Journal of Vocational Behavior 79 (2011) 538–548

Contents lists available at ScienceDirect

Journal of Vocational Behavior

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j v b

Relationships between emotional labor, job performance, and turnover☆

Robyn E. Goodwin ⁎, Markus Groth 1, Stephen J. Frenkel 2
Australian School of Business, University of New South Wales, Sydney, Australia

a r t i c l e i n f o a b s t r a c t

Article history: The present study investigates the relationship between the emotional labor strategies surface
Received 3 December 2010 acting and deep acting and organizational outcomes, specifically, employees' overall job
Available online 4 March 2011 performance and turnover. Call center employees from two large financial service organizations
completed an online survey about their use of surface and deep acting. Their responses were
Keywords: matched with supervisors' ratings of overall job performance and organizational turnover records
Emotional labor obtained 9 months later. Results indicate that surface acting is directly related to employee turnover
Customer service and emotional exhaustion and that the relationship between surface acting and job performance is
Employee performance indirect via employee affective delivery. Deep acting was not linked to these outcomes. Theoretical
Turnover
and practical implications are discussed from the perspective of emotional labor theories.
© 2011 Elsevier Inc. All rights reserved.

Emotional labor refers to employees' use of various strategies to regulate their emotions when interacting with customers in order
to meet organizational emotional display requirements (Grandey, 2000; Hochschild, 1983), also known as display rules (Ekman,
1973). Emotional labor is performed by employees in response to their perceptions of display rules (Diefendorff & Richard, 2003) and
their commitment to these standards (Gosserand & Diefendorff, 2005). Understanding the consequences of emotional labor is
important because both theory and empirical evidence suggest that emotional labor is integral to the daily work experience of many
frontline service employees and is closely linked to indicators of employee well-being (Grandey, 2000; Hochschild, 1983), customer
outcomes such as satisfaction and loyalty (Giardini & Frese, 2008; Grandey, 2000; Grandey, Fisk, Mattila, Jansen, & Sideman, 2005;
Hennig-Thurau, Groth, Paul, & Gremler, 2006), and ultimately, organizational performance (Grandey, 2000).
Emotional labor can be viewed as an application of Gross's (1998b) process model of emotion regulation to the workplace
(Côté, 2005). Thus, what is understood as emotional labor is a set of regulatory cognitions and behaviors enacted by employees at
work in response to actual or anticipated discrepancies between felt emotions and perceptions of expected emotional displays.
According to Gross's (1998b) model, emotion regulation is either antecedent-focused (anticipatory) or response-focused
(reactionary). When applied to employee–customer interactions, these types of emotion regulation strategies mirror what is
commonly known as deep acting (i.e., modifying felt emotions, usually in anticipation of a perceived discrepancy between felt and
required emotions) and surface acting (i.e., the expression of emotions not actually felt by suppressing felt emotions, amplifying
the expression of a weakly felt emotion, or faking unfelt emotions) (Hochschild, 1983).
Empirical evidence shows that surface and deep acting often yield divergent outcomes for employees, particularly regarding
their well-being (Grandey, 2003; Judge, Woolf, & Hurst, 2009). Surface acting typically has more detrimental outcomes than deep
acting. However, the impact of emotional labor strategies on organizational outcomes remains under-researched, hence the
significance of the present study. Specifically, we investigate how the emotional labor strategies of surface and deep acting
correspond to two critical outcomes: supervisor-rated overall job performance and employee turnover. In so doing, we move

☆ This research was funded by a grant of the Australian Research Council (LP0776781). The authors would like to thank Nick Wang and Jochen Reb for their
constructive feedback on earlier versions of this article. An earlier version of the paper was presented at the 24th annual meeting of the Society for Industrial and
Organizational Psychology in New Orleans, Louisiana, April 2009.
⁎ Corresponding author. Fax: +61 2 9313 6775.
E-mail addresses: [email protected] (R.E. Goodwin), [email protected] (M. Groth), [email protected] (S.J. Frenkel).
1
Fax: +61 2 9313 6775.
2
Fax: +61 2 9662 8531.

0001-8791/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jvb.2011.03.001
 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548 539

beyond prior research that either used broad measures of emotional labor (Duke, Goodman, Treadway, & Breland, 2009) or used
self-report (Totterdell & Holman, 2003) or laboratory-based measures of performance (Goldberg & Grandey, 2007) to examine the
link between specific emotional labor strategies and core job performance in an applied setting. The mediating roles of emotional
exhaustion and affective delivery are also explored.

1. Grandey's (2000) model of emotional labor

Building on Hochschild's (1983) work, Grandey (2000) developed a comprehensive conceptual framework of emotional labor
that draws on emotion regulation theory (Gross, 1999). The model suggests that employees' emotional labor strategies will result
in several well-being outcomes for individuals (increased job satisfaction and reduced burnout and stress) and organizations
(improved employee performance and reduced turnover).
Although there is considerable empirical evidence illustrating the effects of emotional labor on employee well-being outcomes, the
relationship between emotional labor and organizational outcomes has received far less empirical attention, despite strong theoretical
support for such a link (Ashforth & Humphrey, 1993; Grandey, 2000; Hochschild, 1983; Totterdell & Holman, 2003). Indeed, a strong
assumption underlying organizations' requirement that employees display positive emotions is that such sincere emotional displays
(or good ‘affective delivery’) are believed to result in favorable organizational outcomes including increased employee and customer
retention, higher employee performance, and ultimately, improved organizational performance (Grandey, 2000; Hochschild, 1983).
Given this theoretical justification, the present research contributes to the emotional labor literature by examining the links
between the emotional labor strategies of surface and deep acting and two critical organizational outcomes: employees' overall
job performance and turnover behavior. In addition, we go beyond Grandey's (2000) predictions and examine affective delivery
(i.e., the authenticity with which an employee expresses required emotions) and emotional exhaustion as two key mediators of
these relationships. Furthermore, this study makes an important methodological contribution by using actual rather than self-
reported measures of turnover and performance. With few exceptions (Chau, Dahling, Levy, & Diefendorff, 2009), research linking
emotional labor to turnover, employee well-being (e.g., stress and burnout) and to performance have relied on employee self-
reports (e.g., Abraham, 1999; Brotheridge & Grandey, 2002; Côté & Morgan, 2002; Pugliesi, 1999; Totterdell & Holman, 2003) or
proxies such as the number of errors committed in a laboratory as a substitute for job performance (Sideman & Grandey, 2007).
We utilize three independent sources of data—employee self-reports, supervisor ratings of job performance, and organizational
turnover records—in order to examine the relationship between emotional labor and organizational outcomes. This approach
reduces the potential effects of common method bias on the results, and improves on previous research by concentrating on two
rather than a single dependent variable. The overall conceptual model guiding the present research is illustrated in Fig. 1.

