Content uploaded by Stefan Diestel
Author content
All content in this area was uploaded by Stefan Diestel on Feb 13, 2015
Content may be subject to copyright.
Journal of Applied Psychology
Sleep Quality and Self-Control Capacity as Protective
Resources in the Daily Emotional Labor Process: Results
From Two Diary Studies
Stefan Diestel, Wladislaw Rivkin, and Klaus-Helmut Schmidt
Online First Publication, December 8, 2014. http://dx.doi.org/10.1037/a0038373
CITATION
Diestel, S., Rivkin, W., & Schmidt, K.-H. (2014, December 8). Sleep Quality and Self-Control
Capacity as Protective Resources in the Daily Emotional Labor Process: Results From Two
Diary Studies. Journal of Applied Psychology. Advance online publication.
http://dx.doi.org/10.1037/a0038373
Sleep Quality and Self-Control Capacity as Protective Resources in the
Daily Emotional Labor Process: Results From Two Diary Studies
Stefan Diestel
International School of Management and Technical University
of Dortmund
Wladislaw Rivkin and Klaus-Helmut Schmidt
Technical University of Dortmund
Daily emotional labor can impair psychological well-being, especially when emotions have to be
displayed that are not truly felt. To explain these deleterious effects of emotional labor, scholars have
theorized that emotional labor can put high demands on self-control and diminishes limited regulatory
resources. On the basis of this notion, we examined 2 moderators of the daily emotional labor process,
namely day-specific sleep quality and individual self-control capacity. In particular, in 2 diary studies
(N
TOTAL
⫽171), we tested whether sleep quality moderates the influence of emotional dissonance (the
perceived discrepancy between felt and required emotions) on daily psychological well-being (ego
depletion, need for recovery, and work engagement). In addition, we examined 3-way interactions of
self-control capacity, sleep quality, and emotional dissonance on indicators of day-specific psychological
well-being (Study 2). Our results indicate that the negative relations of day-specific emotional dissonance
to all day-specific indicators of well-being are attenuated as a function of increasing day-specific sleep
quality and that self-control capacity moderates this interaction. Specifically, compared with low
self-control capacity, the day-specific interaction of emotional dissonance and sleep quality was more
pronounced when trait self-control was high. For those with low trait self-control, day-specific sleep
quality did not attenuate the negative relations of emotional dissonance to day-specific well-being.
Implications for research on emotional labor and for intervention programs are discussed.
Keywords: day-specific emotional dissonance, sleep quality, self-control capacity, day-specific psycho-
logical strain, day-specific, work engagement
Emotional labor, which is a form of emotion regulation in which
individuals create a bodily or facial display to meet organizational
requirements, has become an integral part of the job role in many
occupations (Hochschild, 1983). Whereas most empirical studies
on the effects of emotional labor on employees’ psychological
well-being draw on cross-sectional and longitudinal samples (Hül-
sheger & Schewe, 2011), scholars have recently begun to analyze
the emotional labor process on the day-level (Judge, Woolf, &
Hurst, 2009; Scott, Barnes, & Wagner, 2012). Scholarly interest in
the day-level effects of emotional labor is based on the self-control
strength model, according to which emotion regulation can involve
self-control and deplete limited regulatory resources (Muraven &
Baumeister, 2000). Self-control refers to volitionally inhibiting,
modifying, or overriding automatic and spontaneous response im-
pulses, emotions, and motivational processes. A large body of
experimental and field studies shows that self-control entails psy-
chological costs that immediately manifest as exhaustion, a state
referred to as ego depletion (Hagger, Wood, Stiff, & Chatzisaran-
tis, 2010).
To meet emotional job demands, employees are required to
exert self-control, especially when they have to display emotions
that they do not genuinely feel (Zapf & Holz, 2006). The perceived
discrepancy between emotions truly felt and those required by the
job role is commonly referred to as emotional dissonance (Morris
& Feldman, 1996). Emotional dissonance has been found to pre-
dict job dissatisfaction, burnout symptoms, and absenteeism (Hül-
sheger & Schewe, 2011). Consistent with the self-control perspec-
tive, several diary studies also demonstrate day-level negative
effects of emotional labor on employees’ well-being (Judge et al.,
2009). One of the key findings of these studies is that emotional
labor and associated levels of well-being exhibit substantial fluc-
tuations during the course of a week. That is, on some days
employees are strongly required to regulate their emotions because
of frequent discrepancies between felt and organizationally desired
emotions, and on other days emotional labor is less stressful or less
required because employees’ can express emotions that are more
in line with organizational requirements or employees have less
contact with customers, clients, or patients. Correspondingly, well-
being also shows high day-specific variations that are caused by
job demands, job conditions, and personality traits (Kühnel, Son-
nentag, & Bledow, 2012). For example, extraversion and emo-
tional stability moderate the negative day-specific relations of
emotional labor demands to well-being (Judge et al., 2009).
Stefan Diestel, Department of Psychology and Management, Interna-
tional School of Management and Technical University of Dortmund;
Wladislaw Rivkin and Klaus-Helmut Schmidt, Leibniz-Research Centre
for Working Environment and Human Factors, Technical University of
Dortmund.
Correspondence concerning this article should be addressed to Stefan
Diestel, Department of Psychology and Management, International School
of Management, Otto-Hahn-Str. 19, 44227 Dortmund, Germany. E-mail:
stefan.diestel@ism.de
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Journal of Applied Psychology © 2014 American Psychological Association
2014, Vol. 100, No. 1, 000 0021-9010/14/$12.00 http://dx.doi.org/10.1037/a0038373
1
However, although scholars have proposed that daily emotional
labor can put high demands on self-control (e.g., Scott et al.,
2012), a nuanced analysis of the relevant psychological mecha-
nisms that are delineated in the self-control strength model is
lacking in the literature. This gap in the literature is accentuated by
the lack of knowledge regarding how employees can cope with
daily emotional demands to prevent themselves from impaired
well-being even in cases of high emotional dissonance. Most
moderators examined in past research were job characteristics
(e.g., job control) or personality traits (e.g., extraversion), which
are difficult to be altered (Ilies, Dimotakis, & De Pater, 2010). To
identify moderators that are more likely to be influenced by
employees, our study focuses on day-specific sleep quality, which
should facilitate coping with emotional dissonance. In light of the
finding that day-specific impacts of demands on well-being are
contingent upon stable individual traits (Judge et al., 2009), we
also examine, whether, and to what extent, trait self-control mod-
erates the day-specific interaction of sleep quality and emotional
dissonance on well-being.
One of the conceptual challenges of providing insight into the
underlying nature of the day-specific emotional labor process is
that our understanding of this phenomenon cannot be solely de-
rived from one theory or model because several lines of research
on self-control have developed in recent years that have only rarely
been related to one another thus far. Thus, we integrate proposi-
tions of the self-control strength model (Muraven & Baumeister,
2000) and its theoretical extensions (conservation of resources and
role of motivation, Baumeister & Vohs, 2007), theories on sleep
quality and sleep disruption (Beebe & Gozal, 2002), models of
emotional labor (Hülsheger & Schewe, 2011), and state work
engagement (Breevaart, Bakker, Demerouti, & Hetland, 2012).
Our theoretical integration centers on the well-founded proposition
that processes of self-control (e.g., when emotions have to be
displayed that are not truly felt) tax a limited regulatory resource,
whose day-specific availability does not only determine self-
control functioning, but also affects psychological experience of
exhaustion, vitality, and motivational strength (Muraven &
Baumeister, 2000). The day-specific availability of resources for
self-control functioning can be characterized as a function of
day-specific processes (e.g., sleep quality, Beebe & Gozal, 2002)
and individual differences in the capacity to exert self-control
(Tangney, Baumeister, & Boone, 2004).
On the basis of our theoretical integration, we derive two pre-
dictions (see Figure 1): First, day-specific sleep quality moderates
the negative relations of emotional dissonance to well-being such
that the relationships are attenuated as a function of increasing
sleep quality. Second, self-control capacity moderates the day-
specific interaction of sleep quality and emotional dissonance on
well-being (three-way interaction). In particular, when trait self-
control is high, day-specific sleep quality attenuates the negative
effects of emotional dissonance on well-being. In contrast, in cases
of low trait self-control, sleep quality should not interact with
emotional dissonance. Given that psychological well-being in-
volves both strain and motivational states, we consider need for
recovery (van Veldhoven & Broersen, 2003) and ego depletion
(Muraven & Baumeister, 2000) as strain outcomes, as well as work
engagement (Bakker, 2011) as an indicator of a fulfilling and
motivational state of mind.
In the following, we conceptualize emotional dissonance as a
demand on self-control, and develop the prediction of day-specific
interactions of emotional dissonance and sleep quality on well-
being. We test these predictions in a diary study. In addition, we
elaborate on theoretical extensions of the self-control strength
model to clarify how self-control capacity moderates the interac-
tion of emotional dissonance and sleep quality on well-being.
Finally, we analyze the hypothesized three-way interaction in a
second diary study.
Emotional Dissonance and Self-Control: Depletion of
Limited Regulatory Resources
Compared with other emotional labor demands (e.g., specific
display rules or sensitivity requirements), emotional dissonance
exerts the strongest negative effects on well-being (Zapf & Holz,
2006). In addition, several studies have also demonstrated that
emotional dissonance mediates the relations of emotional display
rules to strain and thus explains why emotional labor can impair
well-being (Cheung & Tang, 2010). To provide a conceptual
framework for emotional dissonance, Zapf and Holz (2006) argued
that the discrepancy between felt and required emotions puts high
demands on self-control (see also Diestel & Schmidt, 2011a,
2011b). Their argument is grounded in the self-control strength
model, according to which self-control processes draw on and
deplete limited regulatory resources (Muraven & Baumeister,
2000). In support of this argument, several studies have revealed
that both suppressing felt emotions and exaggerating a required
emotional display cause ego depletion and, hence, temporarily
impair self-control processes such as working memory operations
and response inhibition (Schmeichel, Vohs, & Baumeister, 2003;
Schmeichel, 2007). In addition, experimentally induced emotional
dissonance also predicts high sympathetic activation and health
impairments (Gross & Levenson, 1997; Robinson & Demaree,
2007). In line with Morris and Feldman’s (1996; p. 992) argument
that “when mismatches between genuinely felt and organization-
ally required emotions exist, then, greater control, skill, and atten-
tive action will be needed,” we propose that coping with emotional
dissonance requires high self-control efforts and affects well-being
through the consumption of limited regulatory resources.
Although the framework for self-control processes has been
widely acknowledged in the literature on emotional labor (e.g.,
Figure 1. Theoretical model. Numbers refer to hypotheses.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2DIESTEL, RIVKIN, AND SCHMIDT
Judge et al., 2009), past research has largely failed to derive
implications for day-level effects of emotional dissonance on
well-being and, more important, protective mechanisms that facil-
itate coping with daily emotional dissonance. Drawing from the
theoretical assertion that day-specific emotional dissonance con-
sumes the limited resource for self-control functioning, we seek to
extend scholarly knowledge on such protective mechanisms by
focusing on processes, which promote the day-specific availability
of the limited resource taxed by emotional labor.
The Protective Function of Sleep Quality: Filling the
Limited Regulatory Resource
Theories on sleep quality and sleep disruption suggest that high
sleep quality buffers the negative day-level effects of emotional
dissonance on well-being because sleep as a psychological process
fuels and restores limited regulatory resources and thus, facilitates
self-control functioning during the following day (Barber, Munz,
Bagsby, & Powell, 2010; Beebe & Gozal, 2002). According to
Buyssé, Reynolds, Monk, Berman, and Kupfer (1989; p. 194),
sleep quality involves “[. . .] quantitative aspects of sleep, such as
sleep duration, sleep latency, or number of arousals, as well as
more purely subjective aspects, such as ‘depth’ or ‘restfulness’ of
sleep.” Because of the storing function of sleep (Killgore, 2010),
high sleep quality stabilizes the cerebral metabolic rate and ensures
adequate resource or energetic supply of the prefrontal cortex,
whose structures are important for executive self-control (Hof-
mann, Schmeichel, & Baddeley, 2012; Tabibnia et al., 2011). In
contrast, low sleep quality results in reduced metabolic activity in
the prefrontal cortex (Boonstra, Stins, Daffertshofer, & Beek,
2007), deficits in self-control functioning (Ghumman & Barnes,
2013) and hence low psychological well-being (Barber et al.,
2010). Experimental findings indicate that impaired sleep causes
self-control deficits, such as impaired decision making, response
inhibition, and attention control (e.g., Chuah, Venkatraman,
Dinges, & Chee, 2006) even after one night’s sleep loss (Nilsson
et al., 2005). To explain these deleterious effects, Barber et al.
(2010) explicitly refer to limited regulatory resources that are
diminished by impaired sleep.
