Who can do this Annotated Bibliography on Job Satisfaction?

Article5.pdf

Home >Business & Finance homework help >Management homework help >Who can do this Annotated Bibliography on Job Satisfaction?

Good acting requires a good cast: A meso-level model of deep acting in work teams

WILLIAM J. BECKER1* AND RUSSELL CROPANZANO2 1Department of Management, Neeley School of Business, Texas Christian University, Ft. Worth, Texas, U.S.A. 2Department of Management and Entrepreneurship, Leeds School of Business, University of Colorado, Boulder, Colorado, U.S.A.

Summary This study proposes and tests a meso-level model of deep acting in work teams that draws on emotional con- tagion theory to explain how shared means of complying with display rules can arise in work teams. We argue that the presence of influential deep actors can lead to greater convergence (lower dispersion) on individual deep acting in the team. That is, team members behave more similarly. When a team has greater convergence, deep acting by individual members should be related to lower emotional exhaustion and higher job satisfaction and in-role performance. In a sample of mature work teams, these hypotheses received general support. Our findings suggested that team-level deep acting effects can foster benefits for team members (lower emotional exhaustion and higher satisfaction) and organizations (higher job performance). Copyright © 2014 John Wiley & Sons, Ltd.

Keywords: emotional labor; deep acting; teams

Since the publication of Hochschild’s (1979) seminal work over two and a half decades ago, scholars have taken an active interest in investigating emotional labor. There is now a considerable body of evidence suggesting that orga- nizations regulate the expression of emotions by employee through formal and informal expectations (i.e., “display rules”; Diefendorff & Greguras, 2009; Rafaeli, 1989). Evidence further suggests that conforming to these display rules frequently conflicts with felt emotions, thereby requiring emotional labor by workers (Cropanzano, Weiss, & Elias, 2004; Diefendorff & Gosserand, 2003; Grandey, 2000). When employees successfully deep act and sincerely express the feelings normatively endorsed by display rules, then important individual and organizational outcomes can be enhanced (Cropanzano et al., 2004; Grandey, 2003). Many organizations have moved toward organizing employees into semi-autonomous work teams (Mathieu,

Maynard, Rapp, & Gilson, 2008). The growing use of teams, rather than individually structured work, has implica- tions for deep acting. When employees work in interdependent groups, deep acting becomes part of the shared ex- perience of their work teams (cf. Ashkanasy, 2003; Gibson, 1997). With this in mind, the present study contributes to the emotional labor literature by proposing and testing a meso-level model of deep acting in the context of work teams. Meso-level models simultaneously investigate multiple levels of analysis in order to understand how team and individual variables are linked to each other (House, Rousseau, & Thomas-Hunt, 1995). As depicted in Figure 1, we propose a meso-level model that includes both individual deep acting and unit level deep acting dispersion. We provide a brief overview of our framework here, before the more detailed literature review that follows. In our framework, which seeks to integrate previous research within the team context, the emotional labor process

is initiated when workers are collectively expected to display situationally appropriate emotions (Shields, 2005). In care and service roles, this usually involves displaying positive emotions while eschewing or reducing negative emo- tions (e.g., Côté & Morgan, 2002; Diefendorff, Richard, & Croyle, 2006). When individuals are able to comply with display rules by deep acting and genuinely changing their underlying feelings, they should experience reduced

*Correspondence to: William J. Becker, Department of Management, Neeley School of Business, Texas Christian University, TCU Box 298530, Ft. Worth, TX 76129, U.S.A. E-mail: [email protected]

Revised 09 September 2014, Accepted 20 September 2014

Journal of Organizational Behavior, J. Organiz. Behav. 36, 232–249 (2015) Published online 23 October 2014 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/job.1978

R esearch

A rticle

emotional exhaustion. Subsequently, emotional exhaustion is posited to affect important work outcomes such as job satisfaction and job performance. We extend this traditional approach to the emotional labor story by adding meso-level effects to the model. Based

on emotional contagion theory, we argue that emotional displays produced by deep acting can spread through recep- tive teams. Over time, this can lead to increased deep acting by other team members. As a result, these teams will demonstrate greater consensus around deep acting (i.e., a dispersion model). We predict that when there is greater team agreement in deep acting, individual deep acting will be negatively related to emotional exhaustion. Next, we will discuss our model in detail, focusing on the evidence for each stage of our framework.

Deep Acting Overview

Organizations govern display rules because the expression of affect has been tied to important organizational out- comes (Diefendorff & Gosserand, 2003). For instance, Tsai (2001; see also Tsai & Huang, 2002) provided evidence that positive emotional displays by store clerks tended to produce favorable customer reactions. Display rules can arise as formal or informal norms and become a shared attribute of social groups (e.g., Diefendorff, Erickson, Grandey, & Dahling, 2011). Individuals who work within such units feel pressure to display expected emotional ex- pressions even in difficult situations. If display rules represent collective expectations for emotional expressions during the performance of one’s work,

then emotional labor has come to describe the methods that employees utilize to conform to those expectations. In this sense, display rules represent demands, and emotional labor represents behaviors necessary to meet these de- mands and perform effectively (Ashkanasy & Humphrey, 2011). Grandey (2000) argued that there were two broad tactics for complying with display rules in the workplace—deep acting and surface acting (for a more recent anal- ysis, see Scott & Barnes, 2011). Deep acting occurs when a desired emotional expression is achieved by changing one’s underlying felt emotion. For instance, a worker in an amusement park might “psych himself up” to interact with customers (Van Maanen & Kunda, 1989). In this case, the employee actually becomes more enthusiastic than would otherwise be the case. In contrast, surface acting refers to producing a desired outward emotional expression without modifying the underlying felt emotion (Gosserand & Diefendorff, 2005; Scott & Barnes, 2011). Our present theoretical model is predicated on the sharing of genuine emotional states (e.g., emotional contagion;

Barsade, Brief, & Spataro, 2003; Barsade & Gibson, 1998), which help to create a shared affective tone (cf. Collins, Lawrence, Troth, & Jordan, 2013). Specifically, our conceptual framework is grounded in research that pertains to

Figure 1. Meso-level model of deep acting in teams

DEEP ACTING IN WORK TEAMS 233

experienced emotions, as opposed to superficial expressions. For this reason, deep acting—emotional labor that pro- duces authentic emotional states—is most relevant (Grandey, 2003). Empirical evidence supports this contention. Deep acting has been consistently linked with emotional contagion (Barsade & O’Neill, 2014; Brotheridge & Lee, 2002; Hennig-Thurau, Groth, Paul, & Gremler, 2006). Relative to surface acting, deep acting also tends to be associated with more favorable affective, social, and professional outcomes than surface acting (Scott & Barnes, 2011). For example, Grandey (2003) found that workers who engaged in deep acting produced more effective service delivery than those who engaged in surface acting. Therefore, in this study, we focused on deep acting.

Meso-level effects on deep acting in teams

Deep acting has generally been conceptualized at the individual level of analysis (Grandey, 2000; Scott & Barnes, 2011). This approach has inspired a great deal of important research but has an important limitation. When individuals are part of a team, this approach does not capture the social nature of emotions or the impact of one employee’s emo- tions on those of their teammates (Ashkanasy & Ashton-James, 2005). As interdependence within an organizational group increases, so does the effect of expressed emotions on others (Ashkanasy, 2003; Barsade & Gibson, 2007; Gibson, 1997). That is, over time teammates are likely to experience convergence in their affective states (Totterdell, 2000; Totterdell, Kellet, Teuchmann, & Briner, 1998). In fact, Rimé (2009) argued convincingly that an exclusively individual-level view of emotional regulation is no longer tenable. Reflecting the growing importance of teams in or- ganizations, there has been a recent call to investigate emotional labor at the team level (Mesmer-Magnus, DeChurch, & Wax, 2012).