2. Emotional labor and employee performance

Employee performance—the actions and behaviors that are controlled by the individual and contribute to the goals of the
organization (Rotundo & Sackett, 2002)—is an important construct since it often informs reward and discipline decisions. In a
service work context employee performance refers to both tangible service delivery and intangible aspects such as interpersonal
behavior and emotional display (Bitner, Booms, & Tetreault, 1990; Bowen & Schneider, 1988). Emotional labor strategies are
particularly important in the context of customer service, because employee behavior during service delivery (e.g., empathy,
responsiveness) is often perceived by customers as the most critical aspect of service quality (Bitner et al., 1990). In particular,
employees' positive emotional displays are typically central to the customer service experience (Bailey, Gremler, & McCollough,
2001) and are associated with favorable customer evaluations (Mattila & Enz, 2002; Tsai & Huang, 2002). Given this centrality, and

Fig. 1. Theoretical model and study hypotheses.

540 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548

the frequent intention of employees' emotional labor to evoke positive customer responses, we suggest that the extent to which
employees engage in emotional labor is positively related to their task-related performance.
Despite the above proposition, there are intrapersonal mechanisms that suggest a negative relationship between emotional
labor and performance. According to Ego depletion theory, self-regulatory efforts, including emotion regulation, draw on, and
deplete, a common resource (Baumeister, 2002). Thus emotion regulation can diminish other proximal self-regulatory efforts
(Tice & Bratslavsky, 2000), such as task performance. Given the demonstrated importance of adaptive self-regulation in
determining performance in the workplace (Tsui, Ashford, St Clair, & Xin, 1995), we expect that engaging in emotional labor limits
employees' ability to perform to their full potential and thus reduces their performance. However, different emotional labor
strategies may vary in the demands placed on employees (Gross, 1998a), thus requiring different levels of effort.
We therefore expect the relationships between different emotional labor strategies and job performance to vary. Specifically, we
expect that surface acting is negatively related to job performance, whereas deep acting is positively related. Surface acting can incur a
cognitive cost: suppression of emotions has been shown to impair memory and subsequent performance of cognitive tasks (Baumeister,
Bratslavsky, Muraven, & Tice, 1998; Richards & Gross, 1999) leading to diminished physical stamina (Muraven, Tice, & Baumeister,
1998). Surface acting, which involves such emotional suppression, can therefore be expected to be negatively related to job performance.
Despite scant empirical evidence regarding the relationship between deep acting and overall job performance, we expect deep acting
to be positively related to performance. Staw, Sutton, and Pelled (1994) found that employees who reported greater experiences of
positive emotions at Time 1 had higher supervisor ratings of performance at Time 2. Chronic deep acting increases employees' positive
emotional experiences, thus it seems reasonable to assume that deep acting plays a role in this relationship. Although deep acting
requires some initial self-regulation, from a cumulative perspective it does not necessitate the continuous—and cognitively demanding—
process of self-monitoring and suppression/masking associated with maintaining conflicting internal feelings and displays of emotion
that surface acting does (Beal, Trougakos, Weiss, & Green, 2006; Gross & Levenson, 1997). On an interpersonal level, displaying authentic
positive emotions in service delivery leads to more favorable perceptions of employees' service performance and related service delivery
outcomes (Grandey et al., 2005; Groth, Hennig-Thurau, & Walsh, 2009; Hennig-Thurau et al., 2006). Deep acting involves the expression
of real emotions, and thus we expect it to be positively related to performance.
Recently, Duke et al. (2009) examined the relationship between emotional display and performance, and found some support for a
significant relationship between emotional display norms and performance in cases where perceived organizational support of
employees was high. However, these authors ulitized a measure of ‘general emotional labor’ which assesses emotional display norms or
burnout rather than emotional labor per se. Nevertheless, these results provide some preliminary support for the claim that emotional
labor is related to performance, although no previous studies have actually examined emotional labor strategies as a predictor of overall
employee performance.
In summary, we expect the relationships between surface and deep acting and job performance to differ; the increased
cognitive load experienced by employees who surface act is expected to reduce job performance, whereas engaging in deep acting,
which does not deplete self-regulatory resources to the same extent, also produces authentic emotions that are more likely to lead
to service success. Therefore, we propose the following:

H1. Surface (deep) acting is negatively (positively) related to overall job performance.