Inspired by research on self-control and sleep quality, scholars
have analyzed the effects of impaired sleep on organizational
behavior. For example, Christian and Ellis (2011) found that
impaired sleep results in workplace deviance through diminished
self-control. Similarly, Wagner, Barnes, Lim, and Ferris (2012)
found a positive relation of impaired sleep to subsequent cyber-
loafing at work, which is associated with lower productivity.
Arguing in line with theories on sleep quality and sleep disruption,
the authors of these studies have suggested that because of their
low resources caused by impaired sleep, individuals are limited in
their ability to exert self-control at work. In particular, Christian
and Ellis (2011; p. 917) noted that “[. . .] sleep deprivation has the
potential to impair emotion regulation [as a form of self-control],
whereby individuals modulate [their] emotions [. . .] and how they
express them.”
Hypothesis Development: Interactions of Sleep Quality
and Emotional Dissonance
As elaborated above, on the basis of the self-control strength
model, we propose that high daily emotional dissonance taxes
limited regulatory resources and causes psychological strain
through decrements in these resources. Consistent with theories on
sleep quality and sleep disruption, we further argue that sleep
quality facilitates coping with daily emotional dissonance because
high sleep quality fuels the limited resource and thus, the day-
specific availability of that resource depends on the level of sleep
quality. Consequently, because of high resource availability, cop-
ing with daily emotional dissonance will be less straining if sleep
quality on the previous night is high. In contrast, in cases of low
sleep quality, emotional labor should become more difficult, and
employees’ effort toward self-control should deplete their regula-
tory resources at a greater rate. Thus, we expect day-specific
emotional dissonance to interact with sleep quality in predicting
psychological strain:
Hypothesis 1: Day-specific sleep quality moderates the posi-
tive relations of day-specific emotional dissonance to day-
specific ego depletion (Hypothesis 1a) and need for recovery
(Hypothesis 1b). The positive relations are attenuated as a
function of high sleep quality.
Work engagement is a fulfilling and motivational state of mind
reflected by perceived energy, vitality and mental resilience
(vigor), strong work-related involvement (dedication), and being
positively engrossed in and focused on work (absorption; Bakker,
2011). Inspired by this conceptualization, Sonnentag, Mojza, De-
merouti, and Bakker (2012; p. 844) argued that day-specific work
engagement is affected by the “availability of energetic and affec-
tive resources” and emotional regulation that consumes these
resources. This argument is consistent with Baumeister and Vohs’s
(2007) proposition that because of adaptive patterns of conserving
limited regulatory resources, high self-control efforts and associ-
ated decrements in these resources result in shifts of motivation,
which should manifest in lower vitality, impaired task involvement
and less absorption (see also Ryan & Deci, 2008). In support of
this proposition, Muraven, Gagné, and Rosman (2008) reported
impaired subjective vitality and diminished concentration after
high external demands on self-control. Consequently, state work
engagement is based upon the same limited resource as experi-
ences of exhaustion and thus, should also be predicted by interac-
tions of emotional dissonance and sleep quality:
Hypothesis 2: Day-specific sleep quality moderates the neg-
ative relations of day-specific emotional dissonance to day-
specific vigor (Hypothesis 2a), dedication (Hypothesis 2b),
and absorption (Hypothesis 2c). The negative relations are
attenuated as a function of high sleep quality.
Study 1
Method
Research design and participants. Because our model (see
Figure 1) is based on the proposition that the negative relations of
emotional dissonance to well-being are not uniform across all
days, we conducted a diary study. The day-specific interaction of
emotional dissonance and sleep quality on well-being was ana-
lyzed on the basis of a sample that involved employees from
different occupational contexts. We recruited our participants from
various organizations in Germany through announcements, indi-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
3
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
vidual contacts, and contacts of undergraduate management stu-
dents. A final sample of 63 participants (response rate: 67%) could
be employed for our study. Among the participants, 54% were
female, and 11.7% worked part time. Mean age was 36.32 (SD ⫽
13.98) years, and mean tenure was 16.11 (SD ⫽14.08) years. Most
of the participants worked in the services sector, with regular
contact with clients, patients, customers, or others. All participants
received €40 (approx. $53.26) as compensation for completing the
study. Three times per day, the participants completed question-
naires over 10 consecutive work days (617 daily measurement
points). On weekends or (public) holidays, the diary study was
interrupted and continued on the next regular work day. At morn-
ing, the participants rated sleep quality of the previous night. At
midday, day-specific emotional dissonance, and after the work
day, indicators of well-being were assessed.
Measures. At the beginning of the diary study, participants
completed a general survey. This survey included age, gender, and
trait negative affect (NA; five items with a 5-point Likert-scale:
distressed,upset,irritable,nervous, and afraid; Watson, Clark, &
Tellegen, 1988). NA predicts emotional mood and thus, may
explain individual differences in emotional dissonance and well-
being during the course of a week (Langeveld, Koot, & Passchier,
1999). In the day-specific questionnaires, we explained that the
items of sleep quality refer to the last night and the items of
emotional dissonance refer to momentary experience and to the
last hours of working time.
Sleep quality (morning). Day-specific sleep quality was mea-
sured with a shortened version of the Pittsburgh Sleep Quality
Index (PSQI, Buyssé et al., 1989), which was adapted for night-
specific assessment. That is, all items referred to the previous
night. For each person and each night, we calculated a day-specific
sleep quality score. In line with Buyssé et al. (1989, p. 194), our
score involves both subjective (sleeping quality and restfulness)
and objective components (sleep efficiency, sleep duration, and
sleep latency). Efficiency, duration, and latency were calculated on
the basis of participants’ reports on their bedtime, number of
minutes required to fall asleep, time of awakening in the morning,
and number of hours of sleep. Two items (4-point Likert scale),
assessed the participants’ sleep quality [“How would you rate the
quality of your previous night’s sleep?”; 0 (very good)to3(very
bad)] and restfulness [“This morning, how much of a problem has
it been for you to keep up enough enthusiasm to get things done?”;
0(not at all)to3(a very big problem)]. Because higher values of
the original PSQI indicate lower sleep quality or impaired sleep,
we recoded our score (range: 0–15), whose higher values now
reflect higher day-specific sleep quality.
Emotional dissonance (noon). We assessed day-specific
emotional dissonance on the basis of five items that reflected the
daytime frequency of experienced discrepancies between genu-
inely felt emotions and those required by participants’ job role
(e.g., “In the last few hours, how often did you have to show
feelings at work that you did not really feel?”). The items were
adapted from the Frankfurt Emotion Work Scales (Zapf, Vogt,
Seifert, Mertini, & Isic, 1999). The response format of this scale
ranged from 1 (never)to5(very often).
Ego depletion (evening). Day-specific ego depletion was as-
sessed using five items related to participants’ current experiences
with resource depletion and low willpower (e.g., “At the moment,
I feel increasingly less able to focus on anything.”). The scale was
developed and validated by Bertrams, Unger, and Dickhäuser
(2011), who intended to assess the psychological state of ego
depletion proposed by Muraven and Baumeister (2000). All items
are scored using a 4-point intensity rating format (1 ⫽not at all,
5⫽a great deal).
Need for recovery (evening). We assessed day-specific need
for recovery using five items from van Veldhoven and Broersen’s
(2003) scale (e.g., “Today, I cannot really show any interest in
other people when I have just come home myself.”). In essence,
this scale indicates the extent to which employees are incapable of
expressing interest in other things and perceive a high requirement
for a rest period to recover from straining activities. Again, all
items are scored using a 4-point intensity rating format (1 ⫽not at
all,5⫽a great deal).
Work engagement (evening). The assessment of day-specific
work engagement was based on the 9-item version of the Utrecht
Work Engagement Scale (Breevaart et al., 2012; Schaufeli, Bak-
ker, & Salanova, 2006), which was adapted for day-specific as-
sessment and involved three facets: vigor (e.g., “Today, I felt
strong and vigorous at my work.”), dedication (e.g., “Today, I was
enthusiastic about my job.”), and absorption (e.g., “Today, I was
immersed in my work.”). The response format ranges from 0
(never)to6(always).
Construct validity. To test validity of our measures, we per-
formed multilevel confirmatory factor analyses (MCFAs). MCFA
for emotional dissonance showed a well data fit:
2
(10) ⫽31.53,
p⬍.01, root mean square error of approximation (RMSEA) ⫽
.059, comparative fit index (CFI) ⫽.981, standardized root-mean-
square residual within-person/between-person (SRMR
w
/SRMR
b
)⫽
.023/.010. The distinctiveness of ego depletion, need for recovery,
and the three work engagement facets was tested in a 5-factor
model. The 5-factor model (
2
(284) ⫽633.74, p⬍.01, RM-
SEA ⫽.045, CFI ⫽.952, SRMR
w/b
⫽.033/.048) best fitted our
data compared with other models (4-factor model with one strain
factor:
2
(292) ⫽825.09, p⬍.01, RMSEA ⫽.054, CFI ⫽.927,
SRMR
w/b
⫽.040/.054; 3-factor model with one engagement fac-
tor:
2
(298) ⫽720.39, p⬍.01, RMSEA ⫽.048, CFI ⫽.942,
SRMR
w/b
⫽.040/.069; 2-factor model with one strain and one
engagement factor:
2
(302) ⫽906.63, p⬍.01, RMSEA ⫽.057,
CFI ⫽.917, SRMR
w/b
⫽.046/.071). Additional tests for work
engagement and both strain outcomes showed that models with
separate facets also best fitted the data (for information, please
consult the first author). Although need for recovery and ego
depletion are highly correlated, they refer to different aspects of
short-term strain: ego depletion reflects inner experience of ex-
haustion (Bertrams et al., 2011), whereas need for recovery is a
behavioral manifestation of exhaustion (van Veldhoven & Bro-
ersen, 2003). In line with studies on daily well-being (Breevaart et
al., 2012; Sonnentag & Zijlstra, 2006), we distinguished between
all measures (ego depletion, need for recovery, vigor, dedication,
and absorption). Our procedure aims at testing whether the hy-
pothesized interaction generalizes across different, albeit highly
correlated indicators of well-being (and is not specific to certain
operationalizations).
Analytical procedure. To test our hypotheses, we used step-
wise multilevel modeling with random intercepts because the
day-level data (Level 1) were nested within the person-level data
(Level 2) and this procedure takes into account the interdepen-
dence of both levels (Hox, 2002). All parameter specifications and
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
4DIESTEL, RIVKIN, AND SCHMIDT
estimations were conducted with the MLwiN program (Rasbash,
Steele, Browne, & Goldstein, 2012). The null model only included
the intercept. In Model 1, we added the Level 2 variables gender,
age, and trait NA; Model 2 included emotional dissonance and
sleep quality at Level 1; in Model 3, we tested the interactions of
emotional dissonance and sleep quality. All Level 1 and Level 2
variables (except gender) were centered around their grand mean
to reduce the risk of confounding effects (Hofmann & Gavin,
1998). To avoid multicollinearity when testing the interaction at
Level 1, we formed the interaction term by multiplying both of the
grand-mean centered predictors (emotional dissonance and sleep
quality) (Aiken & West, 1991). On a conceptual level, our center-
ing decision is grounded in the proposition that the day-specific
levels of emotional dissonance and sleep quality should be inter-
preted in absolute terms. For example, the PSQI has been validated
on the basis of reference samples and thus, is designed to reflect
the individual level of sleep quality relative to other individuals
(Buyssé et al., 1989).
Results
Table 1 displays the descriptive statistics and reliabilities of
the study variables. Before testing our hypotheses, we examined
the within-person (Level 1) variance in all five outcomes. For
ego depletion and need for recovery, the proportion of within-
person variation was 49.3% and 57.1%, respectively. For work
engagement, the Level 1variance was 44.1% (vigor), 45.7%
(dedication), and 45.2% (absorption). In line with our proposi-
tion of day-specific fluctuations, the results of variance decom-
position necessitate the application of multilevel modeling.
Test of hypotheses. Hypothesis 1 proposed day-specific in-
teractions of emotional dissonance and sleep quality on ego
depletion and need for recovery. In line with this proposition,
multilevel estimates revealed that emotional dissonance and
sleep quality significantly interacted in predicting both strain
outcomes at the day-level (see Table 2). Model 3 showed an
improved fit compared with Model 2, as indicated by the
difference in the log likelihood ratios. To facilitate the inter-
pretation of the interactions, we depicted the interactions and
performed simple slope tests, as recommended by Preacher,
Curran, and Bauer (2006). As Figures 2a and 2b show, the
interaction patterns are consistent with Hypothesis 1 (a-b). In
particular, day-specific emotional dissonance was strongly re-
lated to ego depletion and need for recovery (in the evening)
when day-specific sleep quality (last night) was low. In com-
parison, on days with high sleep quality, the positive relations
of day-specific emotional dissonance to both strain outcomes
were weaker. Thus, in support of Hypothesis 1, sleep quality
attenuated the positive effects of emotional dissonance on ego
depletion and need for recovery.