How and why affect is shared Our theoretical basis for deep acting at the team level begins with the principle that teams engage in episodic inter- actions over time (Marks, Mathieu, & Zaccaro, 2001). Affective events and subsequent emotional responses are key aspects of these interactions, and individual team members often consider the reactions and displays of others when evaluating their own response and feelings toward an event (Collins et al., 2013; Rimé, 2009). There are various rea- sons why this is so, but two are especially relevant here. The first comes from research on emotional labor, and the latter from research on emotional contagion. As we shall see, although these mechanisms are drawn from related, but distinct, research literatures, they are mutually reinforcing. First, according to emotional labor research, teams frequently develop display rules that specify the type of affec-

tive display that is most appropriate in particular work settings (Rafaeli & Sutton, 1990; Van Maanen & Kunda, 1989). Team members tend to enforce violations of these display rules and buttress conformance (Barsade & Gibson, 1998; Kelly & Barsade, 2001). Notice that the presence and enforcement of group norms is a team-level phenomenon. This suggests that teammates share display rules and hold one another accountable to them (Diefendorff et al., 2011). As a result, coworkers will often show a degree of convergence in their feelings (George, 2002; Scott & Barnes, 2011). Second, emotional contagion research has shown that authentic and appropriate emotions can be transferred

among group members, often through implicit processes (Barsade, 2002; Hatfield, Cacioppo, & Rapson, 1994; Hatfield, Rapson, & Le, 2011). Even though a felt emotion originates from a teammate, each member feels and expresses the emotion genuinely and perceives it as his or her own (Barsade, Ramarajan, & Westen, 2009). For ex- ample, Totterdell (2000) found that professional cricket players tended to share the feelings of their teammates, while Totteredell et al. (1998) reported similar findings for accountants. In both cases, the agreement in moods could not be entirely explained by shared, objective events. Often, people are not aware of their shared responses (see Chartrand & Bargh, 1999, especially their Study 1). Moreover, this convergence has even been found for autonomic responses (Kaplan, Burch, & Bloom, 1964). Seen in this light, deep acting by individual team members can spread through the team through emotional contagion.

234 W. J. BECKER AND R. CROPANZANO

As a result of both enforcement of display rules and emotional contagion among team members, deep acting should be considered at the team level. Our framework suggests a meso-level paradigm because the individual and group level pro- cesses are inextricably linked and must be considered simultaneously (House et al., 1995). In this case, deep acting by one or more team members gives rise to meso-level processes, such as display rule enforcement and emotional contagion, that produce stable team-level characteristics. Recall that similar points have also been made by other investigators of work- place emotion (Diefendorff et al., 2011; Martin, Knopoff, & Beckman, 1998). In this model, lower-level measures of deep acting also contribute to team-level characteristics. If multiple or influential team members favor deep acting and other team members are susceptible to emotional contagion, then the team as a whole will tend to converge on deep acting. It bears mention that our arguments here are consistent with the concept of “leading with emotional labor”

outlined by Humphrey, Ashkanasy, and their colleagues (Ashkanasy & Humphrey, 2011; Humphrey, 2002; Humphrey, Pollack, & Hawver, 2008). Here, we add that coworkers, rather than just formal leaders, are also able to shape the emotions of their teammates with authentic emotional displays. Barsade and O’Neill (2014) suggested an analogous role for deep acting in the development of a culture of companionate love. Dasborough, Ashkanasy, Tee, and Tse (2009) also proposed a similar, but reciprocal, framework by which negative emotional displays shape attitudes toward leaders through the emotional contagion process (see also Ashkanasy & Nicholson, 2003).

Within-team dispersion and the need for a meso-level model These meso-level processes are dynamic and difficult to measure directly. In this initial test of our framework, we investigated mature and stable work teams where the meso-level processes should have been mostly crystallized into stable team characteristics. Therefore, we adopted a dispersion composition model of team deep acting (Chan, 1998). Conceptually, this team-level construct reflected the extent to which team members converged (or diverged) in their level of deep acting during difficult emotion eliciting events (Bliese, 2000). While display rules and emo- tional contagion often push team members to share affective experiences, this does not necessarily always occur. For example, some groups may be more or less strict about enforcing affective norms. Likewise, a number of factors influence team members’ susceptibility to emotional contagion (Ilies, Wagner, & Morgeson, 2007). For one, individuals are more likely to catch the emotions of others that they care for and identify with (Hatfield et al., 1994). Similarly, the findings of Totterdell and colleagues (2000; 1998) indicated that linkages were particularly strong in mutually interde- pendent teams. Emotional contagion is also more likely to originate from the emotions of influential team members such as charismatic members and formal or informal leaders (Bono & Ilies, 2006; Collins et al., 2013). It is also enhanced when multiple team members produce similar emotion displays (Barsade, 2002; Bartel & Saavedra, 2000). This sug- gests that there will be group differences in the level of convergence. Therefore, we make the following prediction:

Hypothesis 1: In mature teams, some teams will demonstrate greater deep acting convergence than others.

Deep acting dispersion and its consequences

These meso-level effects for deep acting extend our understanding of deep acting by grounding it in the social group, suggesting that mature groups can converge on patterns of mutual deep acting. Sincere changes in one’s feelings that align with display rules should be adaptive for individual team members (Diefendorff & Gosserand, 2003; Gross, 2002). To understand this more fully, our framework suggests that we need to consider the dispersion of deep acting within each team in order to accurately predict when these benefits will accrue.

Deep acting and emotional exhaustion Regardless of the team-level influences, individual behaviors have individual-level consequences. According to Grandey (2000), deep acting involves altering one’s emotions in order to conform to display rules; it ultimately re- duces the tension caused by internal inconsistency and requires expenditure of resources only while the change is

DEEP ACTING IN WORK TEAMS 235

being enacted. Hence, it should be negatively related to emotional exhaustion. Despite the logic of Grandey’s con- ceptual position, the empirical results to date have not been entirely supportive. In a study of working adults from multiple occupations, Brotheridge and Grandey (2002) found that deep acting and emotional exhaustion were unre- lated. This pattern of findings was replicated in research by Brotheridge and Lee (2002) and Judge, Woolf, and Hurst (2009). Some studies have even found that deep acting is positively related to emotional exhaustion (Diefendorff et al., 2011; Grandey, 2003). Meta-analytic results do not reveal a reliable relationship between deep acting and ex- haustion, when deep acting was measured only at the individual level of analysis (Hülsheger & Schewe, 2011; Mesmer-Magnus et al., 2012). However, our framework suggests that Grandey’s (2000) theory would be supported, if the team level of analysis was taken into account. When a difficult situation arises, a focal team member may deep act to change her or his feelings and produce an

emotional display that complies with expectations. Doing so requires a significant expenditure of emotional and cog- nitive resources (Grandey, 2000). If the deep acted emotion spreads through the team by emotional contagion, other team members will deep act with relatively little effort. Subsequently, team affective convergence should improve team rapport and coordination and relieve the strain on the focal team member (Barsade et al., 2009; Collins et al., 2013; Walter & Bruch, 2008). Evidence suggests that there are also direct benefits to individuals who are mim- icked by others such as increased liking and social support (van Baaren, Decety, Dijksterhuis, van der Leij, & van Leeuwen, 2011). As a result, the team should have lower dispersion, and increased deep acting should not increase emotional exhaustion in any of the team members. If, however, the deep acting of the focal team member does not spread through the team, this individual must con-

tinue to expend resources in order to maintain their emotional display in the face of inauthentic or even inappropriate displays by their teammates (Cole, Walter, & Bruch, 2008). Meanwhile, the difficult situation that required deep act- ing is unlikely to improve and remove the demand. Research also suggests that individuals respond negatively and experience stress when others do not mimic or imitate their authentic emotional displays (van Baaren et al., 2011). In this case, the individual deep acts within a team with high dispersion and must continue to expend resources to maintain their emotional display, while they do not receive support from teammates to replenish these resources. Therefore, they should experience increased emotional exhaustion (Grandey, 2000). As we have seen, prior research has primarily investigated the individual level of analysis. We argue here that

there is good reason to believe that a higher level of analysis may be more appropriate. According to our framework, the beneficial effects of deep acting will receive more support when team-level deep acting dispersion is included in the analysis. Therefore, we make the following prediction:

Hypothesis 2: Team-level deep acting dispersion will moderate the effect of individual-level deep acting, such that when dispersion is low, individual deep acting will be negatively related to emotional exhaustion.