In addition to a direct relationship between emotional labor and job performance, we also hypothesize that this relationship is
mediated by the service employee's affective delivery. Affective delivery refers to how well or authentically an employee appears to
express required emotions as judged by observers (Ashforth & Humphrey, 1993; Grandey, 2003). As mentioned earlier, many scholars
believe that authenticity is an important interpersonal aspect of emotional displays toward customers, and evidence indicates that this
is the case. Grandey et al. (2005) manipulated employee emotional authenticity in videos of hotel staff and participants, finding that
customers consistently indicated greater satisfaction with authentic displays than with inauthentic displays. Tsai and Huang (2002)
similarly found that affective delivery was related to positive service performance. In a study of cheerleaders, Beal et al. (2006)
demonstrated that as the degree of employee emotional labor increased, the negative relationship between employees' experience of
negative emotions with supervisor ratings of their affective delivery became less negative.
We expect affective delivery to be positively related to employee performance because it is a judgment about the ‘final product’
of emotional labor: the type and quality of emotions displayed to customers. Presumably, this relationship is substantial in the
context of service work, in light of the centrality of positive and authentic emotional expression to most customer service
interactions. Given that different emotional labor strategies have been shown to be more or less effective in producing consistent
and believable emotional displays to customers (Grandey et al., 2005; Hennig-Thurau et al., 2006), affective delivery is an
important mechanism to consider when examining the relationship between emotional labor and performance.
A mediating role of affective delivery is theoretically supported by Grandey (2000), who posits that, because deep acting involves
expressing real (albeit initially contrived) emotions, observers are more likely to perceive such displays as sincere and thus view them
more positively. On the other hand, surface actors find it difficult to maintain incongruent emotions and expressions and may ‘leak’
their true emotions to customers. In turn, customers may feel betrayed by the employee and react negatively to the service experience.
We also expect that, because emotional displays resulting from deep acting are genuine emotions, the resulting expressions are
more likely to be perceived by customers as sincere. In support, Grandey (2003) found that co-worker-rated affective delivery was
positively related to employee-reported deep acting, yet surface acting was negatively related to co-worker ratings of affective
delivery. In summary, there is ample theoretical support for the role of affective delivery as an important intermediary mechanism in
the relationship between emotional labor and job performance.
 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548 541

H2. The relationship between surface (deep) acting and job performance is mediated by affective delivery.

3. Emotional labor and employee turnover

Employee turnover is a critical concern for many organizations, given the immense costs associated with attracting and
replacing human capital (Dalton, Krackhardt, & Porter, 1981). In the call center industry in particular, turnover remains
persistently high and poses a major challenge for management (Robinson & Morley, 2006). For individual employees, turnover
may lead to personal and interpersonal problems after leaving the organization (Muchinsky & Morrow, 1980). Therefore,
employee turnover has long been a topic of interest to organizational scholars and practitioners alike.
Several researchers have posited that emotional labor potentially contributes to employee turnover (e.g., Côté & Morgan, 2002;
Grandey, 2000). Employees who regularly engage in emotional labor may develop the attitude that they are in an unsuitable work
environment, which may ultimately compel them to leave the organization (Grandey, 2000). This perspective emphasizes the
cumulative effect of performing emotional labor in predicting turnover and is consistent with findings that employees form
attitudes about work in their day-to-day tasks and interactions that, over time, develop into withdrawal cognitions and behaviors,
such as turnover (Holtom, Mitchell, Lee, & Eberly, 2008).
Although there is very limited empirical research that has directly tested the link between emotional labor and turnover, initial
evidence provides preliminary support for this relationship. Abraham (1999) has shown that customer service employees who
experience a discrepancy between their internal and external emotional experience at work are more likely to express intentions
to leave. Similarly, in a longitudinal study of college students with part-time jobs, Côté and Morgan (2002) found that emotion
regulation at work has effects on intentions to quit, rather than the other way around. However, when it comes to actual turnover
behavior rather than self-reported intentions to quit, the empirical evidence is sparse. Chau et al. (2009) conducted what appears
to be the only study on emotional labor in which actual turnover was measured. Based on a sample of bank tellers, their results
suggest that no direct relationship exists between emotional labor strategies and turnover. Rather, their impact on turnover was
found to be indirect, through turnover intentions and emotional exhaustion.
Similar to the relationship between emotional labor and job performance, we expect variation in relationships between surface
and deep acting and turnover. As discussed earlier, prior research has shown that surface acting is often related to negative
intrapersonal outcomes, such as emotional exhaustion and feelings of inauthenticity, whereas deep acting is more likely to lead to
positive outcomes, such as enhanced job satisfaction (e.g., Brotheridge & Grandey, 2002; Simpson & Stroh, 2004). Thus, we expect
surface acting to be positively related to turnover.
The effort involved in maintaining a positive display (and managing the associated emotional dissonance) is an uncomfortable
experience for many employees and may lead them to withdraw from such situations in the long term (Grandey, 2000). As
discussed by Chau et al. (2009), deep acting, which is more likely to evoke an authentic emotional display, leads to more positive
experiences for employees (since they are experiencing more positive emotions over time) and therefore reduces the likelihood of
withdrawal behaviors. This is consistent with research showing a negative relationship between turnover behavior and both
positive job attitudes and positive affective experiences (Hom & Kinicki, 2001).

H3. Surface (deep) acting is positively (negatively) related to employee turnover.

Based on evidence relating emotional labor to employee well-being outcomes noted earlier, we expect the link between
emotional labor strategies and employee turnover to be mediated by emotional exhaustion. The extent to which employees report
engaging in surface acting has consistently emerged as a negative predictor of employee well-being, particularly in regards to
employee burnout. Burnout, defined as a syndrome typical of employees who spend their emotional resources and do not recover
them (Jackson, Schwab, & Schuler, 1986), is often conceptualized as having three dimensions—emotional exhaustion,
depersonalization, and reduced personal accomplishment (Maslach, 1982)—all of which have been positively associated with
surface acting (Brotheridge & Grandey, 2002; Brotheridge & Lee, 2003; Grandey, 2003; Judge et al., 2009; Yagil, Luria, & Gal, 2008;
Zammuner & Galli, 2005). Emotional exhaustion is the “central quality of burnout and the most obvious manifestation of this
complex syndrome” (Maslach, Schaufeli, & Leiter, 2001, p. 402), and thus it is the dimension we focus on in this study.
We expect emotional exhaustion to be a key mechanism in the relationship between emotional labor and turnover. This is
consistent with the conservation of resources perspective, which posits that the burnout phenomenon equates to a loss in
resources available to employees when undertaking their jobs (Hobfoll & Freedy, 1993)—a depletion of various resources which
may result in employee withdrawal behaviors. Not surprisingly, a positive relationship between employee burnout and turnover
intentions is well established in the management literature (Lee & Ashforth, 1996).

H4. The relationship between surface (deep) acting and employee turnover is mediated by emotional exhaustion.