Hypothesis 2 proposed that day-specific sleep quality moderates
(buffers) the negative effect of day-specific emotional dissonance
on work engagement. A significant interaction of emotional dis-
sonance and sleep quality was only found for absorption, whereas
vigor and dedication failed to reflect an interaction of both (see
Table 3). For absorption, the improvement of model fit was sig-
nificant. As Figure 2c shows, the day-specific negative relation of
emotional dissonance to absorption is attenuated as a function of
sleep quality. After nights with high sleep quality, absorption was
less affected by daily emotional dissonance, whereas emotional
dissonance was negatively related to absorption when sleep quality
was low. Thus, Study 1 provides strong support for Hypothesis 2c,
but not for Hypotheses 2a and b.
Supplemental analyses. Because several authors have recom-
mended group-mean (or person-related) centering when testing
Level 1 interactions (Enders & Tofighi, 2007), we reanalyzed the
interactions of emotional dissonance and sleep quality using
person-related centered values of both predictors and their inter-
action. Our results revealed significant interactions of both predic-
tors on ego depletion, need for recovery, and absorption with signs
corresponding to expectations.
In addition, on the basis of person-related centering (Rauden-
bush & Bryk, 2002), we also analyzed slope variability of the
relations of emotional dissonance, sleep quality, and their interac-
tion. Our interest in slope variance results from recent findings,
which suggest that the day-level effects of emotional labor and
sleep quality on well-being largely differ between individuals
(Judge et al., 2009; van Dongen, Vitellaro, & Dinges, 2005). In
support of this suggestion, models with all three random slopes
Table 1
Means, Standard Deviations, Internal Consistencies (Cronbach’s Alpha) and Intercorrelations (Study 1)
Variable 1 2 345678910
1. Emotional dissonance—noon (.95) ⴚ.37 .57 .56 ⴚ.36 ⴚ.22 ⴚ.30
2. Sleep quality—previous night ⴚ.49 (.74) ⴚ.44 ⴚ.43 .43 .33 .37
3. Ego depletion—evening .73 ⴚ.60 (.94) .83 ⴚ.55 ⴚ.43 ⴚ.48
4. Need for recovery—evening .77 ⴚ.57 .88 (.92) ⴚ.49 ⴚ.37 ⴚ.42
5. Vigor—evening ⴚ.39 .52 ⴚ.59 ⴚ.53 (.89) .84 .89
6. Dedication—evening ⫺.21 .42 ⴚ.45 ⴚ.39 .87 (.91) .92
7. Absorption—evening ⴚ.29 .45 ⴚ.49 ⴚ.44 .91 .95 (.91)
8. Negative affect .44 ⴚ.46 .67 .57 ⴚ.58 ⴚ.57 ⴚ.56 (.76)
9. Age ⫺.03 .19 ⴚ.26 ⫺.17 .19 .13 .15 ⫺.18 —
10. Gender
a
⫺.08 .23 ⫺.14 ⫺.13 .23 .24 .28 ⫺.20 .34 —
M2.31 11.06 1.97 2.09 3.01 2.91 2.88 2.66 36.32 1.46
SD 0.74 1.99 0.64 0.62 1.15 1.19 1.13 0.70 13.98 0.50
Note. Cronbach’s alpha for day-level variables are mean internal consistencies averaged over all measurement days. Correlations below the diagonal are
person-level correlations (N⫽63). Correlations above the diagonal are day-level correlations (N⫽617). Numbers in bold p⬍.05.
a
Gender (1 ⫽female, 2 ⫽male).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
5
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
Table 2
Multilevel Estimates for Models Predicting Ego Depletion and Need for Recovery (Study 1)
Parameter
Ego depletion Need for recovery
Null model Model 1 Model 2 Model 3 Null model Model 1 Model 2 Model 3
(SE)(SE)(SE)(SE)(SE)(SE)(SE)(SE)
Fixed effects
␥
00
⫽Intercept 1.97
ⴱⴱ
(0.08) 1.89
ⴱⴱ
(0.20) 1.85
ⴱⴱ
(0.15) 1.87
ⴱⴱ
(0.15) 2.09
ⴱⴱ
(0.08) 2.09
ⴱⴱ
(0.20) 2.04
ⴱⴱ
(0.16) 2.05
ⴱⴱ
(0.16)
␥
01
⫽Gender 0.05 (0.14) 0.08 (0.10) 0.07 (0.10) 0.00 (0.14) 0.04 (0.10) 0.03 (0.10)
␥
02
⫽Age ⫺0.01 (0.00) ⫺0.01
ⴱ
(0.00) ⫺0.01
ⴱ
(0.00) ⫺0.00 (0.01) ⫺0.00 (0.00) ⫺0.00 (0.00)
␥
03
⫽Negative affect 0.60
ⴱⴱ
(0.07) 0.39
ⴱⴱ
(0.06) 0.41
ⴱⴱ
(0.06) 0.50
ⴱⴱ
(0.07) 0.27
ⴱⴱ
(0.06) 0.29
ⴱⴱ
(0.05)
␥
10
⫽Emotional dissonance (EmoDis) 0.31
ⴱⴱ
(0.04) 0.30
ⴱⴱ
(0.04) 0.31
ⴱⴱ
(0.04) 0.30
ⴱⴱ
(0.04)
␥
20
⫽sleep quality (SQ) ⫺0.05
ⴱⴱ
(0.01) ⫺0.05
ⴱⴱ
(0.01) ⫺0.07
ⴱⴱ
(0.01) ⫺0.06
ⴱⴱ
(0.01)
␥
30
⫽EmoDis ⫻SQ ⫺0.03
ⴱⴱ
(0.01) ⫺0.02
ⴱ
(0.01)
Random effects
2
⫽Residual variance at Level 1 0.36 0.36 0.30 0.29 0.44 0.44 0.37 0.36
Intercept
2⫽Residual variance at Level 2 0.37 0.17 0.08 0.08 0.33 0.20 0.09 0.09
—2
ⴱ
log (lh) 1267.62 1226.71 1082.55 1070.76 1375.12 1349.05 1206.94 1200.60
Diff—2
ⴱ
log (lh) 40.91
ⴱⴱ
144.16
ⴱⴱ
11.79
ⴱⴱ
26.07
ⴱⴱ
142.11
ⴱⴱ
6.34
ⴱ
Number of parameters 3 6893689
Note. Gender, age and negative affect are person-level (Level 2) variables; all other predictor variables are day-level (Level 1) variables.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
Figure 2. Interaction effects of emotional dissonance and sleep quality on
ego depletion (Figure 2a), need for recovery (Figure 2b) and absorption
(Figure 2c), Study 1. See the online article for the color version of this
figure.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
6DIESTEL, RIVKIN, AND SCHMIDT
Table 3
Multilevel Estimates for Models Predicting Work Engagement (Study 1)
Parameter
Vigor Dedication Absorption
Null model Model 1 Model 2 Model 3 Null model Model 1 Model 2 Model 3 Null model Model 1 Model 2 Model 3
(SE)(SE)(SE)(SE)(SE)(SE)(SE)(SE)(SE)(SE)(SE)(SE)
Fixed effects
␥00 ⫽Intercept 3.02ⴱⴱ (0.14) 2.65ⴱⴱ (0.39) 2.74ⴱⴱ (0.38) 2.72ⴱⴱ (0.38) 2.91ⴱⴱ (0.15) 2.41ⴱⴱ (0.41) 2.48ⴱⴱ (0.43) 2.46ⴱⴱ (0.43) 2.88ⴱⴱ (0.14) 2.28ⴱⴱ (0.37) 2.35ⴱⴱ (0.39) 2.32ⴱⴱ (0.38)
␥01 ⫽Gender 0.25 (0.24) 0.19 (0.23) 0.20 (0.23) 0.34 (0.24) 0.29 (0.26) 0.30 (0.26) 0.41 (0.23) 0.36 (0.24) 0.38 (0.24)
␥02 ⫽Age 0.01 (0.01) 0.00 (0.01) 0.00 (0.01) ⫺0.00 (0.01) ⫺0.00 (0.01) ⫺0.00 (0.01) ⫺0.00 (0.01) ⫺0.00 (0.01) ⫺0.00 (0.01)
␥03 ⫽Negative Affect ⫺0.90ⴱⴱ (0.19) ⫺0.61ⴱⴱ (0.19) ⫺0.63ⴱⴱ (0.18) ⫺0.94ⴱⴱ (0.18) ⫺0.71ⴱⴱ (0.19) ⫺0.73ⴱⴱ (0.18) ⫺0.84ⴱⴱ (0.19) ⫺0.58ⴱⴱ (0.20) ⫺0.60ⴱⴱ (0.20)
␥10 ⫽Emotional dissonance
(EmoDis) ⫺0.33ⴱⴱ (0.08) ⫺0.32ⴱⴱ (0.08) ⫺0.25ⴱⴱ (0.09) ⫺0.24ⴱⴱ (0.09) ⫺0.33ⴱⴱ (0.08) ⫺0.32ⴱⴱ (0.09)
␥20 ⫽sleep quality (SQ) 0.12ⴱⴱ (0.02) 0.11ⴱⴱ (0.02) 0.09ⴱⴱ (0.02) 0.08ⴱⴱ (0.02) 0.09ⴱⴱ (0.02) 0.08ⴱⴱ (0.02)
␥30 ⫽EmoDis ⫻SQ 0.03 (0.02) 0.03 (0.02) 0.04ⴱ(0.02)
Random effects
2⫽Residual variance at
Level 1 0.94 0.94 0.78 0.77 1.08 1.08 0.98 0.97 0.95 0.95 0.82 0.81
Intercept
2⫽Residual variance
at Level 2 1.19 0.73 0.67 0.66 1.28 0.79 0.84 0.83 1.15 0.72 0.74 0.73
—2ⴱlog (lh) 1876.84 1849.12 1738.95 1733.36 1957.65 1930.64 1877.12 1873.37 1880.35 1853.77 1770.72 1763.42
Diff—2ⴱlog (lh) 27.72ⴱⴱ 110.17ⴱⴱ 5.59ⴱ27.01ⴱⴱ 53.52ⴱⴱ 3.75 26.58ⴱⴱ 83.05ⴱⴱ 7.30ⴱⴱ
Number of parameters 3 6 8 9 3 6 8 9 3 6 8 9
Note. Gender, age and negative affect are person-level (Level 2) variables; all other predictor variables are day-level (Level 1) variables.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
7
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
(emotional dissonance, sleep quality, and their interaction) showed
an improvement in data fit compared with fixed slopes (ego
depletion: ⌬-2
ⴱ
log (⌬df)⫽39.13 (3); need for recovery: ⌬-2
ⴱ
log
(⌬df)⫽21.70 (3); vigor: ⌬-2
ⴱ
log (⌬df)⫽38.84 (3); dedication:
⌬-2
ⴱ
log (⌬df)⫽43.99 (3); absorption: ⌬-2
ⴱ
log (⌬df)⫽42.24 (3);
all ps ⬍.01). In performing the difference tests, we considered
all three slopes simultaneously, because the form of an inter-
action and its variation across higher level units does not only
depend on the varying slope of that interaction, but is also
contingent upon the varying slopes of the main effects (Aiken
& West, 1991). To avoid convergence problems and enable
summation of all three slope variances, we restricted the cova-
riance terms among the random slopes and intercepts to be zero.
The sum of slope variances of emotional dissonance, sleep
quality and their interaction indicate that the moderation effect
of sleep quality may take different shapes as a function of
individual traits (ego depletion (percentage of slope variance
relative to total variance): Slopes
2⫽0.08 (11%); need for recov-
ery: Slopes
2⫽0.06 (8%); vigor: Slopes
2⫽0.21 (10%); dedication:
Slopes
2⫽0.29 (12%); absorption: Slopes
2⫽0.21 (10%)).
Discussion of Study 1
The findings of Study 1 show that day-specific sleep quality
is a protective factor that buffers the effects of emotional
dissonance on psychological well-being. On days with high
emotional dissonance, high sleep quality prevented employees
from impaired well-being in terms of high strain and low
absorption. In contrast, low sleep quality in combination with
high emotional dissonance resulted in disproportionally low
well-being. Our finding highlights the necessity of focusing on
the daily mechanisms that provide employees with a high
availability of regulatory resources and that thus facilitate emo-
tion regulation at work.
Furthermore, our supplemental analyses revealed significant
interindividual differences in the day-specific relations of emo-
tional dissonance and sleep quality to well-being. That is, the form
of the moderating effect of sleep quality may be contingent upon
person-related variables, which may determine whether and to
what extent employees benefit from high sleep quality when they
have to cope with daily emotional dissonance. Our findings on
slope variations join recent observations, according to which the
negative effects of impaired sleep on self-control functioning dif-
fer between individuals (van Dongen et al., 2005). For example,
the effects of impaired sleep on inhibitory processes as a core
mechanism of emotional regulation (Gross, 1998) vary as a func-
tion of individuals’ ability to control inhibitory processes (Chuah
et al., 2006). Both this finding and the self-control perspective
suggest that self-control capacity may moderate the beneficial
impact of sleep quality on coping with emotional dissonance.
