively and using include components outside that model (see Mayer, Salovey, & Caruso, them to do so; (3) Understanding Emotions: the 2000 for a more detailed discussion of these differences). understanding of complicated emotional dynam- 4 The Big Five model of personality traits measures the dimensions of extraversion, conscientiousness, agreeableness, emotional stability (also ics, including how emotions can change from one known as ) and openness to experience in individuals (McCrae to another (e.g., can turn into & Costa, 1987). These factors have emerged in a wide variety of studies of personality dimensions and are widely accepted by personality psychologists anger rather than apology); and (4) Managing (Hogan, Hogan, & Roberts, 1996), and have been found to be relevant to Emotions: the knowledge of how to regulate one’s a variety of organizational outcomes (Barrick & Mount, 1991). 2007 Barsade and Gibson 41 theoretically and methodologically, currently un- bill collectors attempting to calm or browbeat dergoing its own set of growth crises on a variety debtors (Sutton, 1991). This is considered to be of dimensions (see Conte, 2005; Daus & Ashka- labor because part of what these employees are nasy, 2005), including methodological challenges being paid to do is regulating their own emotions within the current ability-based tests (Matthews to produce the appropriate emotional state in oth- et al., 2004). However, overall there is positive ers (Hochschild, 1983). Research in this area has support for the validity of the EI construct and its also been advanced by accentuating how the or- relationship to a variety of life outcomes, includ- ganizational context constrains or encourages ing behavior at work (Mayer, Roberts, & Barsade, emotion norms (e.g., employees showing positive forthcoming), and we predict that the construct of emotions only during less busy times in conve- emotional intelligence, particularly if decon- nience stores or banks; Pugh, 2001; Rafaeli & structed into its component parts (e.g., the four Sutton, 1990). factors), will ultimately have much to offer to our Emotional labor has also been discussed in understanding of organizational life. terms of “surface acting,” when employees show emotions without necessarily feeling them (such Emotion Regulation and Emotional Labor as when an irked airline customer service agent A second focus has been on the degree to which forces himself to smile and be friendly as a cus- employees manage or regulate their emotional ex- tomer becomes increasingly agitated about lost pression through the facial “mask” they present to luggage); and “deep acting,” when employees dis- others. This perspective first notes that an em- play emotions they have actually worked on feel- ployee’s felt emotions can be distinguished from ing (such as if the airline customer service agent his or her displayed emotions, which are the facial actually tries to sympathize with the customer and expressions, gestures, tone of voice, and language show emotions aligned with feeling ). It used to convey feeling (Rafaeli & Sutton, 1989). was initially theorized that a discrepancy between Displaying emotions that differ from what we are individuals’ emotional display and their underly- actually feeling involves emotional regulation, ing feelings (characteristic of surface acting) which is the attempt to influence which emotions would cause “emotional dissonance” and contrib- we have, when we have them, and how these ute to work strain (Hochschild, 1983; Morris & emotions are experienced or expressed (Gross, Feldman, 1996). That is, when a customer service 1998a). It thus includes a broad range of regula- agent continuously forces himself to smile despite tory activities, including, as the introductory vi- feeling negative affect such as irritation, the dis- gnette suggests, controlling anger when a person sonance created may be a source of , de- feels that it will reflect badly on his or her repu- pression, and burnout (Grandey, 2003). However, tation. researchers have also found that for many workers When engaging in emotional labor, an organi- surface acting does not cause strain, particularly if zation-specific type of regulation, employees man- workers are “faking in good ” and believe the age their public displays of emotions to comply act they are putting on is a legitimate part of the with normative “display rules” (Ekman, 1973; work role (Rafaeli & Sutton, 1989: 37; see sum- Hochschild, 1983). Such organizational display mary in Coˆte´, 2005). Given these divergent find- rules or emotion norms can be used as a mecha- ings, researchers are working to understand the nism for increasing performance—such as sales- conditions under which emotion regulation in the people keeping an upbeat, enthusiastic expression form of surface acting does result in strain (Coˆte´, with customers to encourage purchasing behavior 2005), including the role of individual differences (Pugh, 2001; Totterdell & Holman, 2003); law- and organizational context (Grandey, 2000). One yers using an aggressive, angry tone to encourage possible clue to this question comes from emotion compliance in adversaries (Pierce, 1995); medical regulation research suggesting that regulating professionals adopting norms of intentional affec- emotions by anticipating them and engaging in tive neutrality (Smith & Kleinman, 1989); and cognitive re-framing (for example, telling oneself 42 Academy of Management Perspectives February in advance to be objective in a potentially emo- work groups, collective team mood convergence, a tionally charged situation) causes less strain than product of contagion, was found to occur, partic- regulating emotions through suppression (for ex- ularly in those with high group cohesion (Totter- ample, attempting to “surface act” cheerful when dell, Kellet, Teuchmann, & Briner, 1998). In a currently feeling angry) (Gross, 1998b). broader study of 70 work teams across 51 different organizational contexts (including product teams, Emotional Contagion and Collective Affect service teams, strategic planning teams, consult- The idea that affect not only occurs intrapsychi- ing teams, and engineering teams), mood conver- cally but has a strong social component which can gence occurred across all dimensions of the affec- influence dyadic and group interactions is a third tive circumplex (Bartel & Saavedra, 2000). In a emerging area of research (Barsade & Gibson, recent study of group contagion in a naturalistic 1998; Kelly & Barsade, 2001; Keltner & Haidt, team performance setting, not only was contagion 1999). The process of emotional contagion is a shown to occur, but this contagion was stronger primary mechanism through which emotions are for people who had a higher dispositional propensity shared and become social, creating collective toward emotional contagion, and also for those who emotion. had more collectivistic tendencies toward the team Emotional contagion, characterized as pro- (Ilies, Wagner, & Morgeson, in press). cesses that allow the sharing or transferring of emotions from one individual to other group Why Does Affect Matter in Organizations? members, often occurs without conscious knowl- e have laid the groundwork for understand- edge (although it can also be consciously induced; ing and answering this question by outlining Barsade, 2002). This everyday, continuous, auto- the parameters of affect as a construct. We matic process has been described as a tendency to W now focus on how affect of all sorts influences the mimic the nonverbal behavior of others, to “syn- thoughts and behaviors of individuals and groups chronize facial expressions, vocalizations, pos- within organizations. We do so by first examining tures, and movements” with others, and in turn, to a relationship that has long intrigued researchers: “converge emotionally” (Hatfield, Cacioppo, & whether employees’ feelings influence their work Rapson, 1994). Research has suggested this mim- performance. We then examine specific relation- icry can be explained by the facial feedback hy- ships between affect and decision making, creativ- pothesis, such that individuals who model certain ity, group dynamics, and individual behaviors, facial displays corresponding to emotions actually such as turnover, helping behavior, negotiation, begin to experience the same emotions (Larsen & and leadership. Finally, we draw conclusions from Kasimatis, 1990; Strack, Martin, & Stepper, these studies to indicate where we are and what 1988). The contagion process may be modified by challenges we face in exploring future directions a range of factors, such as the degree to which in organizational research. individuals are good senders and receivers of emo- tion (Hatfield et al., 1994; Sullins, 1989). Affect and Performance While studies exploring the influence of emo- The dominant hypothesis about employee emo- tional contagion initially focused primarily on dy- th adic settings, findings from both lab and field tions in the 20 century was that happy workers research suggest that contagion also functions at ought to be productive workers (Staw, Bell, & the group level. For example, in one study, con- Clausen, 1986). Up until the 1980s, however, tagion occurred and influenced group dynamics “happiness” was often measured using attitudinal measures such as job satisfaction, and the results both with the deliberate mood induction by the 5 presence of a trained confederate, and within were decidedly modest. Recently, measures that group dynamics without a confederate (Barsade,

2002). Moving to organizational settings, in an 5 Recent meta-analyses indicate that job satisfaction and performance in-depth daily tracking of nurse and accountant are correlated in the range of .17 (Iaffaldano & Muchinsky, 1985) to .30 2007 Barsade and Gibson 43 more directly measure happiness—those focusing contagion to be a useful mechanism for under- on state and trait positive affect—have been used standing performance outcomes. For example, the to examine this critical relationship. These studies positive mood of bank tellers was found to lead to have produced more compelling results. Indeed, a positive emotional contagion among their cus- comprehensive meta-analysis indicated that an tomers, which was then positively associated with individual’s tendency to experience positive emo- customer evaluations of service quality (Pugh, tions and moods is associated with increases in a 2001). In a cleverly done coder observation study variety of work performance measures, including of emotional contagion within 220 employee-cus- more positive supervisory evaluations, higher in- tomer encounters in coffee shops, behavioral come, enhanced negotiating ability, and perform- mimicry, an underlying mechanism of emotional ing discretionary acts for the benefit of the orga- contagion, was shown to occur (the strength of nization (Lyubomirsky, King, & Diener, 2005). the employees’ smiles predicted customers’ smiles Most studies of affect and work-related perfor- during the purchase encounter, even above the mance have examined employees’ dispositional degree of smiling the customers came into the affect (see Staw & Cohen-Charash, 2005 for a coffee shop with). This emotional contagion also review). For example, an experimental study fo- predicted customers’ satisfaction with the encoun- cusing on managerial performance found that dis- ter (Barger & Grandey, 2006). Similarly, a study positional positive affect was a significant predic- involving a short-term affect measure (though not tor of decision-making effectiveness, interpersonal