4. Method

4.1. Participants and procedure

Employees of two large Australian financial institutions' call centers were invited to participate in the study. These employees
primarily answer inbound calls from customers and are responsible for sales and service. Approximately 600 employees and their
542 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548

supervisors were initially invited to complete an online survey during work time. We received 377 completed surveys from
employee–supervisor dyads during the data collection period of approximately six weeks (63% response rate). The employee
sample comprised 37% male, with a mean age of 30.8 years (SD = 9.4) and a mean job tenure of 2.3 years (SD = 2.0). The
supervisor sample comprised 43% male, with a mean age of 35.2 years (SD = 9.0) and a mean job tenure of 2.0 years (SD = 2.1).
Respondents were entered into a lottery to win one of several retail store gift vouchers.
Initial visits to both call centers confirmed that supervisors had a close working relationship with their employees. In both
workplaces, supervisors had close physical proximity to their employees, regularly interacted with them, and were required as
part of their job to closely monitor employee performance by regularly listening to randomly selected calls on a daily basis.
Therefore, we concluded that the supervisors would be well aware of, and be in a position to accurately rate the interpersonal
components of employees' service performance.

4.2. Measures

The employee survey contained measures of emotional labor (surface acting and deep acting) and emotional exhaustion, along
with demographic and control variables. For each employee who completed the survey, measures of his or her affective delivery as
well as overall job performance were obtained from his or her respective supervisor. In addition, organization records were
accessed 9 months after survey completion to assess whether employees who had completed the survey were still employed by
the organization. All scale items used are shown in the Appendix A.
Emotional labor was measured using four items for surface acting (α = .90) and three items for deep acting (α = .86) developed
by Brotheridge and Lee (1998). Responses were on a frequency scale ranging from (1) = never to (5) = always.
Emotional exhaustion was measured using four items taken from Pines and Aronson (1988) on a five-point Likert scale ranging
from (1) = never to (5) = always (α = .93).
Ratings for both the affective delivery and overall job performance for each employee were independently obtained from his or
her immediate supervisor, during the data collection period. Affective delivery was measured using three items adapted from
McLellan, Schmit, Amundson, and Blake (1998). Responses were on a seven-point Likert-type scale ranging from (1) = strongly
disagree to (7) = strongly agree. The measure had an internal consistency estimate of α = .98. Overall job performance was
measured using two items from Williams and Anderson's (1991) measure of in-role performance as well as one item based on a
measure of service worker performance by Brown, Mowen, Donavan, and Licata (2002). Because the latter item used a different
response scale (from 1–10, with higher ratings indicating higher performance), we standardized all three items by converting
them to z-scores before combing them to form one scale. Thus, higher numbers indicate better overall job performance as rated by
the supervisor. The measure had an internal consistency estimate of α = .82.
Turnover data were obtained from organizational records 9 months after survey completion and were recorded as a binary
measure (1 = still employed; 2 = has left organization).

5. Results

5.1. Reliability and validity assessment

Means, standard deviations, correlation coefficients, and reliability estimates of all variables are shown in Table 1. The
reliability of all scales is satisfactory, with α scores ranging from .86 to .98. Descriptive results show that 24% of the sample had left
the organization in the 9 months subsequent to administering the survey.

Table 1
Means, standard deviations, and correlations. a

M SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

1. Gender b 1.63 0.48 –

2. Job tenure 2.29 1.96 .18** –
3. Customer service workload 65.90 27.25 .09 −.07 –
4. Organization c 1.36 0.48 .15** .17** −.22** –
5. Surface acting 2.50 1.06 .02 −.03 .07 .05 (.90)
6. Deep acting 2.69 1.02 .03 −.04 .07 −.08 .59** (.86)
7. Emotional exhaustion 2.93 0.87 .21** .11* .12* .08 .41** .30** (.93)
8. Affective delivery 6.05 1.40 .00 .10 .06 −.03 −.16** −.07 −.05 (.98)
9. Overall job performance d 0.00 0.86 −.03 −.16** .14** −.50** −.08 .00 −.06 .34** (.82)
10. Turnover e 1.24 0.43 −.09 −.15** .03 .11* .17** .06 .13* −.09 −.06 –

*p b .05 (two-tailed).
**p b .01 (two-tailed).
a
N = 377. Values along diagonal represent internal consistency estimates.
b
Coded as 1 = Male, 2 = Female.
c
Coded as 1 = Organization 1, 2 = Organization 2.
d
Items were standardized (z-scores) to form the scale.
e
Coded as 1 = Active employee, 2 = Left organization.
 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548 543

A measurement model of all multi-item measures was subjected to confirmatory factor analysis (CFA) in order to assess the
convergent and discriminant validity of all constructs. We compared the fit of three different models: a one-factor model; a four-
factor model in which both affective delivery and performance items loaded on a single factor; and the hypothesized five-factor
model (surface acting, deep acting, emotional exhaustion, affective delivery, overall job performance). The overall fit statistics for
our five-factor model indicate a good fit to the data: χ2 (109, N = 377) = 192.24, p b .01; Comparative fit index (CFI) = .99;
Incremental fit index (IFI) = .99; Root mean square error of approximation (RMSEA) = .05. The model fit was significantly
better than that for the one-factor model (Δχ2 [10] = 3983.55, p b .01; CFI = .29, IFI = .29; RMSEA = .30) and the four-factor model
(Δχ2 [6] = 408.36, p b .01; CFI = .92, IFI = .92; RMSEA = .11). The factor loadings of all items are shown in the Appendix A.
Further evidence of the discriminant validity of constructs in our measurement model was provided by investigating whether
the average variance extracted for two constructs exceed the square of the correlation between the constructs as suggested by
Fornell and Larcker (1981). All constructs showed sufficient discriminant validity.