Job satisfaction As we have argued, our framework suggests that low team-level deep acting dispersion should be associated with lower emotional exhaustion. Reframing this in light of Grandey’s (2000) model suggests that deep acting should be positively related to job satisfaction. A number of empirical studies have examined the relationship between individual-level deep acting and job satisfaction. However, meta-analyses have not found conclusive support for a positive relationship between deep acting and job satisfaction (Hülsheger & Schewe, 2011; Mesmer-Magnus et al., 2012; Wang, Seibert, & Boles, 2011). According to our meso-level framework, low levels of deep acting dis- persion result when teams converge on authentic, appropriate emotions despite difficult work situations. The group emotion literature suggests that affective convergence facilitates increased cooperation, reduced conflict, and im- proved working conditions (Collins et al., 2013). Therefore, individual deep acting in a team with low deep acting dispersion should be positively related to job attitudes, like job satisfaction. However, according to our framework, individual deep acting should influence job satisfaction primarily by its

indirect effect through emotional exhaustion. As shown in Figure 1, emotional exhaustion is expected to mediate

236 W. J. BECKER AND R. CROPANZANO

the relationship between deep acting and job satisfaction. This implies, of course, that exhaustion should be related to job satisfaction. The literature on the consequences of emotional exhaustion itself is extensive (for a review, see Halbesleben & Buckley, 2004). The great majority of the studies that investigated job satisfaction (see Singh, Goolsby, & Rhoads, 1994, for one exception) have consistently found that emotional exhaustion was deleterious for job satisfac- tion (e.g., Burke & Greenglass, 1995; Jackson, Turner, & Brief, 1987). This occurs because individuals, who are deplet- ing their emotional resources and not receiving the support necessary from their organization, supervisor, or teammates to restore them, will be unhappy with their working situation. Previously, we predicted that the effect of individual deep acting on emotional exhaustion would depend on the level of deep acting dispersion in the team. Given the framework we have articulated, we are predicting a conditional indirect effect for the relationship between deep acting and job sat- isfaction. That is, the path from deep acting (the predictor variable) to emotional exhaustion (the mediator variable) is moderated by deep acting dispersion (a third variable). This is a case of mediated moderation.

Hypothesis 3: Emotional exhaustion will mediate the relationship between deep acting and job satisfaction. This indirect effect will be moderated by deep acting dispersion such that the indirect relationship between deep acting and job satisfaction through exhaustion will be positive and stronger on teams with low deep acting dispersion.

Job performance In general, deep acting should be positively related to job performance, especially in a service context, because of the authentic nature of displayed emotions (Grandey, 2000). Evidence indicates that people can reliably discern other’s authentic emotional displays and that they react more favorably to appropriate and authentic displays (Ashforth & Tomiuk, 2000). A number of empirical investigations of emotional labor have included deep acting and job performance and found support for this prediction (Grandey, 2003; Groth, Hennig-Thurau, & Walsh, 2009). However, the meta-analytic results are mixed. One analysis found that deep acting was positively related to work, task, and emotion performance (Mesmer-Magnus et al., 2012). Another meta-analysis concluded that deep acting was positively related to customer satisfaction but not to task or emotional performance (Hülsheger & Schewe, 2011). Once again, we suggest that a meso-level approach may be more appropriate for detecting the rela- tionship between deep acting and performance. From this perspective, individual deep acting, while part of a team with greater convergence on deep acting, should have a positive relationship with job performance. It is also important to consider the indirect effect of deep acting through emotional exhaustion. Evidence suggests that

emotional exhaustion has a deleterious effect on job performance (e.g., Halbesleben & Bowler, 2007; Keijsers, Schaufeli, LeBlanc, Zwerts, & Miranda, 1995; Wright & Bonett, 1997; Wright & Cropanzano, 1998). Once again, in- dividuals who deplete their emotional resources should be less able to meet the taxing emotional demands of their job. Overall, there has been a consistent (negative) link between emotional exhaustion and performance (Cropanzano, Rupp, & Byrne, 2003). In this case, there is reason to expect both a direct and an indirect effect between deep acting and job performance. Again, we expect these effects to be conditional on the level of deep acting dispersion in the team. From the preceding discussion and our proposed model of the emotional labor process, we predict the following:

Hypothesis 4a: There will be a positive relationship between individual-level deep acting and job performance.

Hypothesis 4b: The relationship between individual-level deep acting and job performance will be moderated by team-level deep acting dispersion such that the relationship between deep acting and performance will be stronger in teams with low dispersion.

Hypothesis 4c: There will be an indirect effect of deep acting on performance through emotional exhaustion that will be moderated by deep acting dispersion such that the deep acting will be positively related to performance on teams with low deep acting dispersion.

DEEP ACTING IN WORK TEAMS 237

METHOD

Sample and procedure

Testing the theoretical model presented in Figure 1 imposed some fairly demanding requirements for an appropriate sample. First, we needed individuals who were organized into identifiable work teams. Second, these individuals needed to be engaged in work that was emotionally challenging with clear display rules. To meet these demands, this study utilized a sample of nurse teams in a large veteran’s hospital in the southwest USA. An important aspect of nursing is the interdependence within nursing teams (Cox, 2003). Likewise, the core daily activities involved in nursing are both emotionally demanding and complex (Bolton, 2000). Therefore, the present sample of professional nurses, who were organized into work teams, provided a strong setting for testing our predictions. The hospital employed a total of 415 nurses engaged directly in patient care. Over the course of approximately

2months, we were able to meet with 319 (77 percent) of these nurses in small groups during shift turnover and in- vited them to participate in the study. Nurse surveys with unique codes were distributed at these meetings in person- alized envelopes. The nurses then discarded the envelopes and returned the completed surveys via drop boxes located in each ward. Sixty-six percent of the distributed surveys were returned. Once all the surveys were collected, a manager (16 in total) completed an evaluation of each nurse that returned a survey. Each manager was responsible for between one and five nurse teams depending on the size and structure of the unit (the manager with five teams supervised five different clinics). Each manager was provided with a survey for each nurse that they were responsi- ble for. The unique code was then used to link the nurse and manager surveys. Three nurses were dropped from the sample because of excessive missing data. The final sample consisted of 208

nurses. Eighty percent of the nurses who responded were registered nurses with the highest level of nursing educa- tion and qualifications. The average age was 44.16 years (SD = 11.50). The respondents had an average of 13.89years (SD= 11.56) nursing experience and an average tenure at the hospital of 5.67 years (SD = 5.81). Eighty-five percent of the nurses were female. The nurses were organized into multiple functional departments such as emergency room, intensive care, and outpatient clinic. Within each department, nurses generally remained with the same coworkers according to a shift schedule. Therefore, work teams were defined by department and shift. The sample included 35 such work teams. The average tenure of nurses in their work teams was 3.19years (SD= 2.40). The average team size was 5.71 (SD = 3.83).

Measures

Each measure is presented according to the order it appears in the process model shown in Figure 1. All of the mea- sures were completed by the nurses except for the job performance measure, which was completed by each nurse’s direct supervisor.

Independent variables Deep Acting was measured by four items from Grandey et al. (2004). Nurses were asked to think of a time in the previous 2 weeks when they experienced conflicting emotions when dealing with a patient or family mem- ber. Sample items included the following: “I generally tried to look at the positive side of things to change how I feel” and “I tried to see things from the other person’s point of view.” Responses were on a five-point frequency scale ranging from 1 (“almost never”) to 5 (“very frequently”). Team-level Deep Acting Dispersion was represented by the standard deviation of individual team member deep acting at the construct level within each work team (Chan, 1998).