According to the model of self-control strength, people differ in
their capacity to exert self-control (Tangney et al., 2004). Thus,
people with high trait self-control can be expected to more effec-
tively cope with self-control demands compared with those with
low self-control capacity. Because we did not consider person-
related variables in Study 1, we conducted a second diary study
that aimed to disentangle the moderating role of self-control ca-
pacity in the interaction of daily emotional dissonance and sleep
quality on well-being.
Study 2
Self-Control Capacity and Conservation of Resources
As we elaborated above, experimental evidence suggests that
the effect of nightly sleep quality on self-control functioning
depends on one’s capacity to control or regulate attention and
behavior (Alhola & Polo-Kantola, 2007; van Dongen et al., 2005).
In line with Muraven and Baumeister’s (2000) definition of self-
control capacity, some individuals are more effective in modulat-
ing their emotional expression, resisting temptations, overcoming
inner resistance, or controlling their behavior than others (Tangney
et al., 2004). Self-control capacity is considered as a dispositional,
trait-like factor that operates as a protective resource when indi-
viduals face high demands on self-control. Compared with low
trait self-control, people with high trait self-control have greater
academic success, demonstrate better psychological adjustment,
and report higher self-acceptance (for review, de Ridder, Lensvelt-
Mulders, Finkenauer, Stok, & Baumeister, 2012). Less anger and
fewer interpersonal conflicts are also associated with high self-
control capacity, suggesting that trait self-control may facilitate
coping with emotional labor demands. In support of this sugges-
tion, Schmidt, Hupke, and Diestel (2012) revealed that trait self-
control moderates (attenuates) the positive relations of job-related
self-control demands to psychological strain.
To disentangle the potential interplay between trait self-control,
sleep quality, and emotional dissonance in predicting well-being, a
detailed understanding of the underlying mechanism of resource
depletion is needed. In elaborating on and extending the self-
control strength model, Muraven, Shmueli, and Burkley (2006)
proposed that individuals often allocate and conserve their limited
regulatory resources during self-control exertion, especially when
they anticipate further self-control demands (see also Baumeister
& Vohs, 2007; p. 11). One’s motivation to flexibly conserve
resources or selectively invest self-control effort is based upon
inner goals or actual priorities, which are rooted in the intention to
retain and protect resource supply (Hobfoll, 2002). That is, low
self-control performance may result from goal-based conservation
of resources (rather than “total” resource depletion). By demon-
strating that individuals can be highly adaptive in coping with
self-control demands, several experimental studies provide strong
support for the proposition of flexible conservation and selective
allocation of limited regulatory resources (Gröpel, Baumeister, &
Beckmann, 2014; Muraven et al., 2006; Vohs, Baumeister, &
Schmeichel, 2012).
The integration of the proposition of resource allocation with
theories on sleep quality and sleep disruption suggests that self-
control capacity moderates the day-specific interaction of emo-
tional dissonance and sleep quality on well-being. In particular,
compared with low self-control capacity, the buffering effect of
sleep quality on the negative relations of emotional dissonance to
well-being should be more pronounced for those with high trait
self-control. Our prediction of a three-way interaction is derived
from two implications that are grounded in the abovementioned
theories. First, compared with low trait self-control, individuals
with high self-control capacity are thought to be more effective in
monitoring and regulating their goal-directed behavior (Wan &
Sternthal, 2008) and, hence, more efficient in allocating their
limited regulatory resources (Gröpel et al., 2014; see also
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
8DIESTEL, RIVKIN, AND SCHMIDT
Baumeister & Alquist, 2009). Thus, when self-control capacity is
high, greater day-specific availability of resources provided by
high sleep quality should facilitate coping with daily emotional
dissonance and thus reduce the risk of impaired well-being. In
contrast, sleep quality should not facilitate emotional labor for
those with low self-control capacity because they are less able to
control their emotions effectively and to allocate their resources
efficiently.
Second, impaired sleep causes considerable decrements in met-
abolic activity in the prefrontal cortex and affects self-control
functioning by restricting day-specific availability of regulatory
resources (Boonstra et al., 2007; Nilsson et al., 2005). As a result,
individuals are less able to cope with emotional dissonance and to
prevent feelings of exhaustion, even when they are generally
capable of conserving and allocating resources. In other words, the
reduced availability of resources cannot be compensated by effec-
tive regulation of goal-directed behavior or efficient resource al-
location of those with high trait self-control. Thus, regardless of
one’s self-control capacity, low daily sleep quality should impair
emotional control and amplify the negative effects of dissonance
on well-being.
Hypothesis Development: Interactions of Sleep
Quality, Emotional Dissonance and Self-Control
Capacity
According to our lines of reasoning, we propose that when self-
control capacity is low, day-specific sleep quality should not moderate
the positive relations of day-specific emotional dissonance to ego
depletion and need for recovery. Because low trait self-control reflects
limited ability to regulate emotions or comply with display rules,
those with low self-control capacity are expected to experience in-
creases in strain with increasing day-specific emotional dissonance
even when sleep quality is high. Similarly, because of the diminished
resources resulting from sleep impairment, low sleep quality impairs
employees’ ability to cope with emotional dissonance and thus leads
to disproportionally high strain when emotions should be displayed
that are not truly felt, regardless of one’s self-control capacity. In
contrast, when trait self-control is high, increasing day-specific sleep
quality should prevent day-specific emotional dissonance to cause ego
depletion and need for recovery. Self-control capacity facilitates effi-
cient usage of resources, which are provided by high sleep quality,
and thus should protect employees from being strained when emo-
tional dissonance is high.
Hypothesis 3: Three-way interaction: Self-control capacity
moderates the day-specific interaction of sleep quality and
emotional dissonance on day-specific ego depletion (Hypoth-
esis 3a) and need for recovery (Hypothesis 3b). In cases of
high trait self-control, sleep quality moderates the positive
relations of emotional dissonance to both indicators of strain,
whereas in cases of low trait self-control, emotional disso-
nance and sleep quality do not interact in predicting strain.
Because the availability of regulatory resources influences mo-
tivational states, which become manifest in vitality, job involve-
ment, and absorption (Muraven et al., 2008), we hypothesize that
state work engagement should also be predicted by a three-way
interaction of emotional dissonance, sleep quality, and trait self-
control as they determine availability and usage of limited re-
sources. Our hypothesis is consistent with Xanthopoulou, Bakker,
and Fischbach’s (2013) resource-based view on work engagement,
according to which daily experiences of vigor, dedication, and
absorption are strongly influenced by self-regulatory processes of
conscious emotion control, goal-directed behavior, and being en-
grossed in work. Self-control capacity fosters such processes by
effective task-related action regulation and efficient allocation of
limited regulatory resources (de Ridder et al., 2012). Therefore,
state work engagement should also reflect the moderating effect of
self-control capacity on the day-specific interaction between emo-
tional dissonance and sleep quality.
Hypothesis 4: Three-way interaction: Self-control capacity
moderates the day-specific interaction between sleep quality
and emotional dissonance on day-specific vigor (Hypothesis
4a), dedication (Hypothesis 4b) and absorption (Hypothesis
4c). In cases of high trait self-control, sleep quality moderates
the negative relations of emotional dissonance to work en-
gagement, whereas in cases of low trait self-control, emotional
dissonance and sleep quality do not interact in predicting
engagement.
Method
Participants and research design. The procedure for recruit-
ing the participants and completing the diary study was exactly the
same as in Study 1. Again, we ideally asked people who were
employed in the services sector and who had daily work-related
contact with clients, patients, or customers. In sum, we recruited
108 people (response rate: 97.3%; 1,073 daily measurement
points). Among the participants, 49.1% were female, and 22.2%
worked part time. Mean age was 41.64 (SD ⫽13.34) years, and
mean tenure was 20.44 (SD ⫽13.47) years. For completing the
study, the participants received €40 (approx. $53.26).
Measures. We assessed NA, sleep quality (morning), emo-
tional dissonance (noon), ego depletion, need for recovery, and
engagement (evening) with the same scales from Study 1.
Self-control capacity (peer-rating). We assessed self-control
capacity with Tangney et al.’s (2004) self-control capacity scale,
which addresses several self-regulatory domains, namely, control
over thoughts, emotions, and impulses; motivation regulation; and
habit breaking. Because Tangney et al. (2004; pp. 282–283) found
that their scale comprises at least five dimensions, we focused on
seven items, which refer to one’s capacity for conscious attention
and behavioral control and one’s inclination toward deliberative
and nonimpulsive action. Muraven and Baumeister (2000) consid-
ered both aspects as core components of trait self-control. Others
aspects of this scale (work ethic, healthy habits, and reliability) are
domain-specific behavioral manifestations of self-control capacity
(de Ridder et al., 2012). Although we could not replicate the
factorial structure as reported by Tangney et al. (2004), the re-
duced 7-item scale yielded an acceptable model fit (
2
(14) ⫽
20.57, n.s., RMSEA ⫽.066, CFI ⫽.938, SRMR ⫽.055, AIC ⫽
2084.68, BIC ⫽2141.01). Models with all items failed to fit the
data (e.g., all 15 items loading on one factor:
2
(90) ⫽275.18, p⬍
.01, RMSEA ⫽.138, CFI ⫽.467, SRMR ⫽.114, AIC ⫽4428.16,
BIC ⫽4548.85).
For two reasons, we used a peer-rating procedure. First, differ-
ent sources of assessment reduce the risk of common method
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
9
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
biases. Second, meta-analyses on personality traits have shown
that peer-ratings of conscientiousness are valid and reliable (Con-
nelly & Ones, 2010). More importantly, the positive relation of this
trait to academic achievement is even stronger for peer-ratings
than for self-reports. That is, peers are very good at judging a
person’s self-discipline, trustworthiness, and tendency to be orga-
nized and controlled. Factor analyses suggest that conscientious-
ness is a manifestation of self-control capacity (Digman, 1997;
McCrae & Löckenhoff, 2010; Olson, 2005). Thus, peer ratings on
trait self-control should provide valid information about one’s
ability to exert self-control, especially in social interactions. The
participants were asked to select persons for peer rating who knew
them very well (e.g., friends, family members, colleagues; John-
son, 2000; Kolar, Funder, & Colvin, 1996). Each selected person
evaluated trait self-control of the corresponding participant. We
modified the self-control capacity scale items for peer rating by
specifically referring to the third person instead of the first person
(e.g., “She/he is impulsive,” “She/he often interrupts people.”).
The response format ranged from 1 (not at all)to5(very much). The
responses are scored such that higher values indicate higher trait
self-control in the target person. The item scores are averaged to
generate an overall measure (Bertrams & Dickhäuser, 2009).
Construct validity. We were able to replicate the factor struc-
tures provided in Study 1: the MCFA of emotional dissonance
indicated a good fit (
2
(10) ⫽32.41, p⬍.01, RMSEA ⫽.046,
CFI ⫽.984, SRMR
w/b
⫽.019/.003). The MCFAs of the outcomes
(ego depletion, need for recovery, vigor, dedication, and absorp-
tion) showed that the 5-factor model provided the best fit with the
data (
2
(284) ⫽795.66, p⬍.01, RMSEA ⫽.041, CFI ⫽.951,
SRMR
w/b
⫽.034/.053) compared with the other models (4-factor
model with both strain outcomes as one factor:
2
(292) ⫽1114.57,
p⬍.01, RMSEA ⫽.051, CFI ⫽.921, SRMR
w/b
⫽.040/.055;
3-factor model with one engagement factor:
2
(298) ⫽880.45,
p⬍.01, RMSEA ⫽.043, CFI ⫽.944, SRMR
w/b
⫽.038/.061;
2-factor model with one strain factor and one engagement factor:
2
(302) ⫽1194.20, p⬍.01, RMSEA ⫽.053, CFI ⫽.914,
SRMR
w/b
⫽.044/.062). Separate tests for the three engagement
facets and both strain facets confirm distinctiveness of all out-
comes.
Results
Table 4 displays the descriptive statistics and reliabilities for all
the study variables. Analyses of variation suggest substantial Level
1 variance in the outcomes (ego depletion: 47.2%; need for recov-
ery: 50.2%; vigor: 33.3%; dedication: 36.1%; absorption: 35.5%).