5.2. Structural equation modeling and mediation analysis

Structural equation modeling was conducted to test all hypotheses in our theoretical model, shown in Fig. 1. We
operationalized employee turnover as a dichotomous variable (1 = active employee, 2 = left organization). Because turnover was a
single item measure, we made no adjustments for random measurement error and fixed its factor loading to one and its
measurement error to zero (see Markel & Frone, 1998). Overall, the fit statistics for our theoretical model indicate an excellent fit:
χ2 (130, N = 377) = 217.27, p b .01; CFI = .99; IFI = .99; RMSEA = .04. The structural model and standardized parameter estimates
shown in Fig. 2 indicate that neither surface acting nor deep acting were significantly related to supervisors' job performance
ratings. Thus, Hypothesis H1 was not supported. However, the total effect surface acting had on overall job performance was −.10,
suggesting that the relationship between surface acting and performance may be indirect, due to the mediating role of affective
delivery.
Hypothesis H2 suggests that the relationship between emotional labor and job performance is mediated by affective delivery.
Although tests for mediation in the management literature have been dominated by Baron and Kenny's (1986) ‘causal steps’
approach, recent research has advocated structural equation modeling as a superior method of testing mediation hypotheses on
both theoretical and empirical grounds (Iacobucci, 2008; Iacobucci, Saldanha, & Deng, 2007). We followed Iacobucci et al.'s (2007)
advice to fit one structural model (rather than fitting a series of models and employing model comparisons) that includes both
direct and indirect paths simultaneously. The significance of mediating effects and all direct and indirect effects were then tested
using a bootstrapping procedure (using 1000 re-samples) to determine the 95% bias-corrected confidence intervals around these
effects. A confidence interval that did not span zero indicated a statistically significant effect. The use of bias-corrected
bootstrapping confidence intervals has recently been advocated as a superior method and one that is preferred over the Sobel test
(Sobel, 1982) (see Cheung & Lau, 2009; MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002; MacKinnon, Lockwood, &
Williams, 2004 for a discussion).
Fig. 2 shows the direct path between the independent variable (i.e., surface acting) and the dependent variable (overall job
performance) was not significant. However, this path is not a significant prerequisite for a mediating effect to exist (see Kenny,
Kashy, & Bolger, 1998; Shrout & Bolger, 2002). Rather, if mediation paths both from the independent variable to the mediator
and from the mediator to the dependent variable are significant, some mediation may exist. In our case, both paths were
significant, as indicated by their bootstrapped bias-corrected confidence intervals: βsurface acting → affective delivery = −.16, p b .01;
βaffective delivery → overall job performance = .32, p b .01. The total standardized indirect effect was −.05, p b .01, as summarized in Table 2.
Thus, we conclude that affective delivery fully mediates the relationship between surface acting and overall job performance.
Therefore, Hypothesis H2 is supported in regard to surface acting.

Fig. 2. Estimated structural model. Note: Goodness-of-fit summary: χ2 (130, N = 377) = 217.27, pb.05; CFI = .98; IFI = .98; RMSEA = .04. All reported relationships
are statistically significant, pb.01 (two-tailed).
544 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548

Table 2
Results of bootstrapping analysis: Effects on overall performance.

Variable Standardized direct effect Standardized indirect effect 95% Confidence interval

Surface Acting not sig. −.05** −.10 to −.02

Affective Delivery .32** n/a .23 to 43

N = 377. *p b .05 (two-tailed). **p b .01 (two-tailed).

Consistent with Hypothesis H3, the results of our structural model show a significant, positive path for surface acting (β = .18,
p b .01). Thus, Hypothesis H3 is supported in regard to surface acting, but is not supported for deep acting.3 Hypothesis H4
proposed that emotional exhaustion mediates the relationship between both emotional labor strategies and turnover. Given that
the path between emotional exhaustion and turnover was not significant in our structural model, the mediation analysis was
stopped and we conclude that Hypothesis H4 is not supported.

6. Discussion

The aim of this study was to test part of Grandey's (2000) model of emotional labor by examining the relationship between
emotional labor strategies and organizational outcomes, and to test important mediators—affective delivery and emotional
exhaustion—in these relationships. Our results mostly support the model shown in Fig. 1, but this is not the case for deep acting.
We nevertheless make several important contributions to the literature.

6.1. Theoretical contribution

As anticipated, we found that service employees' affective delivery mediated the relationship between surface acting and job
performance. That is, surface acting decreases the likelihood of desirable emotional expression within customer service
interactions, and, in highly interactive customer service jobs such as that of call center agents, affective delivery seems to be an
important predictor of a service agent's overall job performance. As expected, the relationship between affective delivery and
performance was quite strong, consistent with theory suggesting the importance of affective delivery as a key determinant of
successful customer service (Grandey et al., 2005). Since the proportion of overall performance resulting from emotional
expression is likely to differ by occupation, this finding will vary across occupations, a proposition that merits further investigation.
Our study appears to be the first to demonstrate that surface acting exerts a significant indirect effect via affective delivery on
other-rated job performance. This finding highlights the potentially important outcomes of an employee's use of emotional labor
strategies on the job; if employees do not feel the emotions required of them in their interactions with customers, engaging in
surface acting may lead to inconsistent or insincere interpersonal displays of these desired emotions, resulting in reduced service
performance. This conclusion assumes that it is perceptions of inauthenticity on the part of the audience (i.e., customers and
supervisors) that have a detrimental effect on their evaluations. Although empirical evidence supports the possibility that others,
such as co-workers and customers, can detect authenticity and affective delivery during service interactions (Grandey, 2003;
Grandey et al., 2005; Hennig-Thurau et al., 2006), we did not directly assess authenticity in our study.
In line with theory (Grandey, 2000; Hochschild, 1983), but with limited empirical support, our results further demonstrate a
direct negative relationship between surface acting and employee turnover. This result contributes to the emotional labor
literature by showing that emotional labor is not only linked to employee well-being outcomes, but also to critical organizational
outcomes. In itself, this result has clear implications for both employees and employers alike; the more employees engage in
surface acting when interacting with customers, the more likely it is that they will leave their job or be terminated, at a cost to the
organization and employee.
Surprisingly, we did not find support for the mediating role of emotional exhaustion in this relationship. This result differs from
Chau et al. (2009), who found that emotional exhaustion was a mediator in the chain between surface acting and turnover.
However, when considered together, the results of these two studies suggest that a link does exist between emotional labor and
turnover, although the role of potential mediating variables needs to be investigated further.
Contrary to our hypotheses, deep acting did not exhibit a significant relationship with job performance or turnover behavior.
Nevertheless, our null findings echo prior research that has found the impact of surface acting to be more consistent than that of
deep acting (e.g., Brotheridge & Lee, 2003; Judge et al., 2009; Totterdell & Holman, 2003). It is important to note that deep acting
emerges as a more constructive and beneficial emotional labor strategy for employees to engage in than surface acting, if they need
to regulate their emotions at work, although it does not appear to be positively related to well-being; a null relationship is, in a
practical sense, undoubtedly far better than a detrimental one.
One potential explanation for the mixed findings related to deep acting is the possibility of unidentified moderating variables
(Duke et al., 2009). In light of our discussion about the importance of affective delivery, future research might examine the role of