238 W. J. BECKER AND R. CROPANZANO

Outcome variables All responses were on a five-point Likert scale, ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). Emotional Exhaustion was measured using five items from the Oldenburg Burnout Inventory adapted from Halbesleben and Demerouti (2005). A sample item is as follows: “During my work, I often feel emotionally drained.” We used a three-item global Job Satisfaction scale (Seashore, Lawler, Mirvis, & Camman, 1982): “In gen- eral, I like working here,” “In general, I don’t like my job” (reverse coded), and “All in all, I am satisfied with my job.” Supervisors evaluated each nurse’s in-role Job Performance with five items adapted from Williams and Anderson (1991). Sample items included the following: “Meets formal performance requirements of the job” and “neglects aspects of the job he/she is obligated to perform.” Job performance was then centered for each supervisor to account for potential rating differences by individual supervisors.

Control variables We also considered a number of variables that could impact the interpretation of our findings. First, we controlled for surface acting because it represented the other general form of emotional labor that could be used to comply with display rules (see Grandey, 2000). We measured surface acting with two items from Grandey et al. (2004). The same referent and response scale as deep acting was used. The items were “I faked my feelings” and “I tried to keep my feelings from interfering too much.” We also controlled for team tenure, because we ex- pected that greater tenure provided more time for the meso-level effects to take hold. Finally, we controlled for age because emotional experience and expression have been shown to vary over the adult life span (Gross, Carstensen, Tsai, Skorpen, & Hsu, 1997).

RESULTS

The descriptive statistics for the individual-level variables are shown in Table 1. All of the scales were adapted from previous studies, and internal reliabilities suggested that the minor modifications were appropriate for the hospital setting. Even though we focused on deep acting, we did not ignore surface acting, even though it was less concep- tually relevant. When testing our hypotheses, we controlled for surface acting in all of our statistical tests. The in- clusion of surface acting did not appreciably change our findings, except to make them a bit more conservative. Consistent with our theoretical position, surface acting had modest associations with study variables but did not show team-level effects, like those we predicted for deep acting.

Table 1. Descriptive statistics and intercorrelations for individual-level variables.

Student M SD 1 2 3 4 5 6 7

1. Age 44.19 11.71 2. Team tenure 3.25 4.35 .25** 3. Deep acting 3.90 0.70 .23** �.06 (.82) 4. Surface acting 3.23 0.99 .06 �.09 .36** (.70) 5. Emotional exhaustion 3.13 0.76 �.18* .05 .00 .24** (.74) 6. Job satisfaction 4.06 0.72 �.01 �.08 �.05 �.26** �.32** (.82) 7. Performance 4.28 0.64 .12 �.08 .08 �.05 �.12 .15* (.84) Note: N = 208. Coefficient alpha is provided along the diagonal. **p < .01, *p < .05.

DEEP ACTING IN WORK TEAMS 239

Hypothesis 1: deep acting dispersion

As shown in Figure 1, we conceptualized meso-level deep acting as a dispersion model because our theory sug- gested that team convergence on deep acting would occur in some teams but not others and be a meaningful team characteristic (Chan, 1998). We operationalized deep acting dispersion using the standard deviation of individual team member deep acting construct scores within each team and followed the guidance of Cole and colleagues (2011) to verify the appropriateness of our assumptions. Aggregation statistics for deep acting were as follows: ICC(1) =.12, ICC(2) =.40, and average rwg =.86 (range from .13 to .98). The ICC(1) and rwg values suggest that there was modest agreement within groups. These results are consistent with a dispersion model, because all teams are not expected to converge in their deep acting. In fact, some teams demonstrated quite low rwg values. We found a strong correlation between the mean level and dispersion of team deep acting (r= �.75, p < .01). This indicated that when teams converged, they tended to converge on higher mean levels of deep acting that was consistent with our meso-level, emotional contagion account. Because the level and dispersion of team deep acting were not indepen- dent, we controlled for the mean level of team deep acting in all of the analyses. Overall, the results substantiated our theoretical rationale for a dispersion model of team-level deep acting, supporting Hypothesis 1.

Hypotheses 2–4: tests of the deep acting dispersion model

As discussed earlier, we proposed that individual deep acting would be negatively related to emotional exhaustion when team members converged on deep acting. Emotional exhaustion, in turn, would mediate the relationship be- tween deep acting and our two criterion variables—job satisfaction and job performance. These hypotheses posed multilevel relationships between team-level deep acting dispersion and individual-level outcomes. Therefore, hier- archical linear modeling (HLM) was an appropriate analytical approach for investigating these hypotheses (Bryk & Raudenbush, 1996; Hoffman, Morgeson, & Gerras, 2003). Individual deep acting was group mean centered at level one. Team deep acting dispersion and mean level were grand mean centered at level two. This combination of centering prevented the conflation of between-group and within-group variance (Enders & Tofighi, 2007). It is also important to note that our proposed model posed indirect effects of deep acting on job satisfaction and

performance via emotional exhaustion. Because of concerns with traditional mediation tests for multilevel analysis, there has been a renewed emphasis on tests of the indirect paths between variables (MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002). The most basic of these tests remains the Sobel (1982) test. However, more so- phisticated tests of indirect effects such as the simultaneous multilevel regression and the Monte Carlo Method for Assessing Mediation are growing in popularity and relative ease (Bauer, Preacher, & Gil, 2006; Selig & Preacher, 2008). Therefore, we report the results of multiple mediation tests.

Hypotheses 2: emotional exhaustion Hypothesis 2 predicted that deep acting dispersion would moderate the effect of individual deep acting, such that when dispersion was low, individual deep acting would be negatively associated with emotional exhaustion. We first created an individual-level only model of emotional exhaustion Model 1 (Table 2) and found that individual-level deep acting was not significantly related to exhaustion. Next, in Model 2 (Table 2), the team-level deep acting dis- persion was included to create a cross-level interaction. Mean level of deep acting was included as a control as rec- ommended by Cole and colleagues (2011). Consistent with our prediction, team-level deep acting dispersion had a significant moderating effect on the relationship between individual deep acting and emotional exhaustion (γ =0.72, p< .01). Figure 2 illustrates the nature of this cross-level interaction and indicates that when team dispersion was low, in-

dividual deep acting had a strong negative relationship with exhaustion. It also suggests that when dispersion was high, the slope of the relationship was near zero. A simple slopes test indicated that the slope was not significantly

240 W. J. BECKER AND R. CROPANZANO

different from zero (p > .05) for dispersion values above the mean level of dispersion. Therefore, Hypothesis 2 was supported.

Hypothesis 3: job satisfaction Hypothesis 3 predicted that the effect of the deep acting at different levels of deep acting dispersion would be me- diated by emotional exhaustion. Table 2 presents the results of the HLM models that tested these predictions. Model 3 shows that the direct moderated effect of deep acting on job satisfaction was not significant. Moreover, Model 4 found that emotional exhaustion had a strong negative relationship with job satisfaction (γ = �0.39, p < .01). The Sobel test indicated that the indirect effect of the moderated deep acting relationship was significant (z = �2.22, p< .05). Because the mediation took place at level one, we also performed significance tests of the indirect path

Table 2. Hierarchical linear modeling results for hypothesis tests.

Variable

Emotional exhaustion Job satisfaction In-role performance

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

Intercept (γ00) 3.12** 3.10** 4.08** 4.07** �0.02 �0.02 �0.03 Level 1 variables Age (γ10) �0.01 �0.01 0.00 0.00 �0.00 �0.00 �0.00 Team tenure (γ20) 0.02 0.02 �0.01 0.00 �0.02 0.02 0.02 Deep acting (γ30) �0.03 �0.20† 0.07 �0.02 0.08 0.17* 0.18* Surface acting (γ40) 0.25** 0.25** �0.25** �0.15** �0.05 �0.05 �0.06† Emotional exhaustion (γ50) �0.39** 0.01

Level 2 variables Deep acting level (γ01) �0.11 0.03 �0.01 �0.03 �0.04 Deep acting dispersion (γ02) 0.20 �0.33 �0.43 �0.03 �0.01 Dispersion interaction (γ32) 0.72** �0.26 0.02 �0.40* �0.43**

σ2 .45 .44 .39 .32 .19 .19 Pseudo-R2 .22 .23 .27 .39 .27 .28

Note: NL1 = 208. NL2 = 35. Dispersion interaction is the cross-level interaction between individual deep acting and team-level deep acting disper- sion. Pseudo-R

2 calculated based on the proportional difference in level-1 variance.