quite a measure of mood) found that in a sample performance, and ratings of managerial potential of shoe salespeople, engaging in affectively posi- (Staw & Barsade, 1993). Longitudinal field re- tive behaviors with customers such as greeting, search has indicated that employees who tend to smiling, and eye contact was found to correlate meet work obstacles in a positive mood (using with customers’ in-store positive mood, which was both state and trait measures) tend to reap more then related to the amount of time the customers favorable outcomes—including more favorable su- spent in the store and their reported willingness to pervisor evaluations and higher pay 18 months come and shop there again (Tsai & Huang, 2002). after the initial measure of positive emotions— This effect of emotional contagion has also than their more negative counterparts (Staw, Sut- ton, & Pelled, 1994). Within the sales domain, been found at the group level. For example, in a salespeople’s general positive affect toward their simulated managerial group decision-making task, customers was found to significantly predict sales the degree to which individuals within the group performance (Sharma & Levy, 2003). Overall, experienced positive contagion predicted how research shows a consistent, strong relationship positively other group members rated their perfor- between trait positive and various mance (Barsade, 2002). In the same study, looking measures of work performance. This relationship at more collective outcomes, the degree to which has held in experimental, cross-sectional, and lon- groups experienced positive versus negative con- gitudinal studies, even after controlling for possi- tagion led to less conflict and greater cooperation ble confounding variables and using both objec- in the way money was allocated in the salary tive and subjective ratings (see review in decision-making task. Groups in which positive Cropanzano & Wright, 2001). emotional contagion occurred allocated the pot of When examining the influence of mood on money more evenly among the group members, as performance, researchers have found emotional compared to groups in which negative emotional contagion occurred. In a study of the influence of (Judge et al., 2001). A problem with using an attitudinal measure such as the contagion of mood of a group leader on group job satisfaction as a proxy for happiness, however, is that much of job satisfaction also involves a cognitive component—how employees think or members, the positive mood of the leader posi- feel about work, whether they like it or not, which is different from the tively influenced group members at both the in- experience of affect at work (Brief & Weiss, 2002; George, 1989), which is how an employee actually feels while on the job and their emotional dividual and collective level with the opposite for approach to life and work. leader negative mood. The leader’s positive mood 44 Academy of Management Perspectives February also had a subsequent influence on group coordi- Last, a more recent line of research examines nation and effort (Sy, Coˆte´, & Saavedra, 2005). whether an employee’s emotional intelligence Relatedly, while we have been focused on emo- leads to increased job performance, with most tional contagion as a transient state, one’s person- studies to date finding results with overall emo- ality trait propensity toward emotional contagion tional intelligence and the sub-factor of emotional has also begun to be linked to work outcomes. For recognition/perception (see Mayer et al., forth- example, it has been related to positive salesper- coming for a comprehensive review of the influ- son performance, but also to a greater likelihood ence of emotional intelligence on work out- of burnout (Verbeke, 1997), and greater vulnera- comes). Emotional intelligence has been found to bility to when faced with positively influence performance on problem solv- dealing with death and dying among oncology ing tasks (Lam & Kirby, 2002; Lyons & Schnei- care providers (LeBlanc, Bakker, Peeters, Van- der, 2005); and in a variety of managerial simula- Heesch, & Schaufeli, 2001). tions, including problem analysis in a managerial It is not clear whether dispositional affect or in-box , a layoff decision-making task, and mood has a stronger influence on performance, a simulated claims adjustment task (Day & Car- although there are theories which integrate the roll, 2004; Feyerham & Rice, 2002; Matsumoto et two (Weiss & Cropanzano, 1996). There have al., 2004). There have some field studies tying been more studies examining dispositional affect, emotional intelligence with performance, mainly but this may be because dispositional affect is within the domain of skills. A more easily measured than mood, and much easier recent meta-analysis reported a positive relation- than measuring discrete emotions. In an empirical ship with greater emotion perception (emotion test of state versus trait affective measures, a study recognition accuracy) and better work outcomes of public sector employees showed that trait mea- in occupations as diverse as physicians, medical sures (using a measure of positive psychological interns, human service workers, foreign service well-being) related to supervisor performance rat- officers, principals, public service interns, school ings beyond the effects of mood. A replication of teachers, business executives, clinicians and busi- the study with social welfare counselors repeated ness managers (Elfenbein, Foo, White, & Tan, in the trait findings, but also found that negative press). There are significantly fewer field studies mood was predictive of performance while con- looking at emotional intelligence more broadly trolling for trait measures. Thus, both types of than emotional perception. One such recent em- affect mattered in accounting for performance pirical examination studied 44 analysts and cleri- (Wright, Cropanzano, & Meyer, 2004). It has cal employees from the finance department of a been argued that the link between employees’ Fortune 500 insurance company and found that more short-lived feeling states (e.g. moods and the employees with higher emotional intelligence emotions) and performance measures such as su- ability scores (which varied by which of the four pervisory ratings are inconsistent because of the factors was being examined) received greater time lag problem: employees’ moods and emotions merit increases and were employed at a higher may be fleeting and short-term, while perfor- rank in the company. Employees with higher emo- mance measures used tend to reflect longer periods tional intelligence were also rated by both their of evaluation (usually six months to one year; supervisors and teammates as having better social Wright & Staw, 1999). As we review below, when skills than employees with lower emotional intel- performance is measured in more time-delimited ligence ability scores (Lopes et al., 2000). ways (such as through effective decision making, Overall, however, within field settings, the re- creativity, or prosocial behaviors), the results for search evidence tying emotional intelligence abil- positive mood are quite compelling. We encour- ities to work performance is still in its beginning age more research examining the influence of phases and has not yet lived up to the claims of its mood, and especially discrete emotions, on perfor- popular press fame. One possibility may be that mance (see Lazarus & Cohen-Charash, 2001). the measures for this construct need to be im- 2007 Barsade and Gibson 45 proved, or that outcome variables need to be physicians by providing a small gift of candy. The chosen more carefully. Another likely possibility physicians then had to read a description of a is that cognitive intelligence or other personality patient and think aloud (which was recorded and variables (Gohm, Corser, & Dalsky, 2005; Rubin, rated by outside coders) as they tried to determine Munz, & Bommer, 2005) are particularly impor- the correct diagnosis. It was found that while the tant in organizational settings and may interact positive affect-induced doctors considered as with emotional intelligence in a way that then many diagnoses as doctors who did not undergo influences performance. For example, one recent the positive mood induction (received no candy), study made exactly this point and found that the positive affect-induced doctors came to the among staff members in a public university, the correct solution significantly earlier than control influence of high emotional intelligence on per- participants, and were less likely to incorrectly formance was more pronounced for employees anchor on an incorrect hypothesis. This study who had lower cognitive intelligence scores (Coˆte´ suggests that positive affect can facilitate the thor- & Miners, 2006). Finally, it may be that the field ough, efficient, and flexible use of new informa- has not sufficiently considered the links between tion, which increases decision effectiveness. the subcomponents of emotional intelligence, tak- However, studies have also found that negative ing into account the entire “EI profile,” rather affect can lead to more effective decision making. than the skills separately. For example, if there are One set of studies shows that negative affect leads two managers who are both high in ability to read to more concentrated, detailed, and analytic pro- others’ emotions, but one is better than the other cessing (see Schwarz, Bless, & Bohner, 1991 for a in regulating her own emotions or those of others, review) while positive affect can lead to the op- the latter manager may well be more successful at posite (Melton, 1995; Mackie & Worth, 1989). A her job. set of studies in the clinical literature shows a While many of the preceding studies have used “depressive realism effect” in which people who quantitative task output as their measure of per- are depressed (trait affect) have more accurate formance, the direction of emotions research has judgments than nondepressed people (Alloy & been to examine a variety of other outcome vari- Abramson, 1988). Finally, the “mood-as-input” ables that may also be considered measures of model (Martin, Ward, Achee, & Wyer, 1993) “performance,” depending on the task context. predicts that negative affect will lead to more These include effective decision making, creativ- effortful processing. The rationale for this is that ity, turnover, prosocial behaviors, and leadership. people use their moods as indications of the state We turn to these variables next. of their environment. Thus, negative moods serve as a signal to people that something is wrong, so Affect and Decision Making that active cognitive processing will continue as The influence of affect on decision making has people try to solve the problem, whereas positive been an area of active debate focused on whether mood signals that all is well and people should not positive or negative affect leads to better out- continue to analyze. comes. Support for the influence of positive emo- Addressing this debate within the organiza- tions on decision making comes from a variety of tional domain, Staw and Barsade (1993) directly domains. Alice Isen and her colleagues, in a vo- tackled the question of whether positive versus luminous research stream, have consistently negative affect contributes to effective decision shown that positive mood inductions lead to bet- making. In a managerial simulation with MBA ter, more efficient decision making, including de- students, they offered competing hypotheses as to cision