3
We conducted a logistic regression analysis to determine if this result would change as a result of including control variables (gender, job tenure, customer
service workload, organization, and emotional exhaustion). Results were consistent with our structural equation modeling: after controlling for all variables
entered in step 1, the odds ratio (i.e., the odds of employees leaving their organization when the value of self-reported surface acting increases by 1) was 1.37
(Wald statistic: 4.11, p b .05).
 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548 545

social-cognitive motivational constructs, such as emotional labor strategy self-efficacy, in predicting the extent of emotional labor
that employees perform, how successful this is, and how much it may result in detrimental or beneficial outcomes.

6.2. Practical implications

The relationships demonstrated in this research—between surface acting and performance, and between surface acting and
turnover—have direct and important implications for service sector managers. Overall, they indicate that managers need to be
cognizant about the way their employees conform to organizational display rules. The type of emotional labor strategy that
employees engage in is likely to contribute to job performance and organizational attachment.
Overall performance is a metric that is widely used in organizations to reward and punish employees. This study has
demonstrated that performance may be influenced by how employees regulate their emotions. The relationship between
emotional labor strategies and different facets of performance (customer service outcomes, specific forms of task performance,
organizational citizenship behaviors, etc.) is an obvious next step for future theoretical and empirical investigation.
Our results suggest that customer service employees should be discouraged from surface acting. A qualification to this, though
one that is not explored by our research, is that effective surface acting may be a skill that can be learnt. One suggestion is that
training should enable employees to distinguish surface from deep acting and to role play these emotional labor strategies in
encounters that are typical of those they have with customers. Such training may assist employees to avoid engaging in surface
acting or inculcate superior surface acting skills that may conserve emotional resources over the longer term (see below).
Regular performance management processes, including communication, rewards, and recognition procedures, should
incorporate identification and positive reinforcement of emotional labor strategies that enhance customer satisfaction and
employee job performance. These might be summarized and re-enacted for training purposes. Such successful encounters are
likely to contribute to building strategies that enable employees to limit emotional exhaustion and provide customers with
satisfying experiences.
Because the cost of turnover for service organizations is often high (Korczynski, 2002), any means to prevent or discourage
employees from engaging in surface acting may benefit the organization even if, as in our research, the effect sizes of our results
are modest. Our research highlights the complexity that service organizations face in managing the emotional display of their
frontline employees; on the one hand, engineering friendly customer service processes may benefit the organization by increasing
customer satisfaction and sales, but on the other hand, the same processes may take their toll on service employees and result in
higher turnover, thus incurring significant organizational costs.

7. Limitations and recommendations for future research

Our research did not distinguish between the effectiveness of employees in engaging in surface and deep acting. While deep
acting appears to be more effective in the short term, it is possible that over the longer term this strategy leads to emotional
exhaustion and ill-health. Conversely, effective surface acting that is learnt and developed may, over time, result in more effective
organizational and personal outcomes. These possibilities invite longitudinal research under different conditions of emotional
labor strategy skill deployment.
The turnover measure used in this study did not allow us to distinguish between voluntary and involuntary turnover which
precluded further investigation of the relationship between emotional labor strategies and turnover. Future research may benefit
from making this distinction. A noteworthy point is that the turnover measure used in our study is likely to increase random error
in the data; therefore our results may in fact underestimate the actual strength of the relationship. Future research may also
benefit from examining a broader range of employee withdrawal behaviors such as absenteeism and various forms of employee
conflict in order to better understand the relationship between emotional labor and employee behavior, including the mechanisms
that give rise to different types of behavior and their implications for performance and employee well-being.
Research focusing on employee turnover at a more detailed level may contribute toward explaining our unexpected finding
that emotional exhaustion was unrelated to turnover. Perhaps it is not so much a question of whether this relationship exists, but
rather when it exists. Future longitudinal studies that investigate the effect of emotional exhaustion on withdrawal behavior (such
as turnover and absenteeism) might highlight the point at which such withdrawal is likely to occur.
One consideration when attempting to understand the null results for deep acting is that the measure we used – originally
developed by Brotheridge and Lee (1998) and the most commonly used – has limitations. Whereas the items for surface acting
simply capture a judgment about whether or not surface acting was enacted in a given situation, the deep acting items not only
capture whether participants engaged in particular behaviors, but also involve a judgment about the extent of effort that was
applied in the process, regardless of success. In light of this potential conflation, additional scale development and scale refinement
may benefit future studies on emotional labor.

8. Conclusion

In summary, the present study makes important theoretical and methodological contributions to the emotional labor literature
by showing evidence of both direct and indirect links between emotional labor strategies and the organizational outcomes of job
performance and turnover behavior, obtained from multiple sources. Thus, our research provides empirical support for parts of
Grandey's (2000) model of emotional labor that have received limited empirical attention, and also identifies affective delivery as
546 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548

an important mediator. Our study shows that employee emotional labor strategies do in fact impact important organizational
outcomes, a result with substantial implications for emotional labor researchers and managers alike.