**p < .01, *p < .05, and †p < .10.

Figure 2. Moderated relationship between deep acting and emotional exhaustion

DEEP ACTING IN WORK TEAMS 241

using the simultaneous multilevel regression method recommended by Bauer, Preacher, and Gil (2006). We tested the indirect effect at different values of deep acting dispersion and found a significant indirect effect of deep acting on job satisfaction via emotional exhaustion at low to average levels of dispersion (γ = �0.34, p < .05). In addition, the bootstrapped 95 percent CI for the indirect effect at low levels of dispersion (�.57, �.04) did not contain zero. Taken together, these tests indicated that the moderated effect of deep acting on job satisfaction was due largely to its indirect effect through emotional exhaustion. Therefore, Hypothesis 3 was supported.

Hypothesis 4: job performance Hypothesis 4a predicted that there would be a positive relationship between individual deep acting and job perfor- mance. Model 5, a level 1 only model, did not find a significant relationship between individual deep acting and per- formance. Hypothesis 4b predicted that the effect of individual deep acting on job performance would be moderated by team-level deep acting dispersion such that the relationship between deep acting and performance would be pos- itive and stronger in teams with low dispersion. Model 6 shows that there was a significant cross-level interaction between deep acting and team dispersion (γ = �0.40, p < .05). Finally, Hypothesis 4c predicted that the moderated relationship between deep acting and performance would have an indirect effect on performance through emotional exhaustion. Model 7 did not find a significant relationship between emotional exhaustion and performance in the full model. Nonetheless, the cross-level interaction between deep acting and team dispersion continued to be significant (γ= �0.43, p < .01). Thus, our moderated model was successful at predicting job performance ratings (Hypothesis 4b). However, the main effect of exhaustion and its predicted indirect effect were not significant. Figure 3 shows this cross-level interaction at high and low values of the team dispersion. When team dispersion

was low, individual deep acting had a stronger positive relationship with performance than when dispersion was high. A simple slopes test indicated that the slope of the relationship between deep acting and performance was pos- itive and significantly different from zero (p> .05) when team dispersion was .80 or lower. In summary, the results indicated that Hypotheses 4a and 4c were not supported, but Hypotheses 4b was supported.

DISCUSSION

This study continues a recent trend of investigating multilevel research into the importance of emotions and emotion skills in teams (Ashkanasy, 2003; Troth, Jordan, Lawrence, & Tse, 2012). More specifically, we contribute to this

Figure 3. Moderated relationship between deep acting and job performance

242 W. J. BECKER AND R. CROPANZANO

growing literature by developing and testing a meso-level model of deep acting in work teams. Our framework draws on emotional contagion theory to suggest how the emotions and emotional displays of one team member can influence and be influenced by other team members. The results provided general support for our model, and we found evidence that some teams converged on deep acting as a shared method of responding to challenging emotional requirements. The results also suggested that lower team dispersion had beneficial relationships with emo- tional exhaustion, job satisfaction, and job performance. In addition, we found that deep acting by individuals who were members of teams with lower team deep acting

dispersion was negatively related to emotional exhaustion. In teams with high dispersion, the beneficial effects of individual deep acting were much reduced or non-existent. These findings are consistent with earlier theoretical work by Grandey (2000) and extend it to the team level. The current model and findings also suggest that emotional contagion can facilitate greater deep acting without depleting emotional resources. Even though it is extremely dif- ficult to measure emotional contagion in a field setting, these findings are consistent with our meso-level account. They are also consistent with the relationship between deep acting and emotional contagion proposed by others in related topics such as climate (Barsade & O’Neill, 2014) and leadership (Ashkanasy & Humphrey, 2011; Dasborough et al., 2009). Our framework also predicted that emotional exhaustion would have consequences for job satisfaction

(Hypothesis 3) and job performance (Hypothesis 4). These predictions were largely although not entirely supported. Our results found that emotional exhaustion was negatively related with job satisfaction. We did not find a direct relationship between deep acing and job satisfaction, but we did find a significant indirect effect of the moderated deep acting relationship on job satisfaction through emotional exhaustion. Individuals who deep acted as part of a team with low deep acting dispersion reported lower emotional exhaustion and higher job satisfaction. Therefore, the benefits of deep acting as part of a team with low dispersion of deep acting extended to job satisfaction. Team convergence on deep acting was also positively related to in-role job performance. Our support for

Hypothesis 4b suggested that deep acting as part of a team with low dispersion was associated with differences in supervisor-rated performance. On teams with low dispersion, there was a strong positive relationship between indi- vidual deep acting and in-role performance ratings. However, individual deep acting was not significantly related to supervisor-rated performance on teams with high dispersion. Contrary to our expectations, we did not find support for Hypothesis 4a or 4c. No direct relationship was found

between individual-level deep acting and performance. We also did not find a relationship between emotional ex- haustion and job performance. This suggests that exhaustion may have had a stronger effect on job attitudes, with less impact (if any) on supervisor perceptions of performance. Figures 2 and 3 suggest a complex interplay between deep acting, emotional exhaustion, and supervisor-rated performance. This was particularly true on teams with high dispersion, where the linkages between deep acting, exhaustion, and performance tended to break down. It is also true that display rules in nursing are more ambiguous than other service contexts. In some cases, teams may endorse emotional displays that are not perfected aligned with their supervisor’s desires. In this regard, a number of previous studies with nursing samples also failed to detect a significant relationship between deep acting or exhaustion and supervisor-rated performance (i.e., Brotheridge & Lee, 2002; Drach-Zahavy, Somech, Granot, & Spitzer, 2004; Judge et al., 2009; Randall & Scott, 1988).

Theoretical and practical implications

The results of this study provided a number of important theoretical implications and directions for future research. This study was one of the first to conceptualize and test a meso-level framework of emotional labor in work teams. We subsequently found evidence that team-level deep acting effects can manifest in mature teams. Our meso-level paradigm outlined how emotional contagion provided a potential mechanism linking individual deep acting to team- level convergence on deep acting. The level of convergence in the team then reinforced (or weakened) the benefits of individual deep acting and was related, directly or indirectly, to decreased emotional exhaustion and increased job

DEEP ACTING IN WORK TEAMS 243

satisfaction and performance. Our framework and findings provided some insight into the inconsistent effects of deep acting in the individual-level studies to date (Hülsheger & Schewe, 2011). These initial findings will need to be replicated in different types of work teams and in different settings in order to determine the boundary conditions under which team-level deep acting convergence is likely to occur. The teams in this study were unique in their ma- turity and level of interdependence among team members. These conditions were favorable for emotional contagion (Barsade, 2002). These findings also have practical implications. Deep acting in teams with low dispersion was associated with

higher performance and higher employee well-being. This is particularly important in the health care environment because nurse burnout is a persistent problem that impacts patient safety and satisfaction as well as nurse well-being (e.g., Halbesleben, Wakefield, Wakefield, & Cooper, 2008; Vahey, Aiken, Sloane, Clarke, & Vargas, 2004). Our findings suggest that managers could reduce exhaustion by actively seeking to encourage deep acting by team members. Previous studies have demonstrated that training can help individuals be more aware of their emotions and employ different emotional labor strategies (Ashkanasy, Dasborough, & Ascough, 2009; Totterdell & Parkinson, 1999). Therefore, seminars and training that discuss the personal and professional benefits of deep acting and provide strategies for genuinely changing one’s feelings during difficult situations provide promising methods to increase individual deep acting. To consider the matter more generally, our findings are consistent with research on the emotional expressions of