making requiring more careful, systematic, whether high versus low trait positive affect par- and thorough processing (see Isen, 2001; Isen & ticipants would do better in decision making on a Labroo, 2003 for reviews). For example, in one managerial simulation scored by outside raters. study in a hospital setting, Estrada, Isen, and They found a salutary influence of positive affect Young (1997) induced positive affect in practicing on a gamut of detailed, effortful decision making 46 Academy of Management Perspectives February tasks, including: greater decision making accuracy; Affect and Creativity greater amount of additional information re- A very similar set of competing arguments as were quested before making decisions; greater use of presented in the decision-making literature have quantitative indices in the decision making, and been made for the influence of positive versus greater recognition of situational contingencies. negative affect states and traits on creativity (see They thus concluded that positive affect led to James, Brodersen, & Jacob, 2004 for a review). better decision making than negative affect. Pos- Positive affect has been proposed to positively itive affect has also been linked to deeper analytic influence creativity by leading to a state in which processing and efficiency in decision making. In a more cognitive material—more variety in the el- study of graduating university students seeking ements that are considered—is available for pro- employment, those with higher trait positive af- cessing. Then, once those elements are available, fectivity had more clarity about their job search positive affect leads to a more complex, flexible (integrating information more deeply and effi- thinking, allowing a broader choice of elements to ciently), which then led them to look for a job come together and an increased chance that peo- more intensely and ultimately led to more inter- ple will in fact put together all of the cognitive views and job offers (while trait negative affectiv- elements that have become available (see Isen, ity was not found to be associated with job search 1999; Frederickson, 1998). There is strong support clarity; Coˆte´, Saks, & Zikic, 2006). A recent for this theory in laboratory studies of induced meta-analysis showed that the preponderance of positive affect and creativity, with myriad studies evidence indicates that positive emotions are bet- showing that when people are in more positive ter for myriad facets of decision making (see Ly- moods, they are more creative (see Isen, 1999 for ubomirsky, King, & Diener, 2005). Overall, it may a review of this literature). This support has re- be that positive mood allows people to better cently been extended to studies within organiza- process at a level most appropriate to the situation tions. In the first longitudinal study of daily work at hand. Thus, when more heuristic, quick an- creativity, a study examined self-reports, other- swers are needed, people in positive moods can ratings, and daily diary data from 222 employees respond with an appropriate decision making in seven companies over the length of an entire strategy; if the task requires deeper, more analytic project, directly addressing the question of processing, people who are in a good mood recog- whether negative versus positive affect would en- nize this necessity and can do so as well (Isen, hance creativity (Amabile, Barsade, Mueller, & 2004). Staw, 2005). This study found a strong linear There has been less examination of the in- relationship between greater positive mood and fluence of discrete emotions on decision mak- creativity in organizations. Also, the influence of positive affect on creativity lasted up to two days ing. However, a qualitative study of three Brit- after the positive mood had been felt. Another ish professional symphony orchestras offered an recent field study, conducted in the knitwear in- interesting model of how multiple negative dustry, also found facilitative effects for positive emotions such as fear, anxiety, , embar- mood on creative performance work by showing rassment, , anger, and influ- that positive but not negative moods mediated the enced decision making (Maitlis & Ozcelik, relationship between the support employees get 2004). The authors effectively used qualitative for creative work and their actual creative perfor- techniques to capture these phenomena. It mance (Madjar, Oldham, & Pratt, 2002). Overall, would be significantly more difficult to capture most research support is for a positive relationship the same dynamics quantitatively; this is likely between positive affect and creativity.6 one of the reasons there has been less research of discrete emotions conducted in organizations 6 While most evidence points to a strong positive relationship between and particularly the interaction of multiple dis- positive affect and creativity, it is important to note that there have been crete emotions (Maitlis & Ozcelik, 2004). some indications that the influence of negative affect on creativity should 2007 Barsade and Gibson 47

Affect and Turnover/Absence experience high positive moods. This could be Turnover and absence from work are critical or- because they feel higher self-efficacy in their skills ganizational variables, since the cost of replacing and ability to find a new job that will meet their employees and lost employee time is extremely values (George & Jones, 1996). high (see Cascio, 1991; Shaw, Delery, Jenkins, & Gupta, 1998). Several studies have linked affect Affect and Prosocial Behavior to these variables. In general, studies support the Prosocial behaviors are those undertaken to ben- idea that positive affectivity (both state and trait) efit or help another individual, group, or organi- is associated with reduced absence and intention zation (Brief & Motowidlo, 1986). It is well es- to turnover, and that negative affectivity (both tablished in the social psychological literature state and trait) is associated with increased ab- that positive mood is associated with helping be- sence, intention to turnover, and actual turnover havior in general (Isen, Clark, & Schwartz, 1976; (George & Jones, 1996; Pelled & Xin, 1999; Salovey, Mayer, & Rosenhan, 1991).7 Research Thoresen, Kaplan, & Barsky, 2003). Paying atten- has also been directed at the more specific ques- tion to the differing effects of positive and nega- tion of whether an employee’s positive mood will tive affect is, however, important. For example, in enhance prosocial organizational behaviors. Re- one study, the experience of positive moods (mea- searchers have found that employees who experi- sured as how employees felt “during the past ence positive moods at work are more likely to week”) caused employees to be absent less, but the engage in prosocial behavior both in terms of what experience of negative moods had no effect on their job requires (such as superior customer ser- their absence behavior (George, 1989: 321). vice) and aspects that go beyond their job descrip- Another study on the differential influence of tion (such as helping peers, or engaging in altru- positive and negative affect found an answer to a ism—George, 1991). In these studies, more question that has long puzzled researchers: why proximal positive moods have been shown to have does job dissatisfaction result in turnover inten- an effect, while dispositional affect appears to tions for some workers, but not for others? Studies have less of an effect. In one study, employees’ find that workers who are dispositionally higher in positive mood predicted prosocial behaviors such positive affect are more likely to leave their jobs if as altruism and helping, while dispositional affect they are dissatisfied than are people who are char- (measured as PA) had no effect (George, 1991). acterized by low positive affect (Judge, 1993; That is, unlike turnover or absenteeism, these Shaw, 1999). One reason behind this relationship helping behaviors appear to be more affected by is that for low positive affect individuals, the im- aspects of the immediate situation rather than an petus to quit is small because they do not expect a individual’s relatively stable tendency to experi- new job to be more satisfying. Conversely, high ence positive affect. trait positive affect individuals are more likely to Prosocial behavior has also been linked to be willing to change their situations when they are moods expressed by group leaders. A study of sales dissatisfied. Furthermore, the relationship be- associates in 37 retail stores examined whether a tween job satisfaction and the intent to turnover group leader’s positive mood contributed to the is strongest among people who feel that their extent to which the group engaged in prosocial values are not being met at work and tend to behavior and reduced the group’s voluntary turn- over rate. The findings were affirmative (George at least be considered. For example, there is some support for the influence of negative emotions on creativity from studies of affective illness (e.g., and manic-depression; Jamison, 1993; see Feist, 1999 for a 7 However, under the rubric of “motivated cognitive processing theory” review). There is also organizational evidence from a field study based on which indicates that people want to maintain their positive moods and on the “mood-as-input” model. In this cross-sectional study in a large manu- average avoid situations which would reduce their positive emotions (Clark facturing organization, there was a positive relationship between negative & Isen, 1982; Forgas, 1991), there is evidence that people in positive moods affect and creativity when both recognition and rewards for creativity, and are more prone to help if helping does not negatively influence their clarity of feelings (ability to recognize one’s own feelings) are high (George positive mood (Isen & Simmonds, 1978; Forest et al., 1979). But see Parrott & Zhou, 2002). (1993) for an additional perspective. 48 Academy of Management Perspectives February