Test of hypotheses. We tested our hypotheses by comparing
four different models: In the null model, we included the intercept
as the only predictor. In Model 1, we added the control variables
(gender, age, and trait NA). In Model 2, we entered self-control
capacity at Level 2 and day-specific emotional dissonance, sleep
quality and their interaction at Level 1, to determine slope variance
at Level 1 (see Analytical Procedure and Supplemental Analyses
sections for Study 1). Finally, in Model 3, we introduced the
cross-level interactions (emotional dissonance ⫻self-control ca-
pacity, sleep quality ⫻self-control capacity, and emotional disso-
nance ⫻sleep quality ⫻self-control capacity). To provide an
unbiased and pure test of the hypothesized cross-level interactions,
both Level 1 variables and their interaction were centered around
the person-mean (centering within cluster; Enders & Tofighi,
2007; pp. 132–134). Before analyzing the four models, we tested
whether the relations of emotional dissonance and sleep quality to
all five outcomes and their interactions at Level 1 considerably
vary across the persons (Level 2). As in Study 1, the log-likelihood
difference tests indicated significant improvements in data fit of
the model with random slopes as compared with fixed slopes for
emotional dissonance, sleep quality, and their interaction (ego
depletion: ⌬-2
ⴱ
log (⌬df)⫽26.03 (3); need for recovery: ⌬-2
ⴱ
log
(⌬df)⫽19.35 (3); vigor: ⌬-2
ⴱ
log (⌬df)⫽39.02 (3); dedication:
⌬-2
ⴱ
log (⌬df)⫽27.31 (3); absorption: ⌬-2
ⴱ
log (⌬df)⫽29.64 (3);
all ps ⬍.01). In addition, the sum of slope variations also provide
evidence for potential cross-level interactions (ego depletion (per-
centage of slope variance relative to total variance): Slopes
2⫽0.04
(9%); need for recovery: Slopes
2⫽0.04 (8%); vigor: Slopes
2⫽0.14
(7%); dedication: Slopes
2⫽0.14 (7%); absorption: Slopes
2⫽0.14
(7%)).
As in Study 1, day-specific emotional dissonance, sleep quality
and their interaction term predicted all five outcomes with signs
corresponding to expectations (see Table 5 and 6). Hypothesis 3
Table 4
Means, Standard Deviations, Internal Consistencies (Cronbach’s Alpha) and Intercorrelations (Study 2)
Variable 1 2 34567891011
1. Emotional dissonance—noon (.96) ⴚ.14 .44 .44 ⴚ.19 ⴚ.12 ⴚ.16
2. Sleep quality—previous night ⫺.16 (.64) ⴚ.40 ⴚ.39 .26 .16 .19
3. Ego depletion—evening .53 ⴚ.53 (.93) .82 ⴚ.39 ⴚ.30 ⴚ.33
4. Need for recovery—evening .52 ⴚ.52 .88 (.92) ⴚ.40 ⴚ.32 ⴚ.36
5. Vigor—evening ⫺.17 .29 ⴚ.35 ⴚ.37 (.88) .85 .91
6. Dedication—evening ⫺.09 .15 ⴚ.24 ⴚ.28 .88 (.90) .92
7. Absorption—evening ⫺.14 .20 ⴚ.29 ⴚ.33 .95 .95 (.93)
8. Self-control capacity (peer-rating) ⫺.02 .11 ⫺.17 ⫺.13 ⫺.05 .05 ⫺.04 (.75)
9. Negative affect .36 ⴚ.44 .54 .51 ⴚ.27 ⴚ.22 ⴚ.25 ⫺.15 (.77)
10. Age ⫺.03 ⫺.05 .01 .04 .12 .02 .06 ⫺.11 .09 —
11. Gender
a
.06 .15 .06 .04 ⫺.02 .00 .01 ⫺.08 ⫺.16 ⫺.08 —
M2.13 11.99 1.68 1.80 3.21 3.17 3.16 3.57 2.38 41.64 1.51
SD 0.81 1.85 0.51 0.54 1.17 1.21 1.19 0.65 0.62 13.34 0.50
Note. Cronbach’s alpha for day-level variables are mean internal consistencies averaged over all measurement days. Correlations below the diagonal are
person-level correlations (N⫽108). Correlations above the diagonal are day-level correlations (N⫽1,073). Numbers in bold p⬍.05.
a
Gender (1 ⫽female, 2 ⫽male).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
10 DIESTEL, RIVKIN, AND SCHMIDT
Table 5
Multilevel Estimates for Models Predicting Ego Depletion and Need for Recovery (Study 2)
Parameter
Ego depletion Need for recovery
Null model Model 1 Model 2 Model 3 Null model Model 1 Model 2 Model 3
(SE)(SE)(SE)(SE)(SE)(SE)(SE)(SE)
Fixed effects
␥
00
⫽Intercept 1.68
ⴱⴱ
(0.05) 1.46
ⴱⴱ
(0.13) 1.48
ⴱⴱ
(0.13) 1.48
ⴱⴱ
(0.13) 1.80
ⴱⴱ
(0.05) 1.60
ⴱⴱ
(0.14) 1.61
ⴱⴱ
(0.14) 1.61
ⴱⴱ
(0.14)
␥
01
⫽Gender 0.15 (0.08) 0.14 (0.08) 0.14 (0.08) 0.14 (0.09) 0.13 (0.09) 0.13 (0.09)
␥
02
⫽Age ⫺0.00 (0.00) ⫺0.00 (0.00) ⫺0.00 (0.00) ⫺0.00 (0.00) ⫺0.00 (0.00) ⫺0.00 (0.00)
␥
03
⫽Negative Affect 0.46
ⴱⴱ
(0.08) 0.45
ⴱⴱ
(0.07) 0.45
ⴱⴱ
(0.07) 0.46
ⴱⴱ
(0.09) 0.46
ⴱⴱ
(0.09) 0.46
ⴱⴱ
(0.09)
␥
04
⫽Self-control capacity (SCC) ⫺0.06 (0.07) ⫺0.06 (0.07) ⫺0.04 (0.07) ⫺0.04 (0.07)
␥
10
⫽Emotional dissonance (EmoDis) 0.19
ⴱⴱ
(0.03) 0.19
ⴱⴱ
(0.03) 0.24
ⴱⴱ
(0.03) 0.24
ⴱⴱ
(0.03)
␥
20
⫽Sleep quality (SQ) ⫺0.05
ⴱⴱ
(0.01) ⫺0.05
ⴱⴱ
(0.01) ⫺0.06
ⴱⴱ
(0.01) ⫺0.05
ⴱⴱ
(0.01)
␥
30
⫽EmoDis ⫻SQ ⫺0.03
ⴱⴱ
(0.01) ⫺0.03
ⴱⴱ
(0.01) ⫺0.03
ⴱⴱ
(0.01) ⫺0.03
ⴱⴱ
(0.01)
␥
14
⫽EmoDis ⫻SCC ⫺0.13 (0.07) ⫺0.04 (0.08)
␥
24
⫽SQ ⫻SCC 0.01 (0.01) ⫺0.00 (0.01)
␥
34
⫽EmoDis ⫻SQ ⫻SCC ⫺0.03
ⴱⴱ
(0.01) ⫺0.02
ⴱ
(0.01)
Random effects
2
⫽Residual variance at Level 1 0.22 0.22 0.17 0.16 0.27 0.27 0.20 0.20
Slopes
2⫽Residual variance of the slopes
ab
0.04 0.03 0.04 0.04
Intercept
2⫽Residual variance at Level 2 0.24 0.16 0.16 0.16 0.26 0.18 0.19 0.19
—2
ⴱ
log (lh) 1,673.95 1,634.22 1,453.32 1,441.11 1,885.38 1,850.55 1,657.44 1,654.04
Diff—2
ⴱ
log (lh) 39.73
ⴱⴱ
180.90
ⴱⴱ
12.21
ⴱⴱ
34.83
ⴱⴱ
193.11
ⴱⴱ
3.40
Number of paramaters 3 6 13 16 3 6 13 16
Note. Gender, age, negative affect and self-control capacity are person-level (Level 2) variables; all other predictor variables are day-level (Level 1) variables.
a
We summed up all three random slopes (covariances among the random slopes and intercepts were restricted to be zero; see Supplemental Analyses section for Study 1).
b
On the basis of
non-restricted covariances, parameter estimates were identical to those reported (Pinheiro & Bates, 2000).
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
11
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
predicted that trait self-control moderates the day-specific interac-
tion of emotional dissonance and sleep quality on ego depletion
and need for recovery. As shown in Table 5, the three-way inter-
action was significant for both outcomes. For ego depletion, Model
3 provided a significantly better fit compared with Model 2,
whereas no significant improvement in model fit was found for
need for recovery. Figure 3 depicts the three-way interactions on
both outcomes: For those with high self-control capacity, the
interaction of emotional dissonance and sleep quality on both
outcomes was more pronounced as compared with low trait self-
control. That is, in cases of high trait self-control, sleep quality
attenuated the positive relations of emotional dissonance to ego
depletion and need for recovery. In contrast, when trait self-control
was low, emotional dissonance was significantly and positively
related to both outcomes regardless of sleep quality. Thus, data of
Study 2 lend strong support for Hypothesis 3a and partial support
for Hypothesis 3b (because of the insignificant difference test).
Hypothesis 4 proposed that trait self-control moderates the inter-
action of day-specific emotional dissonance and sleep quality on
work engagement. As Table 6 displays, the three-way interactions
were significant for all three engagement facets. Each Model 3
yielded a significant improvement in data fit. Figure 4 demon-
strates patterns of the three-way interactions: When trait self-
control was high, the negative relations of emotional dissonance to
engagement were attenuated as a function of day-specific sleep
quality. In contrast, for those with low self-control capacity, sleep
quality did not moderate the negative relations of emotional dis-
sonance to all three engagement components. In conclusion, the
three-way interactions patterns were also consistent with Hypoth-
eses 4a–c.
1
General Discussion
A growing body of empirical evidence substantiates the theoretical
proposition that emotional labor can involve self-control and thus
consume a limited regulatory resource, especially when employees
perceive high discrepancies between felt and required emotions. On
the basis of this proposition, we sought to identify moderators, which
determine day-specific availability as well as usage of that resource
and thus, have the potential to attenuate the negative relations of
day-specific emotional dissonance to psychological well-being.
Theories on sleep quality and sleep disruption (Beebe & Gozal,
2002) as well as models on self-control (Muraven & Baumeister,
2000) strongly suggest that both day-specific sleep quality and trait
self-control counteract the deleterious effects of emotional dissonance
on strain (ego depletion and need for recovery) and work engagement
(vigor, dedication, and absorption). In two diary studies, we found
day-specific sleep quality to attenuate the negative relations of emo-
tional dissonance to daily well-being. In Study 2, we tested whether
self-control capacity moderates the day-specific interaction of emo-
tional dissonance and sleep quality on well-being. In support of our
predictions, when trait self-control was high, high day-specific sleep
quality attenuated the positive relations of emotional dissonance to
need for recovery and ego depletion. However, in cases of low trait
self-control, sleep quality did not interact with dissonance to predict
1
We also tested the three-way interactions with the full self-control
scale (see Measures section [self-control capacity] for Study 2). Our
supplementary analyses indicated that the three-way interaction was sig-
nificant in the prediction of day-specific ego depletion (p⬍.01), vigor
(p⬍.05), and absorption (p⬍.05) with patterns corresponding to our
hypotheses.
Table 6
Multilevel Estimates for Models Predicting Work Engagement (Study 2)
Vigor Dedication
Null model Model 1 Model 2 Model 3 Null model Model 1
Parameter (SE)(SE)(SE)(SE)(SE)(SE)
Fixed effects
␥
00
⫽Intercept 3.21
ⴱⴱ
(0.11) 3.39
ⴱⴱ
(0.34) 3.42
ⴱⴱ
(0.34) 3.42
ⴱⴱ
(0.34) 3.17
ⴱⴱ
(0.12) 3.29
ⴱⴱ
(0.36)
␥
01
⫽Gender ⫺0.12 (0.21) ⫺0.14 (0.21) ⫺0.14 (0.21) ⫺0.08 (0.22)
␥
02
⫽Age 0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 0.00 (0.01)
␥
03
⫽Negative affect ⫺0.55
ⴱⴱ
(0.19) ⫺0.57
ⴱⴱ
(0.18) ⫺0.57
ⴱⴱ
(0.18) ⫺0.44
ⴱ
(0.17)
␥
04
⫽Self-control capacity (SCC) ⫺0.15 (0.15) ⫺0.15 (0.15)
␥
10
⫽Emotional dissonance (EmoDis) ⫺0.23
ⴱⴱ
(0.06) ⫺0.24
ⴱⴱ
(0.06)
␥
20
⫽Sleep quality (SQ) 0.08
ⴱⴱ
(0.02) 0.08
ⴱⴱ
(0.02)
␥
30
⫽EmoDis ⫻SQ 0.06
ⴱⴱ
(0.02) 0.05
ⴱⴱ
(0.01)
␥
14
⫽EmoDis ⫻SCC 0.25
ⴱ
(0.11)
␥
24
⫽SQ ⫻SCC ⫺0.01 (0.02)
␥
34
⫽EmoDis ⫻SQ ⫻SCC 0.06
ⴱⴱ
(0.02)
Random effects
2
⫽Residual variance at Level 1 0.65 0.65 0.48 0.48 0.77 0.77
Slopes
2⫽Residual variance of the slopes
ab
0.14 0.12
Intercept
2⫽Residual variance at Level 2 1.30 1.17 1.18 1.18 1.37 1.30
—2
ⴱ
log (lh) 2,905.88 2,895.20 2,725.57 2,710.05 3,084.09 3,078.59
Diff—2
ⴱ
log (lh) 10.68
ⴱ
169.63
ⴱⴱ
15.52
ⴱⴱ
5.50
Number of parameters 3 6 13 16 3 6
Note. Gender, age, negative affect and self-control capacity are person-level (Level 2) variables; all other predictor variables are day-level (Level 1) variables.
a
We summed up all three random slopes (covariances among the random slopes and intercepts were restricted to be zero; see Supplemental Analyses
section for Study 1).
b
On the basis of non-restricted covariances, parameter estimates were identical to those reported (Pinheiro & Bates, 2000).