leaders (Humphrey, 2002; Humphrey et al., 2008). This research is extensive, and a complete summary is beyond our present purposes. However, it is worth mentioning that leaders’ emotional expressions can improve the mood of subordinates (Johnson, 2008; Sy, Côté, & Saavedra, 2005). In general, doing so appears to boost subsequent team performance (Johnson, 2009; Volmer, 2012). Our findings are fully consistent with these ideas, although we suggest that leaders should also attend to work team convergence on positive deep acting, because failure to converge can undermine the benefits that have been suggested by earlier work. The current study also had several potential limitations. While this sample was appropriate for testing our theo-

retical model, it would have been useful to have a greater number of work teams. The power of this study to distin- guish significant group level relationships was somewhat low (Maas & Hox, 2005). Of course, the fact that most of our hypotheses were supported suggests that the effects found and reported were particularly strong. In addition, this study investigated a single organization and a single profession. Strong professional and organizational norms are likely limited between team differences in deep acting (George, 1990). Nonetheless, this limitation should also have reduced the ability to detect team-level dispersion effects, rendering our present conclusions more conservative. Future research with a larger number of teams from multiple organizations, and different professions would be beneficial. In addition, this study was unable to directly measure and investigate the nature of the emotional contagion pro-

cesses that we suggest led some teams to converge on deep acting. Of particular interest for practical implementation would be to determine the relative importance of team member characteristics for emotional contagion and conver- gence. For example, it was not clear whether it was more important to have more influential focal members (informal leaders, charismatic members, etc.) who initiate deep acting or to have team members who were more susceptible to emotional contagion. It is possible that both conditions were necessary to produce a beneficial level of convergence. These are important questions for future research to explore. Another limitation involved the cross-sectional nature of the study, which created ambiguity regarding the causal

order of study variables and concern with common method bias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). First, we attempted to minimize susceptibility to this by choosing established measures, structuring the survey into distinct sections with separate instructions and stems, and providing verbal and written assurances of confidentiality. We also obtained a supervisor-reported measure of performance. Post-hoc tests for an overarching common method factor did not indicate that there was a significant common variance factor between the measure items. Lastly, we predicted and found evidence for cross-level interactions and indirect effects, both of which should not occur if the observed relationships between study variables resulted primarily from common method variance (Rupp & Spencer, 2006; Siemsen, Roth, & Oliveira, 2010). A future longitudinal study would help further resolve these

244 W. J. BECKER AND R. CROPANZANO

shortcoming and significantly advance our understanding of the relationship between deep acting and job outcomes. Limitations on the length of the survey also prevented us from collecting several other variables that may have been of interest such as frequency of interaction and social ties between nurses.

CONCLUSION

The purpose of this study was to explore an expanded conceptualization of meso-level deep acting in a field setting for an occupation with high emotional demands and interdependent teams. The findings of this study demonstrated that team-level deep acting dispersion can have important consequences for work outcomes, such as emotional ex- haustion (Hypothesis 2), job satisfaction (Hypothesis 3), and job performance (Hypothesis 4b). The results sug- gested that researchers should consider a richer conceptualization that considers both individual and team-level deep acting. From a practical perspective, these findings suggest that managers could benefit from efforts to help work teams converge on increased deep acting.

Author biographies

William Becker is an assistant professor at Texas Christian University. His research interests include organizational neuroscience, emotion, and turnover. His work has appeared in Journal of Applied Psychology, Personnel Psychology, Journal of Management, and Journal of Organizational Behavior. He is currently co-editing an Organizational Neuro- science column in The Industrial–Organizational Psychologist. Russell Cropanzano is a professor of management and entrepreneurship at the Leeds School of Business. His primary research explores workplace justice and behavioral ethics. He has edited four books and published more than 120 scholarly articles and chapters. He has also authored two books. The first, Organizational Justice and Human Resources Management, won the 1998 Book Award from the International Association of Conflict Man- agement. The second, Social Justice and the Experience of Emotion, was released in November of 2011. He was also a winner of the 2007 Best Paper Award from Academy of Management Perspectives and the 2010 Best Paper Award from the Journal of Management. He is a past editor of the Journal of Management and a fellow in the Society for Industrial/Organizational Psychology, the Southern Management Association, and the Associ- ation for Psychological Science.

REFERENCES Ashforth, B. E., & Tomiuk, M. A. (2000). Emotional labour and authenticity: Views from service agents. In S. Fineman (Ed.), Emotion in organizations (pp. 184–203). London: Sage Publications.

Ashkanasy, N. M. (2003). Emotions in organizations: A multi-level perspective. In F. Dansereau, & F. J. Yammarino (Eds.), Research in multi-level issues, vol. 2: Multi-level issues in organizational behavior and strategy (pp. 9–54). Oxford, UK: Elsevier Science.

Ashkanasy, N. M., & Ashton-James, C. E. (2005). Emotion in organizations: A neglected topic in I/O psychology, but with a bright future. International Review of Industrial and Organizational Psychology, 20, 221–268.

Ashkanasy, N. M., Dasborough, M. T., & Ascough, K. W. (2009). Developing leaders: Teaching about emotional intelligence and training in emotional skills. The Sage handbook of management learning, education and development (pp. 161–183). London: Sage.

DEEP ACTING IN WORK TEAMS 245

Ashkanasy, N. M., & Humphrey, R. H. (2011). A multi-level view of leadership and emotions: Leading with emotional labor. In A. Bryman, D. Collinson, K. Grint, B. Jackson, & M. Uhl-Bien (Eds.), Sage handbook of leadership (pp. 363–377). London: Sage.

Ashkanasy, N. M., & Nicholson, G. J. (2003). Climate of fear in organisational settings: Construct definition, measurement and a test of theory. Australian Journal of Psychology, 55, 24–29.

Barsade, S., Brief, A. P., & Spataro, S. E. (2003). The affective revolution in organizational behavior: The emergence of a par- adigm. In J. Greenberg (Ed.), Organizational behavior: A management challenge (pp. 3–50). Mahwah, NJ: Lawrence Erlbaum.

Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on group behavior. Administrative Science Quarterly, 47, 644–675.

Barsade, S. G., & Gibson, D. E. (1998). Group emotion: A view from top and bottom. In D. Gruenfeld, E. Mannix, & M. Neale (Eds.), Research on managing groups and teams (pp. 81–102). Stamford, CT: JAI Press.

Barsade, S. G., & Gibson, D. E. (2007). Why does affect matter in organizations? Academy of Management Perspectives, 21, 36–59. Barsade, S. G., & O’Neill, O. A. (2014). What’s love got to do with it? A longitudinal study of the culture of companionate love and employee and client outcomes in the long-term care setting. Administrative Science Quarterly. doi:10.1177/ 0001839214538636

Barsade, S. G., Ramarajan, L., & Westen, D. (2009). Implicit affect in organizations. Research in Organizational Behavior, 29, 135–162.

Bartel, C. A., & Saavedra, R. (2000). The collective construction of work group moods. Administrative Science Quarterly, 45, 197–231.

Bauer, D. J., Preacher, K. J., & Gil, K. M. (2006). Conceptualizing and testing random indirect effects and moderated mediation in multilevel models: New procedures and recommendations. Psychological Methods, 11, 142–163.

Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K. J. Klein, & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations (pp. 349–381). San Francisco: Jossey-Bass.

Bolton, S. C. (2000). Who cares? Offering emotion work as a ‘gift’ in the nursing labour process. Journal of Advanced Nursing, 32, 580–586.

Bono, J. E., & Ilies, R. (2006). Charisma, positive emotions and mood contagion. Leadership Quarterly, 17, 317–334. Brotheridge, C. M., & Grandey, A. A. (2002). Emotional labor and burnout: Comparing two perspectives of “people work”. Journal of Vocational Behavior, 60, 17–39.