Appendix A. List of survey items and results of confirmatory factor analysis

Standardized coefficient

Surface acting
“When doing your job, how often do you…?”
Put on an act in order to deal with customers in an appropriate way. .83
Fake a good mood when interacting with customers. .86
Put on a “show” or “performance” when interacting with customers. .85
Just pretend to have the emotions I need to display for my job. .81

Deep acting
“When doing your job, how often do you…?”
Try to actually experience the emotions I must show to customers. .78
Work hard to feel the emotions that I need to show to customers. .86
Make an effort to actually feel the emotions that I need to display .82
towards customers.

Emotional exhaustion
“In general, how often do you experience the following at your job?”
Being tired. .78
Being “wiped out”. .89
Feeling run-down. .92
Being exhausted. .92

Affective delivery
“To what extent do you agree or disagree with the following statements?”
This person seems sincere when dealing with customers. .97
Customers seem to like interacting with this person. .99
This person shows friendliness and warmth to most customers. .99

Overall job performance

This employee performs tasks that are expected of him/her. .90
This employee adequately completes all assigned duties. .85
How does the overall performance of this employee compare to all employees you .58
currently supervise or have supervised in the past?

N = 377. All factor loadings are significant at p b .01.

References

Abraham, R. (1999). The impact of emotional dissonance on organizational commitment and intention to turnover. The Journal of Psychology: Interdisciplinary and
Applied, 133, 441−445.
Ashforth, B. E., & Humphrey, R. H. (1993). Emotional labor in service roles: The influence of identity. Academy of Management Review, 18, 88−115.
Bailey, J. J., Gremler, D. D., & McCollough, M. A. (2001). Service encounter emotional value: The dyadic influence of customer and employee emotions. Services
Marketing Quarterly, 23, 1−26.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical
considerations. Journal of Personality and Social Psychology, 51, 1173−1182.
Baumeister, R. F. (2002). Ego depletion and self-control failure: An energy model of the self's executive function. Self and Identity, 1, 129−136.
Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology,
74(5), 1252−1265.
Beal, D. J., Trougakos, J. P., Weiss, H. M., & Green, S. G. (2006). Episiodic processes in Emotional Labor: Perceptions of affective delivery and regulation strategy.
Journal of Applied Psychology, 91(5), 1053−1065.
Bitner, M. J., Booms, B. H., & Tetreault, M. S. (1990). The service encounter: Diagnosing favorable and unfavorable incidents. Journal of Marketing, 54, 71−84.
Bowen, D. E., & Schneider, B. (1988). Services marketing and management: Implications for organizational behavior. Research in Organizational Behavior, 10,
43−80.
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. (1998). On the dimensionality of emotional labor: Development and validation of an emotional labor scale. The first Conference on
Emotions in Organizational Life, San Diego.
Brotheridge, C. M., & Lee, R. T. (2003). Development and validation of the Emotional Labour scale. Journal of Occupational and Organizational Psychology, 76,
365−379.
Brown, T. J., Mowen, J. C., Donavan, D. T., & Licata, J. W. (2002). The customer orientation of service workers: Personality trait effects on self- and supervisor
performance ratings. Journal of Marketing Research, 19, 110−119.
Chau, S. L., Dahling, J. J., Levy, P. E., & Diefendorff, J. M. (2009). A predictive study of emotional labor and turnover. Journal of Organizational Behavior, 30(8),
1151−1163.
Cheung, G. W., & Lau, R. S. (2009). Testing mediation and suppression effects of latent variables. Organizational Research Methods, 11, 296−325.
Côté, S. (2005). A social interaction model of the effects of emotion regulation on work strain. Academy of Management Review, 30(3), 509−530.
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.
 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548 547