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
12 DIESTEL, RIVKIN, AND SCHMIDT
strain. We found similar three-way interaction patterns in predicting
state work engagement: those with low self-control capacity reported
decrements in day-specific work engagement with increasing emo-
tional dissonance regardless of their sleep quality of the previous
night. When trait self-control was high, the negative relations of
emotional dissonance to engagement were attenuated as s function of
sleep quality.
Theoretical Implications
Our findings have several implications for research on emotional
labor and occupational health. First, the protective function of sleep
quality in the daily emotional labor process extends recent findings on
the restoring function of sleep. For example, Sonnentag and Binnew-
ies (2013) found that sleep quality buffers the positive effects of NA
during the evening on NA on the next morning. According to our
findings, the daytime effects of sleep quality also manifest in high
well-being even when employees experience high discrepancies be-
tween felt and required emotions. Thus, and consistent with theories
on sleep quality and sleep disruption, sleep quality facilitates not only
recovery from high strain on the previous day but also emotion
regulation during the following day.
Second, the interactions of sleep quality and emotional disso-
nance on daily work engagement provide further insight into the
psychological function of regulatory resources for work engage-
ment (Sonnentag & Grant, 2012). Drawing from the finding that
resource depletion at the morning (due to low sleep quantity) is
negatively related to engagement at the afternoon, Lanaj, Johnson,
and Barnes (2014; p. 20) pointed out that “the resource perspective
provides a more holistic understanding of the factors that predict
daily work engagement.” According to our results, work engage-
ment results from a combination of several factors, which influ-
ence the daily resource levels. Thus, and in line with the notion
that regulatory resources influences motivational states of vitality,
involvement, and being engrossed, daily work engagement de-
pends on the interplay of psychological mechanisms, which enable
employees to invest their limited resources in their job role and
ignore nonwork distractions.
Third, the moderating effects of trait self-control advance our
knowledge about individual differences in how people cope with
emotional dissonance. On the one hand, self-control capacity attenu-
ated the negative relations of emotional dissonance to engagement.
This finding is consistent with Schmidt et al.’s (2012) argument that
compared with low trait self-control, those with high self-control
capacity are less vulnerable to the depleting effects of job-related
self-control demands because they are better able to regulate their
behavior, emotions, and motivational impulses. In support of this
argument, our results show that trait self-control stabilizes work
engagement when daily emotional dissonance is high. On the other
hand, trait self-control moderates the day-specific interaction of sleep
quality and emotional dissonance on well-being. This result indicates
that on days with high emotional dissonance, individuals with high
self-control capacity benefit more from high sleep quality than do
individuals with low trait self-control. Interestingly, high self-control
capacity did not prevent employees from being strained when they
experienced high emotional dissonance and low sleep quality. Con-
sistent with the notion of conservation and allocation of resources, we
argue that daily availability and efficient usage of limited regulatory
resources are sine qua non (and together sufficient conditions) for
successful coping with high emotional dissonance and preventing
impaired psychological well-being. High sleep quality increases the
availability of limited resources and trait self-control ensures efficient
usage of that resource.
Dedication Absorption
Model 2 Model 3 Null model Model 1 Model 2 Model 3
(SE)(SE)(SE)(SE)(SE)(SE)
3.28
ⴱⴱ
(0.37) 3.28
ⴱⴱ
(0.37) 3.17
ⴱⴱ
(0.12) 3.25
ⴱⴱ
(0.36) 3.28
ⴱⴱ
(0.36) 3.28
ⴱⴱ
(0.36)
⫺0.08 (0.23) ⫺0.08 (0.23) ⫺0.06 (0.22) ⫺0.08 (0.22) ⫺0.08 (0.22)
0.00 (0.01) 0.00 (0.01) 0.01 (0.01) 0.01 (0.01) 0.01 (0.01)
⫺0.44
ⴱ
(0.17) ⫺0.44
ⴱ
(0.17) ⫺0.51
ⴱⴱ
(0.19) ⫺0.53
ⴱⴱ
(0.19) ⫺0.53
ⴱⴱ
(0.19)
0.03 (0.15) 0.03 (0.16) ⫺0.14 (0.15) ⫺0.14 (0.15)
0.22
ⴱⴱ
(0.06) ⫺0.23
ⴱⴱ
(0.06) ⫺0.22
ⴱⴱ
(0.06) ⫺0.23
ⴱⴱ
(0.06)
0.07
ⴱⴱ
(0.02) 0.07
ⴱⴱ
(0.02) 0.07
ⴱⴱ
(0.01) 0.06
ⴱⴱ
(0.01)
0.07
ⴱⴱ
(0.01) 0.07
ⴱⴱ
(0.01) 0.07
ⴱⴱ
(0.02) 0.07
ⴱⴱ
(0.01)
0.27
ⴱ
(0.12) 0.26
ⴱ
(0.10)
⫺0.00 (0.02) ⫺0.01 (0.02)
0.04
ⴱ
(0.02) 0.06
ⴱⴱ
(0.02)
0.63 0.63 0.77 0.74 0.60 0.60
0.14 0.13 0.14 0.12
1.31 1.31 1.37 1.24 1.24 1.24
2,955.88 2,944.41 3,035.14 3,026.05 2,893.77 2,877.81
122.71
ⴱⴱ
11.47
ⴱⴱ
9.09
ⴱ
132.28
ⴱⴱ
15.96
ⴱⴱ
13 16 3 6 13 16
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
13
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
Finally, our findings invite scholars to reconsider the psychological
role of protective factors in coping with job demands. Most studies on
protective factors have focused on two-way interactions of personal or
job resources and demands on well-being with divergent findings
(Bakker, 2011; Häusser, Mojzisch, Niesel, & Schulz-Hardt, 2010). In
an attempt to integrate these findings in an overarching theory, de
Jonge and Dormann (2006) have proposed the triple-match principle,
according to which a buffering effect of protective factors will be
most likely emerge, if demands, resources, and outcomes conceptu-
ally correspond to each other or draw on the same psychological
domain (that is, match to each other). For example, cognitive re-
sources should rather attenuate the effects of cognitive demands on
cognitive strains than emotional or physical resources (Van den Ven
& Vlerick, 2013).
However, the present interactions involve variables from different
psychological domains (emotional dissonance: emotional domain;
sleep quality: physical domain; trait self-control: behavioral/cognitive
domain) and thus, are not entirely consistent with the idea of triple-
match. Our findings complement a series of studies, which also
demonstrate interactions of variables from different domains (e.g.,
Dollard, Tuckey, & Dormann, 2012; Meier, Semmer, Elfering, &
Jacobshagen, 2008). For example, Dollard et al. (2012) found psy-
chosocial safety climate to moderate the interaction of emotional
demands and resources on distress, whereas Meier et al. (2008)
reported that high self-efficacy enables job control to buffer the
relations of demands to affective strain. In extending the theoretical
notion of triple-match, we argue that the match of demands, protective
factors, and outcomes is not limited to (surface-level) psychological
domains, but is also applicable to (deep-level) underlying psycholog-
ical mechanisms, which determine the observed moderating effects.
In our case, all variables refer to the functioning and availability of
regulatory resources. In comparison, the findings of Dollard et al.
(2012) reflect social-contextual dynamics of managerial support (psy-
chosocial safety) in enabling the usage of protective factors to cope
with job demands, whereas the study of Meier et al. (2008) revealed
the psychological function of perceived action control (self-efficacy
and job control) in facing job demands. In conclusion, scholarly
understanding of occupational health may benefit from a clearly
defined “mechanistic match” of demands, protective factors, and
outcomes.
Limitations and Avenues for Future Research
Our research is subject to several limitations. First, our study
variables were assessed with self-report measures. Thus, common
method variance may have biased the parameter estimates (Pod-
sakoff, MacKenzie, Lee, & Podsakoff, 2003). Scholars often cast
doubt on self-report data because even interactions may partially
Figure 3. Interaction effects of emotional dissonance, sleep quality, and self-control capacity on ego depletion
(Figure 3a) and need for recovery (Figure 3b), Study 2. See the online article for the color version of this figure.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
14 DIESTEL, RIVKIN, AND SCHMIDT
reflect semantic overlaps (Spector, 1994). To address this issue, we
used peer ratings for self-control capacity and thus, limited the
possibility of mutual contamination of the constructs. However,
future research should also consider physiological indicators, such
as cortisol awakening response (e.g., Stetler & Miller, 2005), to
assess stress or impaired sleep.
Second, the negative relations of daily emotional dissonance to
well-being could be partially (or fully) explained by mood, which
may have caused both variables to be related. However, we did not
assess day-specific mood and thus, were not able to rule out that
daily mood drives the emotional labor process. To address this
issue, we examined the role of trait NA, because day-specific
changes in mood vary as a function of interindividual differences
in affectivity (Ilies, Dimotakis, & Watson, 2010). Thus, we con-
trolled for individual variation in the level of daily emotional
dissonance and well-being. In addition, we further tested whether
Figure 4. Interaction effects of emotional dissonance, sleep quality, and self-control capacity on ego depletion
(Figure 4a), vigor (Figure 4b) and absorption (Figure 4c), Study 2. See the online article for the color version
of this figure.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
15
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
trait NA moderated the negative relations of emotional dissonance
to well-being (with all other variables and interactions included).
In Study 1, NA amplified the negative relation of emotional
dissonance to vigor, whereas the positive relation of dissonance to
need for recovery was amplified by NA, in Study 2. Thus, our
results indicate that one’s inclination toward negative mood par-
tially explains the effects of emotional dissonance on well-being.
However, given that mood exhibits high intraindividual variation
(Sonnentag & Grant, 2012), we only provided a proximal test of a
third variable confoundation of the relations examined.
Nevertheless, because all predicted interactions of emotional
dissonance, sleep quality and trait self-control remained significant
in the additional analyses, we argue that regardless of the level of
daily mood, discrepancies between felt and required emotions tax
limited resources and thus, cause impaired well-being. According
to Tice and Bratslavsky (2000), the self-control pressure on limited
regulatory resources indeed increases with negative mood, which
may heighten the likelihood of high emotional dissonance and high
ego depletion during the working day (e.g., Sonnentag & Binnew-
ies, 2013). However, the underlying mechanism of diminished
resources would still explain the effects of emotional dissonance
on well-being, even when daily NA is high (see also Lanaj et al.,
2014; p. 16). Thus, besides sleep quality, daily mood can be
viewed as another factor, which determines day-specific availabil-
ity of resources for self-control functioning (Robinson & Demaree,
2007).
Third, although we assessed sleep quality, emotional disso-
nance, and well-being at different points in time during the work-
ing days, our correlational data structure did not permit strong
causal conclusions. For example, impaired well-being may have
caused emotional dissonance during the day. This concern is
further accentuated by the time frame that was used to measure
work engagement (“during the day today”). In addition, Sonnentag
et al. (2012) provided evidence on the reciprocal relations of
day-specific recovery experiences to engagement, indicating that
low engagement can diminish the availability of resources. To test
opposite causal directions, we analyzed relations of all five out-
comes and sleep quality to emotional dissonance. While the five
outcomes significantly predicted emotional dissonance, no inter-
actions of well-being and sleep quality on emotional dissonance
were found in Study 1. In Study 2, only absorption interacted with
sleep quality to predict emotional dissonance, whereas all other
interactions (including the three-way interactions) were not signif-
icant (with random slopes, some two-way interactions were sig-
nificant at p⬍.05). Although we cannot fully rule out reverse
causation, our analyses provide some evidence for hypothesized
directions of causality. Furthermore, past research has strongly
suggested that emotional dissonance and other demands result in
impaired well-being, not vice versa (Diestel & Schmidt, 2011a,
2011b; Sonnentag, Binnewies, & Mojza, 2010). Thus, low re-
sources owing to, for example, impaired sleep may increase the
likelihood of loss spirals involving reciprocal relations among
demands and well-being. However, job demands most likely drive
these relationships, which are further influenced by impaired well-
being.
Finally, our findings suggest job performance as an outcome of
emotional dissonance, sleep quality, and trait self-control. In our
study, daily engagement reflected the combined effects of these
three variables, and past research has revealed that engagement is
strongly related with performance (Christian, Garza, & Slaughter,
2011). In light of the relevance of self-control for action regulation
at work (Hacker, 2005), fluctuations in performance should also be
influenced by sleep quality and self-control capacity (Halbesleben
& Wheeler, 2011).