Brotheridge, C. M., & Lee, R. T. (2002). Testing a conservation of resources model of the dynamics of emotional labor. Journal of Occupational Health Psychology, 7, 57–67.

Bryk, A. S., & Raudenbush, S. W. (1996). HLM: Hierarchical linear and nonlinear modeling with the HLM/2L and HLM/3L programs. Chicago, IL: SSI, Inc.

Burke, R. J., & Greenglass, E. (1995). A longitudinal study of psychological burnout in teachers. Human Relations, 48, 187–202. Chan, D. (1998). Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models. Journal of Applied Psychology, 83, 234–246.

Chartrand, T. L., & Bargh, J. A. (1999). The chameleon effect: The perception–behavior link and social interaction. Journal of Personality and Social Psychology, 76, 893–910.

Cole, M. S., Bedeian, A. G., Hirschfeld, R. R., & Vogel, B. (2011). Dispersion–composition models in multilevel research: A data-analytic framework. Organizational Research Methods, 14, 718–734.

Cole, M. S., Walter, F., & Bruch, H. (2008). Affective mechanisms linking dysfunctional behavior to performance in work teams: A moderated mediation study. Journal of Applied Psychology, 93, 945–958.

Collins, A. L., Lawrence, S. A., Troth, A. C., & Jordan, P. J. (2013). Group affective tone: A review and future research direc- tions. Journal of Organizational Behavior, 34, 43–62.

Côté, S., & Morgan, L. M. (2002). A longitudinal analysis of the association between emotion regulation, job satisfaction, and intentions to quit. Journal of Organizational Behavior, 23, 947–962.

Cox, K. B. (2003). The effects of intrapersonal, intragroup, and intergroup on team performance effectiveness and work satisfac- tion. Nursing Administration Quarterly, 27, 153–163.

Cropanzano, R., Rupp, D. E., & Byrne, Z. S. (2003). The relationship of emotional exhaustion to work attitudes, job performance, and organizational citizenship behaviors. Journal of Applied Psychology, 88, 160–169.

Cropanzano, R., Weiss, H. M., & Elias, S. M. (2004). The impact of display rules and emotional labor on psychological well-being at work. In P. L. Perrewe & D. C. Ganster (Eds.), Emotional and physiological processes and positive intervention strategies (pp. 45–89). Amsterdam: JAI.

Dasborough, M. T., Ashkanasy, N. M., Tee, E. Y., & Tse, H. H. (2009). What goes around comes around: How meso-level negative emotional contagion can ultimately determine organizational attitudes toward leaders. Leadership Quarterly, 20, 571–585.

246 W. J. BECKER AND R. CROPANZANO

Diefendorff, J. M., Erickson R. J., Grandey, A. A., & Dahling J. J. (2011). Emotional display rules as work unit norms: A multilevel analysis of emotional labor among nurses. Journal of Occupational Health Psychology, 16, 170–186.

Diefendorff, J. M., & Gosserand, R. H. (2003). Understanding the emotional labor process: A control theory perspective. Journal of Organizational Behavior, 24, 945–959.

Diefendorff, J. M., & Greguras, G. J. (2009). Contextualizing emotion display rules: Examining the roles of targets and discrete emotions in shaping display rule perceptions. Journal of Management, 35, 880–898.

Diefendorff, J. M., Richard, E. M., & Croyle, M. H. (2006). Are emotional display rules formal job requirements? Examination of employee and supervisor perceptions. Journal of Occupational and Organizational Psychology, 79, 273–298.

Drach-Zahavy, A., Somech, A., Granot, M., & Spitzer, A. (2004). Can we win them all? Benefits and costs of structured and flex- ible innovation–implementations. Journal of Organizational Behavior, 25, 217–234.

Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel models: A new look at an old is- sue. Psychological Methods, 12, 121–138.

George, J. M. (1990). Personality, affect, and behavior in groups. Journal of Applied Psychology, 75, 107–116. George, J. M. (2002). Affect regulation in groups and teams. In R. G. Lord, R. J. Klimoski, & R. Kanfer (Eds.), Emotions in the workplace: Understanding the structure and role of emotions in organizational behavior (pp. 183–217). San Francisco: Jossey-Bass.

Gibson, D. E. (1997). The struggle for reason: The sociology of emotions in organizations. In R. J. Erickson, & B. Cuthbertson- Johnson (Eds.), Social perspectives on emotion (Vol. 4, pp. 211–256). Greenwich, CT: JAI Press.

Gosserand, R. H., & Diefendorff, J. M. (2005). Emotional display rules and emotional labor: The moderating role of commitment. Journal of Applied Psychology, 90, 1256–1264.

Grandey, A. (2000). Emotion regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 5, 95–110.

Grandey, A. (2003). When “the show must go on”: Surface and deep acting as predictors of emotional exhaustion and service delivery. Academy of Management Journal, 46, 86–96.

Grandey, A., Dickter, D., & Sin, H. P. (2004). The customer is not always right: Customer verbal aggression toward service em- ployees. Journal of Organizational Behavior, 25, 397–418.

Gross, J. J. (2002). Emotion regulation: Affective, cognitive, and social consequences. Psychophysiology, 39, 281–291. Gross, J. J., Carstensen, L. L., Tsai, J., Skorpen, C. G., & Hsu, A. Y. C. (1997). Emotion and aging: Experience, expression, and control. Psychology and Aging, 12, 590–599.

Groth, M., Hennig-Thurau, T., & Walsh, G. (2009). Customer reactions to emotional labor: The roles of employee acting strat- egies and customer detection accuracy. Academy of Management Journal, 52, 958–974.

Halbesleben, J. R. B., & Bowler, W. M. (2007). Emotional exhaustion and job performance: The mediating role of motivation. Journal of Applied Psychology, 92, 93–106.

Halbesleben, J. R. B., & Buckley, M. R. (2004). Burnout in organizational life. Journal of Management, 30, 859–879. Halbesleben, J. R. B., & Demerouti, E. (2005). The construct validity of an alternative measure of burnout: Investigating the English translation of the Oldenburg Burnout Inventory. Work & Stress, 19, 208–220.

Halbesleben, J. R. B., Wakefield, B. J., Wakefield, D. S., & Cooper, L. B. (2008). Nurse burnout and patient safety outcomes— Nurse safety perception versus reporting behavior. Western Journal of Nursing Research, 30, 560–577.

Hatfield, E., Cacioppo, J. T., & Rapson, R. L. (1994). Emotional contagion. New York: Cambridge University Press. Hatfield, E., Rapson, R. L., & Le, Y.-C. L. (2011). Emotional contagion and empathy. In J. Decety, & W. Ickes (Eds.), The social neuroscience of empathy (pp. 19–30). Cambridge, MA: MIT Press.

Hennig-Thurau, T., Groth, M., Paul, M., & Gremler, D. D. (2006). Are all smiles created equal? How emotional contagion and emotional labor affect service relationships. Journal of Marketing, 70, 58–73.

Hochschild, A. R. (1979). Emotion work, feeling rules, and social structure. American Journal of Sociology, 85, 551–575. Hoffman, D. A., Morgeson, F. P., & Gerras, S. J. (2003). Climate as a moderator of the relationship between leader–member ex- change and content specific citizenship: Safety climate as an exemplar. Journal of Applied Psychology, 88, 170–178.

House, R., Rousseau, D. M., & Thomas-Hunt, M. (1995). The meso paradigm: A framework for the integration of micro and macro organizational behavior. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (Vol. 17, pp. 71–114). Greenwich, CT: JAI Press.

Hülsheger, U. R., & Schewe, A. F. (2011). On the costs and benefits of emotional labor: A meta-analysis of three decades of re- search. Journal of Occupational Health Psychology, 16, 361–389.