& Bettenhausen, 1990). It was posited that leaders (Lyubromirsky et al., 2005). Negotiators in a pos- who experience high levels of positive mood at itive mood (usually induced in experimental set- work would tend to feel “active, excited, enthusi- tings) tend to be more cooperative and less likely astic, peppy, and strong,” and this enthusiasm for to engage in conflict, and in some cases, come to the work task would “rub off” on group members, agreements that enhance joint gains more fre- producing an increased incidence of prosocial be- quently (Baron, 1990; Barsade, 2002). Positive haviors (George & Bettenhausen, 1990: 701). mood induces individuals to adopt more innova- The logic behind this emotional connection to tive problem-solving strategies, suggesting that prosocial behaviors is that positive mood leaders these negotiators will be more likely to come to would be more likely to encourage and notice integrative (“win-win”) agreements (Carnevale & positive behaviors performed by the group and Isen, 1986). Positive mood in a negotiator is likely positively reinforce the group. This study suggests to create more positive feelings in his or her coun- the power of the leader’s mood in shaping group terpart, and liking between negotiators has been members’ perceptions of the group and their be- linked to added flexibility in the negotiation havior toward each other. when opponents know each other (Druckman & Overall, there is strong support for the idea that Broome, 1991). Positive mood is also related to positive emotions make prosocial behaviors more persistence and increased levels in ne- likely. Some studies also suggest a corollary: that gotiators, which have been associated with in- negative emotions make anti-social behaviors creased outcomes (Kumar, 1997). Last, a face-to- more likely. One study argued that discrete nega- face negotiation study showed that negotiators tive emotions produced by environments that are can be easily and effectively instructed in how to perceived as unjust or stressful increase the fre- be emotionally strategic in the emotions they dis- quency of anti-social or deviant organizational play (an interesting result in its own right). It also behaviors (Spector & Fox, 2002). This approach showed that positive negotiator emotional display suggests that in organizational environments (or (as compared to neutral or negative displays) led groups) that encourage positive affective states, to a desire by negotiating partners to want to employees are more likely to engage in prosocial, continue doing business with the positive negoti- supportive, and cooperative behaviors. The oppo- ator. The positive negotiator was also better able site is predicted in organizational environments to close a deal in a distributive (win-lose) setting, that foster employees’ negative emotions (see even through increased concessions from the Frost, 2004). other party (Kopelman, Rosette, & Thompson, 2006). Affect and Negotiation and Conflict Resolution Currently, the findings for feeling and display- Conflict is an inherent part of organizational life ing negative emotions in a negotiation are typi- and frequently causes strong emotional responses cally the inverse of the findings regarding the in the conflicting groups and individuals. Negoti- feeling and display of positive moods and emo- ation is the primary means by which organization tions. For example, negotiators in a generally neg- members manage their conflict (Allred, Mallozzi, ative mood were more competitive and received & Matsui, 1997). After many years of either ig- poorer outcomes (Forgas, 1998). However, re- noring emotions or emphasizing practical advice search examining negative moods in conflict to show neutral emotions and use the proverbial management and negotiation are also more likely “poker face” (Gibson & Schroeder, 2002), nego- to examine discrete emotions, which can offer tiation scholars have begun to recognize the im- more nuanced insights. For example, one study portance of emotions and how emotions influence examining the discrete emotions of anger and the negotiation process (see Barry, Fulmer, & Van found that negotiators who felt high Kleef, 2004; Thompson, Nadler, & Kim, 1999 for anger and low compassion for their counterpart reviews). The results of these studies generally achieved fewer joint gains in their show that positive moods help to resolve conflict (Allred et al., 1997). The study also found that 2007 Barsade and Gibson 49 those negotiators’ discrete emotions of anger and One way to conceptualize and compassion influenced their negotiations more its outcomes is a “bottom-up” approach where than their generalized positive or negative moods group emotion is defined by the affective compo- did. There have also been results showing positive sition of the various affective attributes of the effects of anger, finding that a negotiator facing an group’s members (Barsade & Gibson, 1998). For angry counterpart is more likely to concede than a example, a group’s affective tone, the “consistent negotiator facing a happy counterpart. However, or homogenous affective reactions within a group” these effects occurred under high time pressure measured by the group’s mean level of positive and when the negotiator who faced the angry affect, was found in retail sales groups to be posi- counterpart had lower power, and in a computer- tively related to higher levels of customer service mediated setting (Van Kleef, De Dreu, & Man- and lower absenteeism (George, 1995). stead, 2004). Computer-mediated settings are A different way of looking at group emotion is clearly important to organizational life (e.g., in through affective diversity, or the degree of differ- the use of e-mail), but it is critical to see how the ence in affective traits that exists between group effects of anger operate in face-to-face negotia- members. Affective diversity has been shown to tions as well. influence group outcomes. In a sample of 239 top Last, looking at emotional intelligence skills, managers in 62 U.S. corporations, the greater the emotional perception has been shown to influ- degree of trait affective diversity on the senior ence individual negotiator outcomes in a variety management team, the greater the conflict in the of sometimes contradictory ways. More consis- team, the less cooperation and the poorer the firm tently, they have been shown to increase the financial performance (Barsade, Ward, Turner, & integrative outcomes of negotiating dyads (Elfen- Sonnenfeld, 2000). There are a few studies that bein et al., in press; Foo, Elfenbein, Tan, & Aik, examined the effect of discrete emotions in 2004). Emotion understanding skill was shown to groups, such as group envy, which was found to be positively influence how one’s negotiation partner directly associated with decreased group perfor- felt about his/her negotiation outcome, above and mance. It was also associated with more absentee- beyond the amount of money that negotiation ism, less group satisfaction, and poorer group per- partner received and his/her trait positive affect formance via the mechanisms of increased social (Mueller & Curhan, in press). loafing, decreased cohesion, and decreased feel- ings of group potency (Duffy & Shaw, 2000). There is also a “top-down” approach in which Collective Affect and Team Behavior collectively held norms—implicit or explicit— The role of affect has long been an implicit factor about appropriate emotions to express or hold in in studies of groups, for example, in studies of the group and/or organization, shape the type of group cohesiveness (Ashforth & Humphrey, emotions that are allowed and expressed in the 1995) and the progression of group development group context (see Barsade & Gibson, 1998; Kelly (Tuckman, 1965; Wheelan, 1994). However, & Barsade, 2001, for reviews). There has been there are few studies examining how affect oper- much research examining emotion norms within ates as an explicit factor within team development, an emotional labor perspective (e.g., see reviews behavior, and outcomes. This is surprising given by Grandey, 2000; Rafaeli & Sutton, 1989), but that in the process of getting work accomplished, significantly less examining the influence of “af- groups offer a prime place for intense interactions fective culture.” Affective culture can be thought involving individuals with their own emotional of as normative systems which include display histories, emotional agendas, and affective person- rules about expressed emotions at the collective alities confronting positive and negative group level, prescribing the appropriateness or inappro- events. Of the studies that have been conducted, priateness of particular emotional expressions in however, there is very promising evidence for the the organization (Barsade, Brief, & Spataro, 2003; influence of emotion on group outcomes. Barsade & O’Neill, 2004). Overall, while we 50 Academy of Management Perspectives February would predict that the influence of affective cul- service field setting, leaders’ positive moods were ture on group and individual dynamics would be a found to be associated with higher performance of powerful one, of the many areas we have exam- the leader’s group (George, 1995). ined showing the influence of collective affect on Transformational leadership is a setting in workplace outcomes, this is currently one of the which the importance of emotions in leadership least studied and most open for development. effectiveness has been specifically emphasized (Ashkanasy & Tse, 2000).8 For example, a recent study examining the effects of emotional intelli- Affect and Leadership gence and personality traits on transformational We conclude with a critical—but complex—pro- leadership behavior found that leaders with high cess within organizations, the process of leader- trait positive affectivity were more likely to be ship. It has become increasingly apparent that rated as engaging in transformational leadership emotions permeate the leadership process, both in behaviors. As might be expected, trait PA was not terms of the emotions leaders feel and express, and a significant predictor of transactional or contin- the emotions followers feel toward their leaders gent reward behavior (Rubin, Munz, & Bommer, (see George, 2000). Leaders must substantially 2005). This study also found an intriguing inter- regulate their own emotions. For example, they action: the researchers focused on ability to read often must express a positive or upbeat mood others emotions (using the DANVA test) as a about the future, while suppressing expressions of particularly important dimension of emotional in- anxiety or sadness that might de-motivate follow- telligence that should relate to transformational ers. They must also manage the emotions of oth- leadership behavior. But as they predicted—and ers, for example, by understanding and empathiz- found—the leadership personality trait of extra- ing with employees’ emotions about change so version (being outgoing and deriving energy from that change efforts will be accepted (Huy, 2002). other people) moderated this relationship. Specif- While the notion of emotions as critical to the ically, while extraversion alone did not have a leadership process is not new (see, for example, direct effect on transformational leadership be- Ashforth & Humphrey, 1995; Wasielewski, havior, extraversion combined with emotion rec- 1985), recent advances in emotions research and ognition skills did have an effect. Thus, high ex- emotional intelligence in particular have sparked traversion provided a clear benefit to leaders who an increase in the study of leadership and emo- also possess the ability to accurately recognize tion. emotion. Conversely, leaders who “possessed low In terms of positive affectivity and leadership, extraversion and high abili- the work cited above on prosocial behavior, cre- ties did not seem to reap the benefits of their ativity, and decision making suggest that PA emotion recognition ability” (Rubin, Munz, & should contribute to leader effectiveness. There is Bommer, 2005: 854). These findings point to an some empirical support for this connection. In a important characteristic of emotional intelligence simulated managerial setting, high trait positive that we referred to earlier: it does not operate affect MBA students were rated by their peers and separately, but rather in conjunction with other outside observers as being better leaders (Staw & abilities and personality traits. Barsade, 1993). Trait positive affectivity was re- While leadership researchers have emphasized lated to leader-follower liking and perceived sim- the critical place of followers in determining lead- ilarity in a simulated interview setting (Fox & Spector, 2000), and in a lab study, leaders in 8 Transformational leadership is characterized by a leader’s ability to positive versus negative moods had groups who articulate a shared vision of the future, intellectually stimulate employees, motivate colleagues and followers to look beyond their own interests and performed better in their task, expending less un- towards group interests, and provide individual consideration and support necessary effort and more coordination in com- for followers (Bass, 1998; Lowe, Kroeck, & Sivasubramaniam, 1996). Transformational leadership has been contrasted with transactional lead- pleting the task (Sy, Coˆte´, & Saavedra, 2005). ership, which is based on motivating followers by emphasizing reward and Testing this within organizations, in a customer exchange relationships. 