Practical Implications
Our results suggest that people who sleep better and have high
trait self-control cope best with emotional dissonance. In the long
run, sleep quality can be enhanced by sleep hygiene programs that
address sleep sufficiency (average sleep duration) and consistency
(low variation in sleep duration). Such programs may include
preventive tactics, such as sleep awareness, sleep rituals, and
insomnia-reduction strategies (e.g., Brown, 2004). Whereas our
theoretical framework that explains the moderating effect of sleep
quality has been only tentatively tested on the basis of natural
variation in sleep quality, such programs would also provide an
internally valid design, which allows for testing causality on the
basis of comparisons between those, who undergo a sleep training
and a waiting group.
However, in occupations with stressful time schedules, shifts, or
serviceability, organizational support systems and psychosocial
safety climate (Dollard et al., 2012) are imperative to stabilize
well-being when sleep quality is low (Christian & Ellis, 2011).
Support systems and safety climate foster awareness of impaired
sleep and the associated risks of low self-control, and provide
opportunities to recover from high stress. Because sleep consis-
tency reduces day-specific fluctuations in sleep quality (Barber et
al., 2010), our results also suggest that organizations with night
shifts provide constant shift schedules (e.g., over the period of a
week), which enable employees to adapt their sleep routines and
thus their circadian rhythms to required attendance times. Given
that many organizations (hospitals or nursing homes) change shift
schedules quite frequently (Gumenyuk, Roth, & Drake, 2012),
service delivery or other forms of interaction with clients should
account for disproportionally high strain because of high emo-
tional labor demands. Thus, those organizations are well advised to
ensure high sleep quality through constant shift schedules.
In addition, although self-control capacity is considered as a
stable trait, training efforts may improve one’s ability to exert
self-control. Intervention studies aiming to enhance trait self-
control have shown that repeated and controlled exertion of self-
control can lead to improved behavioral regulation, executive
functioning, and emotion control (Baumeister, Gailliot, DeWall, &
Oaten, 2006). For example, Oaten and Cheng (2007) reported that
students who participated in such an intervention showed higher
performance on laboratory self-control tasks, not only in the
trained self-control domain but also in a wide range of other
domains, such as emotion control. Thus, the development of train-
ing interventions tailored to, for example, emotional labor in
service jobs would be a promising direction for future research.
Finally, self-control capacity and sleep quality are inherently
related to the psychological domains of human functioning and,
together, form a strong resource basis for effective, goal-directed
self-control (Hagger et al., 2010). In particular, improvements in
both sleep consistency and sleep sufficiency enhance self-control
capacity in the long run, whereas high self-control capacity fosters
sleep hygiene, which promotes daily resource recovery (Barber et
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
16 DIESTEL, RIVKIN, AND SCHMIDT
al., 2010). Thus, to improve both sleep hygiene and self-control
capacity, occupational health programs may benefit from our find-
ing that both sleep quality and trait self-control jointly interact with
job demands to influence psychological well-being.
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
interpreting interactions. Newbury Park, CA: Sage.
Alhola, P., & Polo-Kantola, P. (2007). Sleep deprivation: Impact on
cognitive performance. Neuropsychiatric Disease and Treatment, 3,
553–567.
Bakker, A. (2011). An evidence-based model of work engagement. Cur-
rent Directions in Psychological Science, 20, 265–269. http://dx.doi.org/
10.1177/0963721411414534
Barber, L. K., Munz, D. C., Bagsby, P. G., & Powell, E. D. (2010). Sleep
consistency and sufficiency: Are both necessary for less psychological
strain? Stress & Health, 26, 186–193. http://dx.doi.org/10.1002/smi
.1292
Baumeister, R. F., & Alquist, J. L. (2009). Is there a downside to good
self-control? Self and Identity, 8, 115–130. http://dx.doi.org/10.1080/
15298860802501474
Baumeister, R. F., Gailliot, M., DeWall, C. N., & Oaten, M. (2006).
Self-regulation and personality: How interventions increase regulatory
success, and how depletion moderates the effects of traits on behavior.
Journal of Personality, 74, 1773–1802. http://dx.doi.org/10.1111/j
.1467-6494.2006.00428.x
Baumeister, R. F., & Vohs, K. D. (2007). Self-Regulation, ego depletion,
and motivation. Social and Personality Psychology Compass, 1, 115–
128. http://dx.doi.org/10.1111/j.1751-9004.2007.00001.x
Beebe, D. W., & Gozal, D. (2002). Obstructive sleep apnea and the
prefrontal cortex: Towards a comprehensive model linking nocturnal
upper airway obstruction to daytime cognitive and behavioral deficits.
Journal of Sleep Research, 11, 1–16. http://dx.doi.org/10.1046/j.1365-
2869.2002.00289.x
Bertrams, A., & Dickhäuser, O. (2009). Messung dispositioneller Selbst-
kontrollkapazität. Eine deutsche Adaptation der Kurzform der Self-
Control Scale (SCS-K-D). [Measuring dispositional self-control capac-
ity: A German adaptation of the short form of the Self- Control Scale
(SCS-K-D)]. Diagnostica, 55, 2–10. http://dx.doi.org/10.1026/0012-
1924.55.1.2
Bertrams, A., Unger, A., & Dickhäuser, O. (2011). Momentan verfügbare
Selbstkontrollkraft: Vorstellung eines Messinstruments und erste Be-
funde aus pädagogisch-psychologischen Kontexten. [Momentarily avail-
able self-control strength: Introduction of a measure and first findings
from educational-psychological contexts.]. Zeitschrift fur Padagogische
Psychologie, 25, 185–196. http://dx.doi.org/10.1024/1010-0652/
a000042
Boonstra, T. W., Stins, J. F., Daffertshofer, A., & Beek, P. J. (2007).
Effects of sleep deprivation on neural functioning: An integrative re-
view. Cellular and Molecular Life Sciences, 64, 934–946. http://dx.doi
.org/10.1007/s00018-007-6457-8
Breevaart, K., Bakker, A. B., Demerouti, E., & Hetland, J. (2012). The
measurement of state work engagement: A multilevel factor analytic
study. European Journal of Psychological Assessment, 28, 305–312.
http://dx.doi.org/10.1027/1015-5759/a000111
Brown, M. (2004). Taking care of business: Self-help and sleep medicine
in American corporate culture. Journal of Medical Humanities, 25,
173–187. http://dx.doi.org/10.1023/B:JOMH.0000036636.43638.03
Buyssé, D. J., Reynolds, C. F., III, Monk, T. H., Berman, S. R., & Kupfer,
D. J. (1989). The Pittsburgh Sleep Quality Index: A new instrument for
psychiatric practice and research. Psychiatry Research, 28, 193–213.
http://dx.doi.org/10.1016/0165-1781(89)90047-4
Cheung, F., & Tang, C. (2010). The influence of emotional dissonance on
subjective health and job satisfaction: Testing the stress–strain–outcome
model. Journal of Applied Social Psychology, 40, 3192–3217. http://dx
.doi.org/10.1111/j.1559-1816.2010.00697.x
Christian, M. S., & Ellis, A. P. J. (2011). Examining the effects of sleep
deprivation on workplace deviance: A self-regulatory perspective. Acad-
emy of Management Journal, 54, 913–934. http://dx.doi.org/10.5465/
amj.2010.0179
Christian, M. S., Garza, A. S., & Slaughter, J. E. (2011). Work engage-
ment: A quantitative review and test of its relations with task and
contextual performance. Personnel Psychology, 64, 89–136. http://dx
.doi.org/10.1111/j.1744-6570.2010.01203.x
Chuah, Y. M., Venkatraman, V., Dinges, D. F., & Chee, M. W. (2006). The
neural basis of interindividual variability in inhibitory efficiency after
sleep deprivation. The Journal of Neuroscience, 26, 7156–7162. http://
dx.doi.org/10.1523/JNEUROSCI.0906-06.2006
Connelly, B. S., & Ones, D. S. (2010). An other perspective on personality:
Meta-analytic integration of observers’ accuracy and predictive validity.
Psychological Bulletin, 136, 1092–1122. http://dx.doi.org/10.1037/
a0021212
de Jonge, J., & Dormann, C. (2006). Stressors, resources, and strain at
work: A longitudinal test of the triple-match principle. Journal of
Applied Psychology, 91, 1359–1374. http://dx.doi.org/10.1037/0021-
9010.91.5.1359
de Ridder, D. T. D., Lensvelt-Mulders, G., Finkenauer, C., Stok, F. M., &
Baumeister, R. F. (2012). Taking stock of self-control: A meta-analysis
of how trait self-control relates to a wide range of behaviors. Personality
and Social Psychology Review, 16, 76–99. http://dx.doi.org/10.1177/
1088868311418749
Diestel, S., & Schmidt, K.-H. (2011a). The moderating role of cognitive
control deficits in the link from emotional dissonance to burnout symp-
toms and absenteeism. Journal of Occupational Health Psychology, 16,
313–330. http://dx.doi.org/10.1037/a0022934
Diestel, S., & Schmidt, K.-H. (2011b). Costs of simultaneous coping with
emotional dissonance and self-control demands at work: Results from
two German samples. Journal of Applied Psychology, 96, 643–653.
http://dx.doi.org/10.1037/a0022134
Digman, J. M. (1997). Higher-order factors of the Big Five. Journal of
Personality and Social Psychology, 73, 1246–1256. http://dx.doi.org/
10.1037/0022-3514.73.6.1246
Dollard, M. F., Tuckey, M. R., & Dormann, C. (2012). Psychosocial safety
climate moderates the job demand-resource interaction in predicting
workgroup distress. Accident Analysis & Prevention, 45, 694–704.
http://dx.doi.org/10.1016/j.aap.2011.09.042
Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in
cross-sectional multilevel models: A new look at an old issue. Psycho-
logical Methods, 12, 121–138. http://dx.doi.org/10.1037/1082-989X.12
.2.121
Ghumman, S., & Barnes, C. M. (2013). Sleep and prejudice: A resource
recovery approach. Journal of Applied Social Psychology, 43, E166–
E178. http://dx.doi.org/10.1111/jasp.12045
Gröpel, P., Baumeister, R. F., & Beckmann, J. (2014). Action versus state
orientation and self-control performance after depletion. Personality and
Social Psychology Bulletin, 40, 476–487. http://dx.doi.org/10.1177/
0146167213516636
Gross, J. J. (1998). Antecedent- and response-focused emotion regulation:
Divergent consequences for experience, expression, and physiology.
Journal of Personality and Social Psychology, 74, 224–237. http://dx
.doi.org/10.1037/0022-3514.74.1.224
Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: The acute effects
of inhibiting negative and positive emotion. Journal of Abnormal Psy-
chology, 106, 95–103. http://dx.doi.org/10.1037/0021-843X.106.1.95
Gumenyuk, V., Roth, T., & Drake, C. L. (2012). Circadian phase, sleepi-
ness, and light exposure assessment in night workers with and without
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
17
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
shift work disorder. Chronobiology International, 29, 928–936. http://
dx.doi.org/10.3109/07420528.2012.699356
Hacker, W. (2005). Allgemeine Arbeitspsychologie: Psychische Regulation
von Wissens-, Denk- und körperlicher Arbeit (2. vollständig überarbe-
itete und erweiterte Ausgabe) [General work psychology: Psychological
regulation of knowledge, mental and physical work]. Bern, Switzerland:
Huber.
Hagger, M. S., Wood, C., Stiff, C., & Chatzisarantis, N. L. D. (2010). Ego
depletion and the strength model of self-control: A meta-analysis. Psy-
chological Bulletin, 136, 495–525. http://dx.doi.org/10.1037/a0019486
Halbesleben, J. R. B., & Wheeler, A. R. (2011). I owe you one: Coworker
reciprocity as a moderator of the day-level exhaustion-performance
relationship. Journal of Organizational Behavior, 32, 608–626. http://
dx.doi.org/10.1002/job.748
Häusser, J. A., Mojzisch, A., Niesel, M., & Schulz-Hardt, S. (2010). Ten
years on: A review of recent research on the job demand-control
(-support) model and psychological well-being. Work & Stress, 24,
1–35. http://dx.doi.org/10.1080/02678371003683747
Hobfoll, S. E. (2002). Social and psychological resources and adaptation.
Review of General Psychology, 6, 307–324. http://dx.doi.org/10.1037/
1089-2680.6.4.307
Hochschild, A. (1983). The managed heart. Berkeley, CA: University of
California Press.
Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in hierar-
chical linear models: Implications for research organizations. Journal of
Management, 24, 623–641. http://dx.doi.org/10.1177/01492063
9802400504
Hofmann, W., Schmeichel, B. J., & Baddeley, A. D. (2012). Executive
functions and self-regulation. Trends in Cognitive Sciences, 16, 174–
180. http://dx.doi.org/10.1016/j.tics.2012.01.006
Hox, J. (2002). Multilevel analysis: Techniques and applications. Mahwah,
NJ: Erlbaum.