Humphrey, R. H. (2002). The many faces of emotional leadership. Leadership Quarterly, 13, 493–504. Humphrey, R. H., Pollack, J. M., & Hawver, T. (2008). Leading with emotional labor. Journal of Managerial Psychology, 23, 151. Ilies, R., Wagner, D. T., & Morgeson, F. P. (2007). Explaining affective linkages in teams: individual differences in susceptibility to contagion and individualism–collectivism. Journal of Applied Psychology, 92, 1140–1150.

Jackson, S. E., Turner, J. A., & Brief, A. P. (1987). Correlates of burnout among public-service lawyers. Journal of Occupational Behaviour, 8, 339–349.

DEEP ACTING IN WORK TEAMS 247

Johnson, S. K. (2008). I second that emotion: Effects of emotional contagion and affect at work on leader and follower outcomes. The Leadership Quarterly, 19, 1–19.

Johnson, S. K. (2009). Do you feel what I feel? Mood contagion and leadership outcomes. The Leadership Quarterly, 20, 814–827.

Judge, T. A., Woolf, E. F., & Hurst, C. (2009). Is emotional labor more difficult for some than for others? A multilevel, experience-sampling study. Personnel Psychology, 62, 57–88.

Kaplan, H. B., Burch, N. R., & Bloom, S. W. (1964). Physiological covariation and sociometric relationships in small peer groups. In P. H. Leiderman, & D. Shapiro (Eds.), Psychobiological approaches to social behavior (pp. 92–109). Stanford, CA: Stanford University Press.

Keijsers, G. J., Schaufeli, W. B., LeBlanc, P., Zwerts, C., & Miranda, D. R. (1995). Performance and burnout in intensive care units. Work and Stress, 9, 513–527.

Kelly, J. R., & Barsade, S. G. (2001). Mood and emotions in small groups and work teams. Organizational Behavior and Human Decision Processes, 86, 99–130.

Maas, C. J. M., & Hox, J. J. (2005). Sufficient sample sizes for multilevel modeling. Methodology: European Journal of Research Methods for the Behavioral and Social Sciences, 1, 86–92.

MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A comparison of methods to test me- diation and other intervening variable effects. Psychological Methods, 7, 83–104.

Marks, M. A., Mathieu, J. E., & Zaccaro, S. J. (2001). A temporally based framework and taxonomy of team processes. Academy of Management Review, 26, 356–376.

Martin, J., Knopoff, K., & Beckman, C. (1998). An alternative to bureaucratic impersonality and emotional labor: Bounded emo- tionality at The Body Shop. Administrative Science Quarterly, 43, 429–469.

Mathieu, J., Maynard, M. T., Rapp, T., & Gilson, L. (2008). Team effectiveness 1997–2007: A review of recent advancements and a glimpse into the future. Journal of Management, 34, 410–476.

Mesmer-Magnus, J., DeChurch, L. A., & Wax, A. (2012). Moving emotional labor beyond surface and deep acting: A discordance–congruence perspective. Organizational Psychology Review, 2, 6–53.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879–903.

Rafaeli, A. (1989). When cashiers meet customers: An analysis of the role of supermarket cashiers. Academy of Management Journal, 32, 245–273.

Rafaeli, A., & Sutton, R. I. (1990). Busy stores and demanding customers: How do they affect the display of positive emotion? Academy of Management Journal, 33, 623–637.

Randall, M., & Scott, W. A. (1988). Burnout, job satisfaction, and job performance. Australian Psychologist, 23, 335–347. Rimé, B. (2009). Emotion elicits the social sharing of emotion: Theory and empirical review. Emotion Review, 1, 60–85. Rupp, D. E., & Spencer, S. (2006). When customers lash out: The effects of customer interactional injustice on emotional labor and the mediating role of discrete emotions. Journal of Applied Psychology, 91, 971–978.

Scott, B. A., & Barnes, C. M. (2011). A multilevel field investigation of emotional labor, affect, work withdrawal, and gender. Academy of Management Journal, 54, 116–136.

Seashore, S. E., Lawler, E. E., Mirvis, R., & Camman, C. E. (1982). Observing and measuring organizational change: A guide to field practice. New York: John Wiley & Sons.

Selig, J. P., & Preacher, K. J. (2008). Monte Carlo method for assessing mediation: An interactive tool for creating confidence intervals for indirect effects. Available from http://quantpsy.org/.

Shields, S. A. (2005). The politics of emotion in everyday life: “Appropriate” emotion and claims on identity. Review of General Psychology, 9, 3–15.

Siemsen, E., Roth, A., & Oliveira, P. (2010). Common method bias in regression models with linear, quadratic, and interaction effects. Organizational Research Methods, 13, 456–476.

Singh, J., Goolsby, J. R., & Rhoads, G. K. (1994). Behavioral and psychological consequences of boundary spanning burnout for customer service representatives. Journal of Marketing Research, 31, 558–569.

Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations models. In S. Leinhart (Ed.), Sociological methodology (pp. 290–312). San Francisco: Jossey–Bass.

Sy, T., Côté, S., & Saavedra, R. (2005). The contagious leader: Impact of the leader’s mood on the mood of group members, group affective tone, and group processes. Journal of Applied Psychology, 90, 295–305.

Totterdell, P. (2000). Catching moods and hitting runs: Mood linkage and subjective performance in professional sports teams. Journal of Applied Psychology, 85, 848–859.

Totterdell, P., Kellet, S., Teuchmann, K., & Briner, R. B. (1998). Evidence of mood linkages in work groups. Journal of Personality and Social Psychology, 74, 1504–1515.

Totterdell, P., & Parkinson, B. (1999). Use and effectiveness of self-regulation strategies for improving mood in a group of trainee teachers. Journal of Occupational Health Psychology, 4, 219–232.

248 W. J. BECKER AND R. CROPANZANO

http://quantpsy.org/

Troth, A. C., Jordan, P. J., Lawrence, S. A., & Tse, H. H. (2012). A multilevel model of emotional skills, communication performance, and task performance in teams. Journal of Organizational Behavior, 33, 700–722.

Tsai, W. C. (2001). Determinants and consequences of employee displayed positive emotions. Journal of Management, 27, 497–512.

Tsai, W. C., & Huang, Y. M. (2002). Mechanisms linking employee affective delivery and customer behavioral intentions. Journal of Applied Psychology, 87, 1001–1008.

Vahey, D. C., Aiken, L. H., Sloane, D. M., Clarke, S. P., & Vargas, D. (2004). Nurse burnout and patient satisfaction. Medical Care, 42, 57–66.

van Baaren, R. B., Decety, J., Dijksterhuis, A., van der Leij, A., & van Leeuwen, M. L. (2011). Being imitated: Consequences of nonconsciously showing empathy. Cambridge, MA: MIT Press.

Van Maanen, J., & Kunda, G. (1989). Real feelings: Emotional expression and organizational culture. Research in Organizational Behavior, 11, 43–103.

Volmer, J. (2012). Catching leaders’ mood: Contagion effects in teams. Administrative Sciences, 2, 203–220. Walter, F., & Bruch, H. (2008). The positive group affect spiral: A dynamic model of the emergence of positive affective simi- larity in work groups. Journal of Organizational Behavior, 29, 239–261.

Wang, G., Seibert, S. E., & Boles, T. L. (2011). Synthesizing what we have known and looking ahead: A meta-analytic review of 30 years of emotional labor research. In C. E. J. Hartel, N. M. Ashkanasy, & W. J. Zerbe (Eds.), Research on emotion in or- ganizations: What have we learned? Ten years on (Vol. 7). Bingley, UK: Emerald.

Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and organizational commitment as predictors of organizational citizen- ship and in-role behaviors. Journal of Management, 17, 601–617.

Wright, T. A., & Bonett, D. G. (1997). The contribution of burnout to task performance. Journal of Organizational Behavior, 18, 491–499.

Wright, T. A., & Cropanzano, R. (1998). Emotional exhaustion as a predictor of job performance and voluntary turnover. Journal of Applied Psychology, 83, 486–493.

DEEP ACTING IN WORK TEAMS 249

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