2007 Barsade and Gibson 51 ership style and shaping leadership behavior, tional state. The leader must then craft a response much less work has focused on followers’ emotions to the situation that takes into account the emo- in response to leadership. Several researchers em- tional tone, and communicates that response ef- phasize that leader effectiveness is at least partially fectively (Pescosolido, 2002). defined by the satisfaction and emotional liking of Overall, work in emotions and leadership is followers (Ashforth & Humphrey, 1995; Conger emerging as a very exciting area that will enhance & Kanungo, 1987; Dasborough & Ashkanasy, our knowledge of what leadership means and how 2002). There is also evidence that followers are leaders can be effective, but it is also an area in influenced by leaders’ displays of emotions. As which work needs to be done to sharpen con- noted above, leaders’ expressions of positive emo- structs more effectively so that our understanding tions are thought to arouse positive emotions in of the intersection of these two domains can be others through the mechanism of emotional con- better understood. tagion, where the positive, upbeat emotions of the leader are emulated by followers, resulting in pos- Conclusions itive outcomes (George & Bettenhausen, 1990; his article offers a review of “what we know” Hatfield et al., 1994). about emotions in organizations at the present Laboratory studies have also examined the ef- Ttime. The review, albeit not exhaustive, indi- fects of leader displays of negative emotions and cates that the study of affect in organizations is a found more complex results. Following a conta- vibrant and growing area. It is characterized by a gion argument, leader displays of negative emo- wide breadth of approaches, developing measures, tions could cause followers to similarly feel and and refinement of variables and outcomes. Orga- display negative emotions, potentially hindering nization researchers are increasingly recognizing morale and motivation. Leader expressions of neg- that affect is inherent to the human experience, ative emotions such as sadness and anger, for and thus inherent to any situation in which hu- example, have been shown to influence how em- mans interact with each other and their environ- ployees view the leader, reducing their percep- ment, including at work. We draw the following tions of leader effectiveness (Lewis, 2000). How- conclusions from this wide range of studies. ever, recent studies also show that a leader expressing anger may increase perceptions of the Affect influences critical organizational variables leader’s power, while a leader expressing sadness This article has identified a range of ways that may decrease those perceptions (Tiedens, 2001). affect is critical to explaining outcomes that con- Displays of negative emotion by the leader may cern managers in organizations. We have outlined also focus followers’ attention on situations that effects on performance, decision making, turn- require attention. For example, a leader’s anger over, prosocial behavior, negotiation and conflict about an issue of discrimination or fairness may resolution behavior, group dynamics, and leader- direct resources to solving the problem (George, ship. These are discrete categories that help us to 2000). group variables as scientific studies have con- Leadership and emotion studies are also just ceived them, but it is our view that affect perme- beginning to examine the more detailed processes ates virtually every aspect of organizational life, and interactions involved in a leaders’ manage- even those areas that have been traditionally ment of their teams’ emotional responses (e.g., thought of as the exclusive province of cognitive Huy, 2002). A recent model examines how lead- behavior, such as decision making and task per- ers can “set the emotional tone” of a group and use formance. The evidence is overwhelming that ex- emotional skills to focus group members on goals. periencing and expressing positive emotions and In order to do this, the emergent leader of a group moods tends to enhance performance at individ- must first empathize and identify the collective ual, group, and organizational levels. As a recent emotional state of a group and also understand the meta-analysis has shown, positive affect is funda- aspects of the situation that are causing this emo- mentally linked with an individual’s “active in- 52 Academy of Management Perspectives February volvement with goal pursuits and with the envi- (Lewis, 2000), and potentially lead to aggression ronment” (Lyubomirsky et al., 2005: 804). This or violence (Fox & Spector, 1999). However, desire to develop new goals and engage with them negative emotions (especially anger) may also is linked with confidence, , self-efficacy, draw our attention to situations of unfairness and likability, activity, energy, flexibility, and coping injustice (George, 2000), enhance perceptions of with challenges and stress, among other abilities power (Tiedens, 2001), and enhance negotiating and behaviors. The evidence is compelling that outcomes (Van Kleef, De Dreu, & Manstead, feeling and expressing positive affect is critical to 2004). Research and practice should be directed success in organizations and in life. to the important questions of, “Under what con- It is particularly ironic that while positive affect ditions can negative affective responses lead to has been found to show greater influence on work- positive organizational outcomes?” To do so, it place outcomes, it has been studied significantly would be helpful for emotion scholars to focus on less than negative affect (Lyubomirsky et al., examining the various discrete negative emotions, 2005; Thoresen et al., 2003). A rationale for why as the outcomes that will come from angry versus positive affect has been found to have a relatively anxious versus sad employees, for example, are stronger effect on these outcomes could be that likely going to be very different. positive affect has been shown to consistently be related more strongly than negative affect to so- Constructs and methods are advancing cially related processes (McIntyre, Watson, Clark, No longer is there a “one size fits all” way to & Cross, 1991; Watson, Clark, McIntyre, & Ha- measure work-related affect, such as using general maker, 1992), which are particularly critical to attitudinal measures like job satisfaction. Drawing effective organizational interactions. Negative af- from and contributing to the robust literature of fect, on the other hand, is more strongly related to affect in psychology, our understanding of affect non-social intrapsychic outcomes, such as stress has been both expanded and refined via the study and burnout (Watson et al., 1988). Thus both sets of discrete emotions, the affective circumplex, of emotions serve a role in important outcomes, emotional labor, emotional contagion, and emo- but in different arenas. tional intelligence. Methods are becoming more varied and sophisticated to match the variety and The influence of negative affect is complex complexity of the phenomena, so that in addition Conclusions about the meaning and influence of to surveys, methods include controlled mood in- negative affect on organizational life are far more ductions, diary studies, daily experience sampling complex. The history of reactions to negative em- research, coding of behavior in-situ and video ployee affect has tended to be simplistic: managers coding. ought to avoid negative affect in their employees and suppress negative affect in themselves (see Future Directions Stearns & Stearns, 1986). Given the power of hile much current work is being directed to positive affect identified above, this approach is refining the variables and relationships we understandable. However, current research has Whave examined above, we also anticipate helped us to appreciate more of the complexity of new approaches to studying affect. First, extant negative affective responses, allowing us to be studies of affect assume that most important affec- more nuanced in our approach. First, we must tive experiences arise through face-to-face inter- acknowledge that the evidence for the deleterious actions, and that most of emotional communica- effects of individual negative affect is substantial, tion occurs through facial, or at least auditory particularly since they tend to be strongly felt by communication, with very little occurring employees (Miner, Glomb, & Hulin, 2005). Neg- through text (Mehrabian, 1972). However, the ative affective expressions can poison organiza- impact of an entirely text-based technology on tional cultures (Aquino, Douglas, & Martinko, emotions must be explored. Significant communi- 2004), negatively influence perceptions of leaders cation in organizations now takes place through 2007 Barsade and Gibson 53 synchronous (e.g., instant messaging) or asynchro- experience (Kihlstrom, 1999). That is, rather than nous (e.g., e-mail) text-based means, which re- being consciously driven, individuals’ current moves critical nonverbal sources of emotion and emotions may be reflections of their “implicit tone. How can emotions be best conveyed via memory” of past events, which may create moods these media? What is the effect of conveying that are out of our awareness (e.g., Singer & emotionally charged messages via text, when Salovey, 1988) and implicit perceptions and emo- these messages are more likely to be misconstrued? tions (see Kihlstrom, 1999, for a review). How must we re-think emotional contagion and Much of this future work will likely need to other social processes in an organizational world consider that people do not walk into organiza- in which many meetings take place online? In- tions as tabula rasa, but rather have life and work deed, a recent study examining e-mail versus face- experiences that may shape current behavior— to-face communication suggests that individuals either consciously or unconsciously. People may tend to be overconfident in their ability to accu- not always be aware of this, as is exemplified in rately convey the emotions they wish via e-mail, the phenomenon of transference, where “repre- particularly when they are trying to be sarcastic or sentations of significant others, stored in memory, humorous (Kruger, Epley, Parker, & Ng, 2005). are activated and used in new social encounters on The use of emoticons (:-)) may be somewhat the basis of a new person’s resemblance to a given helpful in this regard, but are also open to sub- significant other” (Berk & Andersen, 2000: 546; stantial misinterpretation or may be perceived as also known as the “You vaguely remind me of that unprofessional in the business context. Video con- kid in elementary school who I hated, and I don’t ferencing, also increasing in its use, has more cues, like you much either” phenomenon–Kelly & Bar- but is also not yet the same as interacting face to sade, 2001: 109). Such inquiry may well spark new face, particularly in group situations. Given that research on long-ignored constructs such as trans- these technologies continue to grow as a primary