Hülsheger, U. R., & Schewe, A. F. (2011). On the costs and benefits of
emotional labor: A meta-analysis of three decades of research. Journal
of Occupational Health Psychology, 16, 361–389. http://dx.doi.org/
10.1037/a0022876
Ilies, R., Dimotakis, N., & De Pater, I. E. (2010). Psychological and
physiological reactions to high workloads: Implications for well-being.
Personnel Psychology, 63, 407–436. http://dx.doi.org/10.1111/j.1744-
6570.2010.01175.x
Ilies, R., Dimotakis, N., & Watson, D. (2010). Mood, blood pressure, and
heart rate at work: An experience-sampling study. Journal of Occupa-
tional Health Psychology, 15, 120–130. http://dx.doi.org/10.1037/
a0018350
Johnson, J. A. (2000). Predicting observers’ ratings of the Big Five from
the CPI, HPI, and NEO–PI–R: A comparative validity study. European
Journal of Personality, 14, 1–19. http://dx.doi.org/10.1002/(SICI)1099-
0984(200001/02)14:1⬍1::AID-PER349⬎3.0.CO;2-E
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. http://dx.doi.org/10.1111/j
.1744-6570.2008.01129.x
Killgore, W. D. (2010). Effects of sleep deprivation on cognition. Progress
in Brain Research, 185, 105–129. http://dx.doi.org/10.1016/B978-0-
444-53702-7.00007-5
Kolar, D. W., Funder, D. C., & Colvin, C. R. (1996). Comparing the
accuracy of personality judgments by the self and knowledgeable others.
Journal of Personality, 64, 311–337. http://dx.doi.org/10.1111/j.1467-
6494.1996.tb00513.x
Kühnel, J., Sonnentag, S., & Bledow, R. (2012). Resources and time
pressure as day-level antecedents of work engagement. Journal of Oc-
cupational and Organizational Psychology, 85, 181–198. http://dx.doi
.org/10.1111/j.2044-8325.2011.02022.x
Lanaj, K., Johnson, R. E., & Barnes, C. M. (2014). Beginning the workday
yet already depleted? Consequences of late-night smartphone use and
sleep. Organizational Behavior and Human Decision Processes, 124,
11–23. http://dx.doi.org/10.1016/j.obhdp.2014.01.001
Langeveld, J. H., Koot, H. M., & Passchier, J. (1999). Do experienced
stress and trait negative affectivity moderate the relationship between
headache and quality of life in adolescents? Journal of Pediatric Psy-
chology, 24, 1–11. http://dx.doi.org/10.1093/jpepsy/24.1.1
McCrae, R. R., & Löckenhoff, C. E. (2010). Self-regulation and the
five-factor model of personality traits. In R. H. Hoyle (Ed.), Handbook
of personality and self-regulation (pp. 145–168). Malden, MA: Black-
well Publishing. http://dx.doi.org/10.1002/9781444318111.ch7
Meier, L. L., Semmer, N. K., Elfering, A., & Jacobshagen, N. (2008). The
double meaning of control: Three-way interactions between internal
resources, job control, and stressors at work. Journal of Occupational
Health Psychology, 13, 244–258. http://dx.doi.org/10.1037/1076-8998
.13.3.244
Morris, J. A., & Feldman, D. C. (1996). The dimensions, antecedents, and
consequences of emotional labor. The Academy of Management Review,
21, 986–1010.
Muraven, M., & Baumeister, R. F. (2000). Self-regulation and depletion of
limited resources: Does self-control resemble a muscle? Psychological
Bulletin, 126, 247–259. http://dx.doi.org/10.1037/0033-2909.126.2.247
Muraven, M., Gagné, M., & Rosman, H. (2008). Helpful self-control:
Autonomy support, vitality, and depletion. Journal of Experimental
Social Psychology, 44, 573–585. http://dx.doi.org/10.1016/j.jesp.2007
.10.008
Muraven, M., Shmueli, D., & Burkley, E. (2006). Conserving self-control
strength. Journal of Personality and Social Psychology, 91, 524–537.
http://dx.doi.org/10.1037/0022-3514.91.3.524
Nilsson, J. P., Söderström, M., Karlsson, A. U., Lekander, M., Akerstedt,
T., Lindroth, N. E., & Axelsson, J. (2005). Less effective executive
functioning after one night’s sleep deprivation. Journal of Sleep Re-
search, 14, 1–6. http://dx.doi.org/10.1111/j.1365-2869.2005.00442.x
Oaten, M., & Cheng, K. (2007). Improvements in self-control from finan-
cial monitoring. Journal of Economic Psychology, 28, 487–501. http://
dx.doi.org/10.1016/j.joep.2006.11.003
Olson, K. R. (2005). Engagement and self-control: Superordinate dimen-
sions of Big Five traits. Personality and Individual Differences, 38,
1689–1700. http://dx.doi.org/10.1016/j.paid.2004.11.003
Pinheiro, J. C., & Bates, D. M. (2000). Mixed-effects models in S and
S-Plus. New York, NY: Springer. http://dx.doi.org/10.1007/978-1-4419-
0318-1
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. http://dx.doi.org/10.1037/0021-9010.88.5.879
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools
for probing interaction effects in multiple linear regression, multilevel
modeling, and latent curve analysis. Journal of Educational and Behav-
ioral Statistics, 31, 437–448. http://dx.doi.org/10.3102/10769
986031004437
Rasbash, J. R., Steele, F. A., Browne, W. J., & Goldstein, H. (2012). A
user’s guide to MLwiN Version 2.26. Bristol, UK: University of Bristol,
Centre for Multilevel Modelling.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models (2nd
ed.). Thousand Oaks, CA: Sage.
Robinson, J. L., & Demaree, H. A. (2007). Physiological and cognitive
effects of expressive dissonance. Brain and Cognition, 63, 70–78.
http://dx.doi.org/10.1016/j.bandc.2006.08.003
Ryan, R. M., & Deci, E. L. (2008). From ego depletion to vitality: Theory
and findings concerning the facilitation of energy available to the self.
Social and Personality Psychology Compass, 2, 702–717. http://dx.doi
.org/10.1111/j.1751-9004.2008.00098.x
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
18 DIESTEL, RIVKIN, AND SCHMIDT
Schaufeli, W. B., Bakker, A. B., & Salanova, M. (2006). The measurement
of work engagement with a short questionnaire: A cross-national study.
Educational and Psychological Measurement, 66, 701–716. http://dx.doi
.org/10.1177/0013164405282471
Schmeichel, B. J. (2007). Attention control, memory updating, and emo-
tion regulation temporarily reduce the capacity for executive control.
Journal of Experimental Psychology: General, 136, 241–255. http://dx
.doi.org/10.1037/0096-3445.136.2.241
Schmeichel, B. J., Vohs, K. D., & Baumeister, R. F. (2003). Intellectual
performance and ego depletion: Role of the self in logical reasoning and
other information processing. Journal of Personality and Social Psy-
chology, 85, 33–46. http://dx.doi.org/10.1037/0022-3514.85.1.33
Schmidt, K.-H., Hupke, M., & Diestel, S. (2012). Does dispositional
capacity for self-control attenuate the relation between self-control de-
mands at work and indicators of job strain? Work & Stress, 26, 21–38.
http://dx.doi.org/10.1080/02678373.2012.660367
Scott, B., Barnes, C. M., & Wagner, D. T. (2012). Chameleonic or
consistent? A multilevel investigation of emotional labor variability and
self-monitoring. Academy of Management Journal, 55, 905–926. http://
dx.doi.org/10.5465/amj.2010.1050
Sonnentag, S., & Binnewies, C. (2013). Daily affect spillover from work to
home: Detachment from work and sleep as moderators. Journal of
Vocational Behavior, 83, 198–208. http://dx.doi.org/10.1016/j.jvb.2013
.03.008
Sonnentag, S., Binnewies, C., & Mojza, E. J. (2010). Staying well and
engaged when demands are high: The role of psychological detachment.
Journal of Applied Psychology, 95, 965–976. http://dx.doi.org/10.1037/
a0020032
Sonnentag, S., & Grant, A. M. (2012). Doing good at work feels good at
home, but not right away: When and why perceived prosocial impact
predicts positive affect. Personnel Psychology, 65, 495–530. http://dx
.doi.org/10.1111/j.1744-6570.2012.01251.x
Sonnentag, S., Mojza, E. J., Demerouti, E., & Bakker, A. B. (2012).
Reciprocal relations between recovery and work engagement: The mod-
erating role of job stressors. Journal of Applied Psychology, 97, 842–
853. http://dx.doi.org/10.1037/a0028292
Sonnentag, S., & Zijlstra, F. R. H. (2006). Job characteristics and off-job
activities as predictors of need for recovery, well-being, and fatigue.
Journal of Applied Psychology, 91, 330–350. http://dx.doi.org/10.1037/
0021-9010.91.2.330
Spector, P. E. (1994). Using self-report questionnaires in OB research: A
comment on the use of a controversial method. Journal of Organiza-
tional Behavior, 15, 385–392. http://dx.doi.org/10.1002/job.4030150503
Stetler, C., & Miller, G. E. (2005). Blunted cortisol response to awakening
in mild to moderate depression: Regulatory influences of sleep patterns
and social contacts. Journal of Abnormal Psychology, 114, 697–705.
http://dx.doi.org/10.1037/0021-843X.114.4.697
Tabibnia, G., Monterosso, J. R., Baicy, K., Aron, A. R., Poldrack, R. A.,
Chakrapani, S.,...London, E. D. (2011). Different forms of self-control
share a neurocognitive substrate. The Journal of Neuroscience, 31,
4805–4810. http://dx.doi.org/10.1523/JNEUROSCI.2859-10.2011
Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control
predicts good adjustment, less pathology, better grades, and interper-
sonal success. Journal of Personality, 72, 271–324. http://dx.doi.org/
10.1111/j.0022-3506.2004.00263.x
Tice, D. M., & Bratslavsky, E. (2000). Giving in to feel good: The place
of emotion regulation in the context of general self-control. Psycholog-
ical Inquiry, 11, 149–159. http://dx.doi.org/10.1207/S1532796
5PLI1103_03
Van de Ven, B., & Vlerick, P. (2013). Testing the triple-match principle
among technology employees. European Journal of Work and Organi-
zational Psychology, 22, 658–669. http://dx.doi.org/10.1080/1359432X
.2012.682359
Van Dongen, H. P., Vitellaro, K. M., & Dinges, D. F. (2005). Individual
differences in adult human sleep and wakefulness: Leitmotif for a
research agenda. Sleep, 28, 479–496.
van Veldhoven, M., & Broersen, S. (2003). Measurement quality and
validity of the “need for recovery scale.” Occupational and Environ-
mental Medicine, 60 (Suppl 1), i3–i9. http://dx.doi.org/10.1136/oem.60
.suppl_1.i3
Vohs, K. D., Baumeister, R. F., & Schmeichel, B. J. (2012). Motivation,
personal beliefs, and limited resources all contribute to self-control.
Journal of Experimental Social Psychology, 48, 943–947. http://dx.doi
.org/10.1016/j.jesp.2012.03.002
Wagner, D. T., Barnes, C. M., Lim, V. K. G., & Ferris, D. L. (2012). Lost
sleep and cyberloafing: Evidence from the laboratory and a daylight
saving time quasi-experiment. Journal of Applied Psychology, 97,
1068–1076. http://dx.doi.org/10.1037/a0027557
Wan, E. W., & Sternthal, B. (2008). Regulating the effects of depletion
through monitoring. Personality and Social Psychology Bulletin, 34,
32–46. http://dx.doi.org/10.1177/0146167207306756
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
dation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality and Social Psychology, 54, 1063–1070.
http://dx.doi.org/10.1037/0022-3514.54.6.1063
Xanthopoulou, D., Bakker, A. B., & Fischbach, A. (2013). Work engage-
ment among employees facing emotional demands: The role of personal
resources. Journal of Personnel Psychology, 12, 74–84. http://dx.doi
.org/10.1027/1866-5888/a000085
Zapf, D., & Holz, M. (2006). On the positive and negative effects of
emotion work in organizations. European Journal of Work and Orga-
nizational Psychology, 15, 1–28. http://dx.doi.org/10.1080/13594
320500412199
Zapf, D., Vogt, C., Seifert, C., Mertini, H., & Isic, A. (1999). Emotion
work as a source of stress: The concept and development of an instru-
ment. European Journal of Work and Organizational Psychology, 8,
371–400. http://dx.doi.org/10.1080/135943299398230
Received June 26, 2013
Revision received September 10, 2014
Accepted October 3, 2014 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
19
EMOTIONAL LABOR, SLEEP QUALITY, AND WELL-BEING
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Applied Psychology
This content is subject to copyright. Terms and conditions apply.