Dalton, D. R., Krackhardt, D. M., & Porter, L. W. (1981). Functional turnover: An empirical assessment. Journal of Applied Psychology, 66, 716−721.
Diefendorff, J. M., & Richard, E. M. (2003). Antecedents and consequences of emotional display rule perceptions. Journal of Applied Psychology, 88(2), 284−294.
Duke, A. B., Goodman, J. M., Treadway, D. C., & Breland, J. W. (2009). Perceived organizational support as a moderator of emotional labor/outcomes relationships.
Journal of Applied Social Psychology, 39(5), 1013−1034.
Ekman, P. (1973). Cross culture studies of facial expression. In P. Ekman (Ed.), Darwin and Facial Expression: A Century of Research in Review (pp. 162−222). New
York: Academic.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18,
39−50.
Giardini, A., & Frese, M. (2008). Linking service employees' emotional competence to customer satisfaction: A multilevel approach. Journal of Organizational
Behavior, 29, 155−170.
Goldberg, L. S., & Grandey, A. A. (2007). Display rules versus display autonomy: Emotion regulation, emotional exhaustion, and task performance in a call center
simulation. Journal of Occupational Health Psychology, 12, 301−318.
Gosserand, R. H., & Diefendorff, J. M. (2005). Emotional display rules and Emotional Labor: The moderating role of commitment. Journal of Applied Psychology, 90(6),
1256−1264.
Grandey, A. A. (2000). Emotion regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 1, 95−110.
Grandey, A. A. (2003). When “the show must go on”: Surface acting and deep acting as determinants of emotional exhaustion and peer-rated service delivery.
Academy of Management Journal, 46(1), 86−96.
Grandey, A. A., Fisk, G. M., Mattila, A. S., Jansen, K. J., & Sideman, L. A. (2005). Is “service with a smile” enough? Authenticity of positive displays during service
encounters. Organizational Behavior and Human Decision Processes, 96, 38−55.
Gross, J. J. (1998a). Antecedent- and response-focused emotion regulation: Divergent consequences for experience, expression, and physiology. Journal of
Personality and Social Psychology, 74(1), 224−237.
Gross, J. J. (1998b). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2, 271−299.
Gross, J. J. (1999). Emotion and emotion regulation. In L. A. Pervin, & O. P. John (Eds.), Handbook of Personality: Theory and Research (pp. 525−552). New York:
Guilford.
Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: The acute effects of inhibiting negative and positive emotion. Journal of Abnormal Psychology, 106(1), 95−103.
Groth, M., Hennig-Thurau, T., & Walsh, G. (2009). Customer reactions to emotional labor: The roles of employee acting strategies and customer detection accuracy.
Academy of Management Journal, 52(5), 958−974.
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.
Hobfoll, S. E., & Freedy, J. (1993). Conservation of resources: A general stress theory applied to burnout. In W. B. Schaufeli, C. Maslach, & T. Marek (Eds.), Professional
Burnout: Recent Developments in Theory and Research (pp. 115−129). Washington, DC: Taylor & Francis.
Hochschild, A. (1983). The Managed Heart - Commercialization of Human Feeling. Berkeley, CA: University of California Press.
Holtom, B. C., Mitchell, T. R., Lee, T. W., & Eberly, M. B. (2008). Turnover and retention research: A glance at the past, a closer review of the present, and a venture
into the future. Academy of Management Annals, 21, 231−274.
Hom, P. W., & Kinicki, A. J. (2001). Toward a greater understanding of how dissatisfaction drives employee turnover. Academy of Management Journal, 44, 975−987.
Iacobucci, D. (2008). Mediation Analysis. Thousand Oaks: Sage.
Iacobucci, D., Saldanha, N., & Deng, X. (2007). A meditation on mediation: Evidence that structural equations models perform better than regressions. Journal of
Consumer Psychology, 17, 139−153.
Jackson, S. E., Schwab, R. L., & Schuler, R. S. (1986). Toward and understanding of the burnout phenomenon. Journal of Applied Psychology, 71, 630−640.
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.
Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social psychology. In D. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook of Social Psychology
(pp. 233−265). New York: McGraw-Hill.
Korczynski, M. (2002). Human Resource Management in Service Work. London: Palgrave MacMillan.
Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic examination of the correlates of the three dimensions of job burnout. Journal of Applied Psychology, 81(2),
123−133.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A comparison of methods to test the significance of mediation and other
intervening variable effects. Psychological Methods, 7, 83−104.
MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence limits for the indirect effect: Distribution of the product and resampling methods. Multivariate
Behavioral Research, 39, 99−128.
Markel, K. S., & Frone, M. R. (1998). Job characteristics, work–school conflict, and school outcomes among adolescents: Testing a structural model. Journal of Applied
Psychology, 83, 277−287.
Maslach, C. (1982). Burnout: The Cost of Caring. Englewood Cliffs, NJ: Prentice-Hall.
Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of Psychology, 52, 397−422.
Mattila, A. S., & Enz, C. A. (2002). The role of emotions in service encounters. Journal of Service Research, 4(4), 268−277.
McLellan, R. A., Schmit, M. J., Amundson, M., & Blake, R. (1998). Secret shopper ratings as an individual level criterion for validation studies. The 13th Annual
Conference of the Society for Industrial and Organizational Psychology, Dallas, TX Dallas, TX.
Muchinsky, P. M., & Morrow, P. C. (1980). A multidisciplinary model of voluntary employee turnover. Journal of Vocational Behavior, 17, 263−290.
Muraven, M., Tice, D. M., & Baumeister, R. F. (1998). Self-control as limited resource: Regulatory depletion patterns. Journal of Personality and Social Psychology, 74(3),
774−789.
Pines, A., & Aronson, E. (1988). Career Burnout: Causes and Cures. New York: Free Press.
Pugliesi, K. (1999). The consequences of emotional labor: Effects of work stress, job satisfaction, and well-being. Motivation and Emotion, 23(2), 125−154.
Richards, J. M., & Gross, J. J. (1999). Composure at any cost? The cognitive consequences of emotion suppression. Personality and Social Psychology Bulletin, 25,
1033−1044.
Robinson, G., & Morley, C. (2006). Call centre management: Responsibilities and performance. International Journal of Service Industry Management, 17, 284−300.
Rotundo, M., & Sackett, P. R. (2002). The relative importance of task, citizenship, and counterproductive performance to global ratings of job performance: A policy-
capturing approach. Journal of Applied Psychology, 87, 66−80.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7,
422−445.
Sideman, L. A., & Grandey, A. A. (2007). Display rules versus display autonomy: Emotion regulation, emotional exhaustion, and task performance in a call center
simulation. Journal of Occupational Health Psychology, 12(3), 301−318.
Simpson, P. A., & Stroh, L. K. (2004). Gender differences: Emotional expression and feelings of personal inauthenticity. Journal of Applied Psychology, 89, 715−721.
Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equation models. Sociological Methodology, 13, 290−312.
Staw, B. M., Sutton, R. S., & Pelled, L. H. (1994). Employee positive emotion and avorable outcomes at the workplace. Organization Science, 5, 51−70.
Tice, D. M., & Bratslavsky, E. (2000). Given in to feel good: The place of emotion regulation in the context of general self control. Psychological Inquiry, 11(3),
149−159.
Totterdell, P., & Holman, D. (2003). Emotion regulation in customer service roles: Testing a model of emotional labor. Journal of Occupational Health Psychology, 8(1),
55−73.
Tsai, W., & Huang, Y. (2002). Mechanisms linking employee affective delivery and customer behavioral intentions. Journal of Applied Psychology, 87(5),
1001−1008.
548 R.E. Goodwin et al. / Journal of Vocational Behavior 79 (2011) 538–548

Tsui, A. S., Ashford, S. J., St Clair, L., & Xin, K. R. (1995). Dealing with discrepant expectations: Response strategies and managerial effectiveness. Academy of
Management Journal, 38, 1515−1543.
Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of
Management, 17, 601−617.
Yagil, D., Luria, G., & Gal, I. (2008). Stressors and resources in customer service roles. International Journal of Service Industry Management, 19(5), 575−595.
Zammuner, V. L., & Galli, C. (2005). The relationship with patients: “Emotional Labor” and its correlates in hospital employees. In C. E. J. Hartel, W. J. Zerbe, & N. M.
Ashkanasy (Eds.), Emotions in Organizational Behavior (pp. 251−285). Mahwah, NJ: Lawrence Erlbaum Associates.