ference, ego defensive routines, and attachment means of communication within the business relationships and their effect on individuals’ be- world, it is crucial that we understand how the haviors in organizations. The benefit now is our interpretation and communication of affect occurs ability to conduct rigorous empirical research to in these contexts. better help us understand how these phenomena Second, research on affect has primarily fo- occur (Glassman & Andersen, 1999; Westen & cused on conscious feelings and expressions, those Gabbard, 2002). moods and emotions we are aware of and can Last, the research findings we cite for negotia- possibly trace to their source and are thus amena- tion and social influence suggest that affect can be ble to regulation. However, there is also substan- used strategically: individuals can “put on” partic- tial developing research on affective processes ex- ular emotional expressions in order to influence isting at a level below consciousness: emotions others (Gibson & Schroeder, 2002; Kopelman et existing at the subconscious or unconscious level al., 2006). However, the emotion labor literature that nonetheless have an impact on our conscious has also emphasized that there is a cost to masking feelings and behavior. Subconscious affective pro- authentic emotions—by acting like we’re feeling cesses include the automatic mimicry of others’ something we actually aren’t, we may experience emotions characterizing emotional contagion, as emotional dissonance and lose touch with our we have discussed earlier (Hatfield, Cacioppo, & authentic selves (see Hochschild, 1983). This re- Rapson, 1994), and automatic emotion regulation search suggests that employees would be better off (Mauss, Evers, Wilhelm, & Gross, 2006). These if they could engage in less emotion regulation and processes also include our “emotional uncon- that employees need organizations where they can scious,” which can be explained as an individual express themselves more authentically (Erickson being consciously aware of his or her current emo- & Wharton, 1997). There is a paradox, however, tional state, but not being aware of the source of in the assumption that authenticity is the desired that state, which may come from a current or past state for employees. We know that authenticity in 54 Academy of Management Perspectives February our feeling and expression of affect is desirable; ate and influence organizational outcomes is an however, we also know that regulating emotions is essential piece in understanding how work is done often essential: in order for managers to be en- and how to do it better. couraging, inspiring, and motivating to their em- ployees (despite having a bad day, for example) References they must engage in regulation in order to be effective. Part of the job is to be strategic with our Alloy, L.B., & Abramson, L.Y. (1988). Depressive realism: Four theoretical perspectives. In L.B. Alloy (Ed.), Cog- emotions; indeed, emotional regulation of self and nitive process in depression (pp. 223–265). New York: others is an important part of emotional intelli- Guilford. gence construct (Mayer & Salovey, 1997). The Allred, K.G., Mallozzi, J.S., & Matsui, F. (1997). The influ- question is, at what point do individuals “over- ence of anger and compassion on negotiation perfor- mance. Organizational Behavior and Human Decision Pro- regulate” their emotions? What is the point at cesses, 70(3), 175–187. which regulated emotion is too far removed from Amabile, T.M., Barsade, S.G., Mueller, J.S., & Staw, B.M. authenticity? Current studies of the differing an- (2005). Affect and creativity at work. Administrative Sci- tecedents of surface and deep acting (e.g., ence Quarterly, 50(3), 367–403. Aquino, K., Douglas, S., & Martinko, M. J. (2004). Overt Grandey, 2000) may hold a clue to these ques- anger in response to victimization: Attributional style tions, but more work remains. Thus, the paradox and organizational norms as moderators. Journal of Oc- we need to explore is that authenticity may be cupational Psychology, 9, 152–164. desirable, but regulation is essential to meeting Ashforth, B.E., & Humphrey, R.H. (1995). Emotion in the workplace: A reappraisal. Human Relations, 48(2), 97– personal and organizational goals. 125. To that end, we would like to see researchers Ashkanasy, N.M., & Tse, B. (2000). Transformational lead- explore to what degree and under what conditions ership as management of emotion: A conceptual review. individuals in organizations can and should ex- In N. Ashkanasy, C. Ha¨rtel, & W. Wilfred (Eds.), Emo- tions in the workplace: Research, theory, and practice (pp. press their authentic emotions, and how an orga- 221–235). US: Quorum Books/Greenwood Publishing nization’s affective culture and the national cul- Group, Inc. ture in which it is embedded may influence these Barger, P., & Grandey, A. (2006). “Service with a smile” and encounter satisfaction: Emotional contagion and processes. We also urge researchers to explore the appraisal mechanisms. Academy of Management Journal, ethical implications of being inauthentic as part of 49(6), 1229–1238. the work role. Is emotional labor, which involves Baron, R.A. (1990). Environmentally induced positive regulating and changing emotions to fit work re- affect: Its impact on self–efficacy, task performance, ne- gotiation, and conflict. Journal of Applied Social Psychol- quirements, something that organizations should ogy, 20(5, Pt 2), 368–384. be able to expect from their employees as a nec- Barrick, M.R., & Mount, M.K. (1991). The Big Five per- essary part of the job? We are inclined to think sonality dimensions and job performance: A meta-anal- that this is acceptable as long as employees know ysis. Personality Psychology, 44, 1–26. Barry, B., Fulmer, I.S., & Van Kleef, G. (2004). I laughed, I what they have signed up for, and that this emo- cried, I settled: The role of emotion in negotiation. In tional labor has logical performance outcomes fa- M. J. Gelfand and J. M. Brett (Eds.), The Handbook of vorable to the company (see Rafaeli & Sutton, Negotiation and Culture: Theoretical Advances and Cross– 1989), but could see opposing views to this and Cultural Perspectives (pp. 71–94). Palo Alto, CA: Stan- ford University Press. encourage a thorough discussion of this issue Barsade, S.G. (2002). The ripple effect: Emotional conta- within the field. gion and its influence on group behavior. Administrative Why does affect matter in organizations? The Science Quarterly, 47(4), 644–675. state of the literature shows that affect matters Barsade, S., Brief, A.., & Spataro, S. (2003). The affective revolution in organizational behavior: The emergence of because employees are not isolated “emotional a paradigm. In J. Greenberg (Ed.), Organizational islands.” Rather, they bring all of themselves to Behavior: The State of the Science (pp. 3–52). London: work, including their traits, moods, and emotions, Lawrence Erlbaum Associates, Publishers. and their affective experiences and expressions Barsade, S.G., & Gibson, D.E. (1998). Group emotion: A view from top and bottom. In D. Gruenfeld (Ed.), Com- influence others. Thus, an understanding of how position (pp. 81–102). US: Elsevier Science/JAI Press. these affective experiences and expressions oper- Barsade, S.G., & O’Neill O.A.. (2004). Affective Organi- 2007 Barsade and Gibson 55

zational Culture. Academy of Management Presenta- worker is really a ‘productive’ worker: A review and tion. New Orleans, LA. further refinement of the happy-productive worker the- Barsade, S.G., Ward, A.J., Turner, J.D.F., & Sonnenfeld, sis. Consulting Psychology Journal: Practice and Research, J.A. (2000). To your heart’s content: A model of affec- 53(3), 182–199. tive diversity in top management teams. Administrative Dasborough, M.T., & Ashkanasy, N.M. (2002). Emotion Science Quarterly, 45(4), 802–836. and attribution of intentionality in leader-member rela- Bartel, C., & Saavedra, R. (2000). The collective construc- tionships. Leadership Quarterly, 13(5), 615–634. tion of work group moods. Administrative Science Quar- Daus, C.S., & Ashkanasy, N.M. (2005). The case for the terly, 45(2), 197–231. ability-based model of emotional intelligence in organi- Bass, B.M. (1998). Transformational leadership: Industrial, zational behavior. Journal of Organizational Behavior, 26, military, and educational impact. Mahwah, NJ: Lawrence 453–466. Erlbaum Associates, Publishers. Day, A.L., & Carroll, S.A. (2004). Using an ability-based Berk, M.S., & Andersen, S.M. (2000). The impact of past measure of emotional intelligence to predict individual relationships on interpersonal behavior: Behavioral con- performance, group performance, and group citizenship firmation in the social-cognitive process of transference. behaviours. Personality and Individual Differences, 36, Journal of Personality and Social Psychology, 79, 546–562. 1443–1458. Brackett, M.A., Mayer, J.D., & Warner, R.M. (2004). Emo- Doherty, R.W. (1997). The emotional contagion scale: A tional intelligence and its relation to everyday behavior. measure of individual differences. Journal of Nonverbal Personality and Individual Differences, 36(6), 1387–1402. Behavior, 21, 131–154. Brackett, M.A., Rivers, S.E., Shiffman, S, Lerner, N., & Druckman, D., & Broome, B.J. (1991). Value differences Salovey, P. (2006). Relating emotional abilities to social and conflict resolution: Familiarity or liking? Journal of functioning: A comparison of self-report and perfor- Conflict Resolution, 35(4), 571–593. mance measures of emotional intelligence. Journal of Duffy, M.K., & Shaw, J.D. (2000). The Salieri syndrome: Personality and Social Psychology, 91, 780–95. Consequences of envy in groups. Small Group Research, Brief, A.P., & Motowidlo, S.J. (1986). Prosocial organiza- 31(1), 3–23. tional behaviors. Academy of Management Review, 11(4), Ekman, P. (1973). Cross-culture studies of facial expression. 710–725. In P. Ekman (Ed.), Darwin and facial expression: A century Brief, A.P., & Weiss, H.M. (2002). Organizational behavior: of research in review (pp. 169–222). New York: Academic Affect in the workplace. Annual Review of Psychology, Press. 53(1), 279–307. Ekman, P. (1992). An argument for basic emotions. Cogni- Carnevale, P.J., & Isen, A.M. (1986). The influence of tion & Emotion, 6(3-4), 169–200. positive affect and visual access on the discovery of Elfenbein, H.A., Foo, M.D., White J., & Tan, H..H. (in integrative solutions in bilateral negotiation. Organiza- tional Behavior and Human Decision Processes, 37(1), press). Reading your counterpart: The benefit of emotion 1–13. recognition accuracy for effectiveness in negotiation. Cascio, W.F. (1991). Applied psychology to human resource Journal of Nonverbal Behavior. management (5th ed.). Upper Saddle River, NJ: Prentice- Elfenbein, H.A., & Ambady, N. (2002). Predicting work- Hall. place outcomes from the ability to eavesdrop on feelings. Clark, M.S., & Isen, A.M. (1982). Towards understanding Journal of Applied Psychology, 87(5), 963–971. the relationship between feeling states and social behav- Erickson, R. J., & Wharton, A. S. (1997). Inauthenticity ior. In Albert H. Hastorf and Alice M. Isen (Eds.), and depression: Assessing the consequences of interac- Cognitive social psychology (pp. 73–108). New York: tive service work. Work and Occupations, 24(2), 188– Elsevier-North Holland. 213. Conger, J.A., & Kanungo, R.N. (1987). Toward a behav- Estrada, C.A., Isen, A.M., & Young, M.J. (1997). Positive ioral theory of charismatic leadership in organizational affect facilitates integration of information and decreases settings. Academy of Management Review, 12(4), 637– anchoring in reasoning among physicians. Organizational 647. Behavior and Human Decision Processes, 72(1), 117–135. Conte, J.M. (2005). A review and critique of emotional Feist, G.J. (1999). The influence of personality on artistic intelligence measures. Journal of Organizational Behavior, and scientific creativity. In R. Sternberg (Ed.), Handbook 26(4), 433–440. of creativity (pp. 273–296). New York, NY: Cambridge Coˆte´, S. (2005). A social interaction model of the effects of University Press. emotion regulation on work strain. Academy of Manage- Feldman Barrett, L., & Russell, J.A. (1998). Independence ment Review, 30(3), 509–530. and bipolarity in the structure of current affect. Journal of Coˆte´, S., & Miners, C.T.H. (2006). Emotional intelligence, Personality and Social Psychology, 74(4), 967–984. cognitive intelligence, and job performance. Administra- Feyerherm, A.E., & Rice, C.L. (2002). Emotional intelli- tive Science Quarterly, 51, 1–28. gence and team performance: The good, the bad and the Coˆte´, S., Saks, A.M., & Zikic, J. (2006). Trait affect and job ugly. International Journal of Organizational Analysis, 10, search outcomes. Journal of Vocational Behavior, 68(2), 343–362. 233–252. Foo, M.D., Elfenbein, H.A., Tan, H.H., & Aik, V.C. Cropanzano, R., & Wright, T.A. (2001). When a ‘happy’ (2004). Emotional intelligence and negotiation: The 56 Academy of Management Perspectives February

tension between creating and claiming value. Interna- given. Journal of Personality and Social Psychology, 77, tional Journal of Conflict Management, 15, 411–29. 1146–1162. Forest, D., Clark, M.S., Mills, J., & Isen, A.M. (1979). Gohm C.L., Corser G.C., & Dalsky, D.J.(2005). Emotional Helping as a function of feeling state and nature of the intelligence under stress: Useful, unnecessary, or irrele- helping behavior. Motivation & Emotion, 3, 161–169. vant? Personality and Individual Differences, 39, 1017–28. Forgas, J.P. (1998). On feeling good and getting your way: Grandey, A.A. (2000). Emotional regulation in the Mood effects on negotiator cognition and bargaining workplace: A new way to conceptualize emotional labor. strategies. Journal of Personality and Social Psychology, 74, Journal of Occupational Health Psychology, 5(1), 95–110. 565–577. Grandey, A.A. (2003). When “the show must go on”: Sur- Forgas, J.P. (1991). Affect and person perception. In Joseph face acting and deep acting as determinants of emotional P. Forgas (Ed.), Emotion and social judgments (pp. 263– exhaustion and peer-rated service delivery. Academy of 290). Oxford: Pergamon. Management Journal, 46(1), 86–96. Fox, S., & Spector, P.E. (2000). Relationship of emotional Gross, J.J. (1998a). The emerging field of emotion intelligence, practical intelligence, general intelligence, regulation: An integrative review. Review of General Psy- and trait affectivity with interview outcomes: It’s not all chology, 2(3), 271–299. just “g.” Journal of Organizational Behavior, 21(2), 203– Gross, J.J. (1998b). Antecedent and response-focused emo- 220. tion regulation: Divergent consequences for experience, Fox, S., & Spector, P.E. (1999). A model of work frustra- expression, and physiology. Journal of Personality and tion-aggression. Journal of Organizational Behavior, 20(6), Social Psychology, 74(1), 224–237. 915–931. Hatfield, E., Cacioppo, J., & Rapson, R. (1994). Emotional Frederickson, B.L. (1998). What good are positive emo- contagion. New York: Cambridge University Press. tions? Review of General Psychology, 2, 300–319. Hochschild, A. (1983). The managed heart: Commercializa- Frijda, N.H.. (1986). The emotions. Cambridge, UK: Cam- tion of human feeling. Berkeley: University of California bridge University Press. Press. Frost, P.J. (2004). Handling Toxic Emotions: New chal- Hogan, R., Hogan, J., & Roberts, B. W. (1996). Personality lenges for leaders and their organization. Organizational measurement and employment decisions. American Psy- chologist, 51(5), 469–477. Dynamics, 33(2), 111–127. Huy, Q.N. (2002). Emotional balancing of organizational George, J.M. (1989). Mood and absence. Journal of Applied continuity and radical change: The contribution of mid- Psychology, 74(2), 317–324. dle managers. Administrative Science Quarterly, 47(1), George, J.M. (1991). State or trait: Effects of positive mood 31–69. on prosocial behaviors at work. Journal of Applied Psy- Iaffaldano, M.T., & Muchinsky, P.M. (1985). Job satisfac- chology, 76(2), 299–307. tion and job performance: A meta-analysis. Psychological George, J.M. (1995). Leader positive mood and group Bulletin, 97(2), 251–273. performance: The case of customer service. Journal of Ilies, R., Wagner, D.T., & Morgeson, F.P. In press. Explain- Applied Social Psychology, 25(9), 778–794. ing affective linkages in teams: Individual differences in George, J.M. (2000). Emotions and leadership: The role of susceptibility to contagion and individualism/collectiv- emotional intelligence. Human Relations, 53(8), 1027– ism. Journal of Applied Psychology. 1055. Isen, A. (1999). On the relationship between affect and George, J.M., & Bettenhausen, K. (1990). Understanding creative problem solving. In S.W. Russ (Ed.), Affect, prosocial behavior, sales performance, and turnover: A creative experience and psychological adjustment (pp. 3–18). group-level analysis in a service context. Journal of Ap- Philadelphia: Brunner/Mazel. plied Psychology, 75(6), 698–709. Isen, A.M. (2001). An influence of positive affect on deci- George, J.M., & Jones, G.R. (1996). The experience of work sion making in complex situations: Theoretical issues and turnover intentions: Interactive effects of value at- with practical implications. Journal of Consumer Psychol- tainment, job satisfaction, and positive mood. Journal of ogy, 11(2), 75–85. Applied Psychology, 81(3), 318–325. Isen, A.M. (2004). Some perspectives on positive feelings George, J.M., & Zhou, J. (2002). Understanding when bad and emotions: Positive affect facilitates thinking and moods foster creativity and good ones don’t: The role of problem solving. In A. S. R. Manstead, N. Frijda, & A. context and clarity of feelings. Journal of Applied Psychol- Fischer (Eds.), Feelings and emotions: The Amsterdam sym- ogy, 87(4), 687–697. posium (pp. 263–281). Cambridge, UK: Cambridge Uni- Gibson, D.E., & Schroeder, S. (2002). Grinning, frowning, versity Press. and emotionless: Agent perceptions of power and their Isen, A.M., Clark, M., & Schwartz, M. F. (1976). Duration effect on felt and displayed emotions in influence at- of the effect of good mood on helping: “Footprints on the tempts. In N. Ashkanasy, C. Hartel, and W. Zerbe sands of time.” Journal of Personality and Social Psychology, (Eds.), Managing Emotions in the Workplace (pp. 184– 34, 385–393. 211). Armonk, NY: M. E. Sharpe. Isen, A.M., & Labroo, A.A. (2003). Some ways in which Glassman, N. S., & Andersen, S. M. (1999). Activating positive affect facilitates decision making and judgment. transference without consciousness: Using significant- In S.L. Schneider, & J. Shanteau (Eds.), Emerging per- other representations to go beyond what is subliminally spectives on judgment and decision research. Cambridge se- 2007 Barsade and Gibson 57

ries on judgment and decision making (pp. 365–393). New LeBlanc, P.M., Bakker, A.B., Peeters, M.C.W., Van Heesch, York, NY: Cambridge University Press. N.C.A., & Schaufeli, W.B. (2001). Emotional job de- Isen, A. M., & Simmonds, S. F. (1978). The effect of feeling mands and burnout among oncology care providers. Anx- good on a helping task that is incompatible with good iety, Stress and Coping, 14, 243–263. mood. Social Psychology, 41, 346–349 Lewis, K.M. (2000).When leaders display emotion: How James, K., Brodersen, M., & Jacob, E. (2004). Workplace followers respond to negative of affect and workplace creativity: A review and prelimi- male and female leaders. Journal of Organizational Behav- nary model. Human Performance, 17(2), 169–194. ior, 21, 221–234. Jamison, K.R. (1993). Touched with fire: Manic-depressive Lopes, P.N., Grewal, D., Kadis, J., Gall, M., & Salovey, P. illness and the artistic temperament. New York: The (2006). Evidence that emotional intelligence is related Free Press. to job performance and affect and attitudes at work. Judge, T.A. (1993). Does affective disposition moderate the Psicothema, 18, 132–138. relationship between job satisfaction and voluntary turn- Lowe, K.B., Kroeck, K.G., & Sivasubramaniam, N. (1996). over? Journal of Applied Psychology, 78(3), 395–401. Effectiveness correlates of transformation and transac- Judge, T.A., Thoresen, C.J., Bono, J.E., & Patton, G.K. tional leadership: A meta-analytic review of the MLQ (2001). The job satisfaction-job performance relation- literature. Leadership Quarterly, 7(3), 385–425. ship: A qualitative and quantitative review. Psychological Lyons, J.B., & Schneider, T.R. (2005). The influence of Bulletin, 127(3), 376–407. emotional intelligence on performance. Personality and Kelly, J., & Barsade, S. (2001). Mood and emotions in small Individual Differences, 39(4), 693–703. groups and work teams. Organizational Behavior and Hu- Lyubomirsky, S., King, L., & Diener, E. (2005). The benefits man Decision Processes, 86 (1), 99–130. of frequent positive affect: Does happiness lead to suc- Keltner, D., & Haidt, J. (1999). Social functions of emo- cess? Psychological Bulletin, 131(6), 803–855. tions at four levels of analysis. Cognition & Emotion, Mackie, D.M., & Worth, L.T. (1989). Differential recall of 13(5), 505–521. subcategory information about in-group and out-group Kihlstrom, J. F. (1999). The psychological unconscious. In members. Personality and Social Psychology Bulletin, 15(3), L. A. Pervin & O. P. John (Eds.), Handbook of 401–413. personality: Theory and research (2 ed., pp. 424–442). Madjar, N., Oldham, G.R., & Pratt, M.G. (2002). There’s New York: Guilford. no place like home? The contributions of work and Kopelman, S., Rosette, A.S., & Thompson, L. (2006). The nonwork creativity support to employees’ creative per- three faces of Eve: Strategic displays of positive, nega- formance. Academy of Management Journal, 45(4), 757– tive, and neutral emotions in negotiations. Organiza- 767. tional Behavior and Human Decision Processes, 99(1), 81– Maitlis, S., & Ozcelik, H. (2004). Toxic decision processes: 101. A study of emotion and organizational decision making. Kring, A.M., Smith, D.A., & J.M. Neale. (1994). Individual Organization Science, 15(4), 375–393. differences in dispositional expressiveness: Development Martin, L.L., Ward, D.W., Achee, J.W., & Wyer, R.S. and validation of the Emotional Expressivity Scale. Jour- (1993). Mood as input: People have to interpret the nal of Personality and Social Psychology, 66(5), 934–949. motivational implications of their moods. Journal of Per- Kruger, J., Epley, N., Parker, J., & Ng, Z. (2005). Egocen- sonality and Social Psychology, 64(3), 317–326. trism over e-mail: Can we communicate as well as we Matsumoto, D., LeRoux, J.A., Bernhard, R., & Gray, H. think? Journal of Personality and Social Psychology, 89(6), (2004). Unraveling the psychological correlates of inter- 925–936. cultural adjustment potential. International Journal of In- Kumar, R. (1997). The role of affect in negotiations: An tercultural Relations, 28, 281–309. integrative overview. Journal of Applied Behavioral Sci- Matthews, G., Roberts, R.D., & Zeidner, M. (2004). Seven ence, 33(1), 84–100. myths about emotional intelligence. Psychological In- Lam, L.T., & Kirby, S.L. (2002). Is emotional intelligence quiry, 15(3),179–196. an advantage? An exploration of the impact of emo- Mauss, I. B., Evers, C., Wilhelm, F. H., & Gross, J. J. (2006). tional and general intelligence on individual perfor- How to bite your tongue without blowing your top: mance. Journal of Social Psychology, 142(1), 133–143. Implicit evaluation of emotion regulation predicts affec- Larsen, R. J., & Diener, E. (1987). Affect intensity as an tive responding to anger provocation. Personality and individual difference characteristic: A review. Journal of Social Psychology Bulletin, 32(5), 589–602. Research in Personality, 21, 1–39. Mayer, J. D., Roberts, R. D., & Barsade, S. G. (forthcom- Larsen, R., & Kasimatis, M. (1990). Individual differences ing). Emerging research in emotional intelligence. An- in entrainment of mood to the weekly calendar. Journal nual Review of Psychology, 59. of Personality and Social Psychology, 58, 164–171. Mayer, J.D. & Salovey, P. (1997). What is emotional intel- Lazarus, R. S. (1991). Emotion and adaptation. New York: ligence? In P. Salovey & D. Sluyter (Eds.), Emotional Oxford University Press. development and emotional intelligence: Educational impli- Lazarus, R. S., & Cohen-Charash, Y. (2001). Discrete emo- cation (pp. 3–31). New York: Basic Books. tions in organizational life. In R. Payne & C. L. Cooper Mayer, J.D., Salovey, P., Caruso, D.R., & Sitarenios, G. (Eds.), Emotions in organizations (pp.45–81). Chichester, (2003). Measuring emotional intelligence with the MS- UK: Wiley. CEIT V2.0. Emotion, 3(1), 97–105. 58 Academy of Management Perspectives February

Mayer, J.D., Salovey, P., & Caruso, D. R. (2000). Models of intelligence. Current Directions in Psychological Science, emotional intelligence. In R.J. Sternberg (Ed.), Hand- 14(6), 281–285. book of Intelligence (pp.396–420). Cambridge, England: Salovey, P., & Mayer, J. (1990). Emotional intelligence. Cambridge University Press. Imagination, Cognition and Personality, 9, 185–211. McIntyre, C.W., Watson, D., Clark, L.A., & Cross, S.A. Salovey, P., Mayer, J.D., & Rosenhan, D.L. (1991). Mood (1991). The effect of induced social interaction on pos- and helping: Mood as a motivator of helping and helping itive and negative affect. Bulletin of the Psychonomic So- as a regulator of mood. In M.S. Clark (Ed.), Prosocial ciety, 29(1), 67–70. behavior (pp. 215–237). Thousand Oaks, CA: Sage. McCrae, R.R.., & Costa, P.T. (1987). Validation of the Schwarz, N., Bless, H., & Bohner, G. (1991). Response five-factor model of personality across instrument and scales as frames of reference: The impact of frequency observers. Journal of Personality and Social Psychology, range on diagnostic judgments. Applied Cognitive Psychol- 52(1), 81–90. ogy, 5(1), 37–49. Mehrabian, A. (1972). Nonverbal communication. Chicago: Sharma, A., & Levy, M. (2003). Salespeople’s affect toward Aldine-Atherton. customers: Why should it be important for retailers? Melton, R.J. (1995). The role of positive affect in syllogism Journal of Business Research, 56(7), 523–528. performance. Personality and Social Psychology Bulletin, Shaw, J.D. (1999). Job satisfaction and turnover intentions: 21(8), 788–794. The moderating role of positive affect. Journal of Social Miner, A.G., Glomb, T.M., & Hulin, C. (2005). Experience Psychology, 139(2), 242–244. sampling mood and its correlates at work. Journal of Shaw, J. D., Delery, J. E., Jenkins, Jr., G. D., & Gupta, N. Occupational and Organizational Psychology, 78(2), 171– (1998). An organization-level analysis of voluntary and 193. involuntary turnover. Academy of Management Journal, Morris, J.A., & Feldman, D.C. (1996). The dimensions, 41(3), 511–535. antecedents, and consequences of emotional labor. Singer, J. A., & Salovey, P. (1988). Mood and memory: Academy of Management Review, 21(4), 986–1010. Evaluating the network theory of affect. Clinical Psychol- Ortony, A., & Turner, T. J. (1990). What’s basic about basic ogy Review, 8, 211–251. emotions? Psychological Review, 97(3), 315–331. Smith, A.C., & Kleinman, S. (1989). Managing emotions in Mueller, J., & Curhan, J. (in press). Emotional Intelligence medical school: Students’ contacts with the living and and Counterpart Mood Induction in a Negotiation. In- the dead. Social Psychology Quarterly, 52(1), 56–69. ternational Journal of Conflict Management. Spector, P. E., & Fox, S. (2002). An emotion-centered Parrott, W.G. (1993). Beyond : Motives for inhib- model of voluntary work behavior: Some parallels be- iting good moods and for maintaining bad moods. In tween counterproductive work behavior and organiza- D.M. Wegner & J.W. Pennebaker (Eds.), Handbook of tional citizenship behavior. Human Resource Management mental control. Englewood Cliffs, NJ: Prentice-Hall. Review, 12, 269–292. Pelled, L.H., & Xin, K.R. (1999). Down and out: An inves- Staw, B.M., & Barsade, S.G. (1993). Affect and managerial tigation of the relationship between mood and employee performance: A test of the sadder-but-wiser vs. happier- withdrawal behavior. Journal of Management, 25(6), and-smarter hypothesis. Administrative Science Quarterly, 875–895. 38(2), 304–331. Pescosolido, A.T. (2002). Emergent leaders as managers of Staw, B.M., Bell, N.E., & Clausen, J.A. (1986). The dispo- group emotion. The Leadership Quarterly, 13, 583–599. sitional approach to job attitudes: A lifetime longitudi- Pierce, J.L. (1995). Gender trials: Emotional lives in contem- nal test. Administrative Science Quarterly, 31(1), 56–77. porary law firms. Berkeley, CA: University of California Staw, B.M., & Cohen-Charash, Y. (2005). The dispositional Press. approach to job satisfaction: More than a mirage, but not Pugh, S.D. (2001). Service with a smile: Emotional conta- yet an oasis: Comment. Journal of Organizational Behav- gion in the service encounter. Academy of Management ior, 26(1), 59–78. Journal, 44(5), 1018–1027. Staw, B.M., Sutton, R.I., & Pelled, L.H. (1994). Employee Rafaeli, A., & Sutton, R. I. (1989). The expression of positive emotion and favorable outcomes at the work- emotion in organizational life. In B. M. Staw & L. L. place. Organization Science, 5(1), 51–71. Cummings (Eds.), Research in organizational behavior (pp. Stearns, C.Z., & Stearns, P.N. (1986). Anger: The struggle for 1–42). Greenwich, CT: JAI Press. emotional control in America’s history. Chicago: University Rafaeli, A., & Sutton, R.I. (1990). Busy stores and demand- of Chicago Press. ing customers: How do they affect the display of positive Strack, F., Martin, L., & Stepper, S. (1988). Inhibiting and emotion? Academy of Management Journal, 33(3), 623– facilitating conditions of the human smile: A nonobtru- 637. sive test of the facial feedback hypothesis. Journal of Rubin, R.S., Munz, D.C., & Bommer, W.H. (2005). Leading Personality and Social Psychology, 54, 768–776. from within: The effects of emotion recognition and Sullins, E. (1989). Perceptual salience as a function of personality on transformational leadership behavior. nonverbal expressiveness. Personality and Social Psychol- Academy of Management Journal, 48(5), 845–858. ogy Bulletin, 15, 584–595. Russell, J. A. (1980). A circumplex model of affect. Journal Sutton, R.I. (1991). Maintaining norms about expressed of Personality and Social Psychology, 39, 1161–1178. emotions: The case of bill collectors. Administrative Sci- Salovey, P., & Grewal, D. (2005). The science of emotional ence Quarterly, 36, 245–268. 2007 Barsade and Gibson 59

Sy, T., Coˆte´, S., & Saavedra, R. (2005). The contagious proach. Journal of Personality and Social Psychology, 87(4), leader: Impact of the leader’s mood on the mood of group 510–528. members, , and group processes. Jour- Verbeke, W. (1997). Individual differences in emotional nal of Applied Psychology, 90(2), 295–305. contagion of salespersons: Its effect on performance and Tellegen, A.. (1985). Structures of mood and personality burnout. Psychology & Marketing, 14(6), 617–636. and their relevance to assessing anxiety, with an empha- Wasielewski, P.L. (1985). The emotional basis of charisma. sis on self-report. In A.H. Tuma, & J.D. Maser (Eds.), Symbolic Interaction, 8(2), 207–222. Anxiety and the anxiety disorders (pp. 681–706). Hillsdale, Watson, D., & Clark, L. A. (1984). Negative affectivity: NJ: Erlbaum. The disposition to experience negative emotional states. Tett, R.P., Fox, K.E., & Wang, A. (2005). Development and Psychological Bulletin, 96, 465–490. validation of a self-report measure of emotional intelli- Watson, D., Clark, L.A., McIntyre, C.W., & Hamaker, S. gence as a multidimensional trait domain. Personality and (1992). Affect, personality and social activity. Journal of Social Psychology Bulletin, 31(7), 859–888. Personality and Social Psychology, 63, 1011–1025. Thompson, L.L., Nadler, J., & Kim, P.H. (1999). Some like Watson, D., Clark, L. A., & Tellegen, A. (1988). Develop- it hot: The case for the emotional negotiator. In L.L. ment and validation of brief measures of positive and Thompson, J.M. Levine, & D.M. Messick (Eds.), Shared negative affect: The PANAS scales. Journal of Personality cognition in organizations: The management of knowledge and Social Psychology, 54, 1063–1070. (pp. 139–161). Mahwah, NJ: Erlbaum. Watson, D., & Tellegen, A. (1985). Toward a consensual Thoresen, C.J., Kaplan, S.A., & Barsky, A.P. (2003). The structure of mood. Psychological Bulletin, 98(2), 219–235. affective underpinnings of job perceptions and attitudes: Weiss, H. M. (2002). Deconstructing job satisfaction: Sep- A meta-analytic review and integration. Psychological arating evaluations, beliefs and affective experiences. Bulletin, 129(6), 914–945. Human Resources Management Review, 12, 173–194. Tiedens, L.Z. (2001). Anger and advancement versus sad- Weiss, H. M., & Cropanzano, R. (1996). Affective events ness and subjugation: The effect of negative emotion theory: A theoretical discussion of the structure, causes expressions on social status conferral. Journal of Person- and consequences of affective experiences at work. In ality and Social Psychology, 80(1), 86–94. B.M. Staw & L. L. Cummings (Eds.), Research in organi- Totterdell, P., & Holman, D. (2003). Emotion regulation in zational behavior (Vol. 18, pp. 1–74). Greenwich, CT: customer service roles: Testing a model of emotional JAI Press. labor. Journal of Occupational Health Psychology, 8(1), Westen, D., & Gabbard, G. O. (2002). Developments in 55–73. cognitive neuroscience: I. Conflict, compromise, and Totterdell, P., Kellet, S., Teuchmann, K., & Briner, R. connectionism. Journal of the American Psychoanalytic (1998). Evidence of mood linkage in work groups. Jour- Association, 50(1), 53–98. nal of Personality and Social Psychology, 74, 1504–1515. Wheelan, S.A. (1994). Group processes: A developmental Tsai, W. C., & Huang, Y. M. (2002). Mechanisms linking perspective. Needham Heights, MA: Allyn & Bacon. employee affective delivery and customer behavioral in- Wright, T.A., Cropanzano, R., & Meyer, D.G. (2004). State tentions. Journal of Applied Psychology, 87(5), 1001– and trait correlates of job performance: A tale of two 1008. perspectives. Journal of Business and Psychology, 18(3), Tuckman, B.W. (1965). Developmental sequence in small 365–383. groups. Psychological Bulletin, 63(6), 384–399. Wright, T. A., & Staw, B. M. (1999). Affect and favorable Van Kleef, G.A., De Dreu, C.K.W., & Manstead, A.S.R. outcomes: Two longitudinal tests of the happy-produc- (2004). The interpersonal effects of emotions in tive worker thesis. Journal of Organizational Behavior, negotiations: A motivated information processing ap- 20(1), 1–23.