Daily Micro-Breaks and Job Performance: General Work Engagement as a Cross-Level Moderator

Abstract

Despite the growing research on work recovery and its well-being outcomes, surprisingly little attention has been paid to at-work recovery and its job performance outcomes. The current study extends the work recovery literature by examining day-level relationships between prototypical microbreaks and job performance as mediated by state positive affect. Furthermore, general work engagement is tested as a cross-level moderator weakening the indirect effects of microbreaks on job performance via positive affect. Using multisource experience sampling method, the authors collected two daily surveys from 71 call center employees and obtained objective records of daily sales performance for two consecutive weeks (n = 632). Multilevel path analysis results showed that relaxation, socialization, and cognitive microbreaks were related to increased positive affect at work which, in turn, predicted greater sales performance. However, breaks for nutrition-intake (having snacks and drinks) did not show significant effects. Importantly, microbreaks had significant indirect effects on job performance via positive affect only for workers who had lower general work engagement, whereas the indirect effects did not exist for workers who had higher general work engagement. Furthermore, Bayesian multilevel analyses confirmed the results. Theoretical and practical implications, limitations, and future research directions are discussed.

Full text

©American Psychological Association. This paper is not the copy of record and may not exactly
replicate the authoritative document published in the APA journal. The final article is available
at http://dx.doi.org/10.1037/apl0000308
Citation:
Kim, S., Park, Y., & Headrick, L. (2018, March 29). Daily Micro-Breaks and Job Performance: General
Work Engagement as a Cross-Level Moderator. Journal of Applied Psychology. Advance online
publication. http://dx.doi.org/10.1037/apl0000308
Daily Micro-Breaks and Job Performance: General Work Engagement as a Cross-Level
Moderator
Sooyeol Kim
School of Labor and Employment Relations
University of Illinois at Urbana-Champaign
Champaign, IL 61820
Telephone: 703-999-9969
e-mail: sooyeolkim@gmail.com
YoungAh Park, Ph.D.
School of Labor and Employment Relations
University of Illinois at Urbana-Champaign
504 East Armory Avenue
Champaign, IL, 61820
Tel: (217) 333-1482
yapark15@illinois.edu
Lucille Headrick
School of Labor and Employment Relations
University of Illinois at Urbana-Champaign
504 East Armory Avenue
Champaign, IL, 61820
Tel: (217) 333-1482
headric2@illinois.edu
Author Note:
We acknowledge the presentation of an earlier version of this study at the 2015 Academy of
Management annual conference in Atlanta, Georgia and the publication of this study at the 2015
Academy of Management Proceeding.

Running Head: MICRO-BREAKS AND JOB PERFORMANCE
Abstract
Despite the growing research on work recovery and its well-being outcomes, surprisingly little
attention has been paid to at-work recovery and its job performance outcomes. The current study
extends the work recovery literature by examining day-level relationships between prototypical
micro-breaks and job performance as mediated by state positive affect. Furthermore, general
work engagement is tested as a cross-level moderator weakening the indirect effects of micro-
breaks on job performance via positive affect. Using multi-source experience sampling method,
the authors collected two daily surveys from 71 call center employees and obtained objective
records of daily sales performance for two consecutive weeks (n = 632). Multilevel path analysis
results showed that relaxation, socialization, and cognitive micro-breaks were related to
increased positive affect at work which, in turn, predicted greater sales performance. However,
breaks for nutrition-intake (having snacks and drinks) did not show significant effects.
Importantly, micro-breaks had significant indirect effects on job performance via positive affect
only for workers who had lower general work engagement, whereas the indirect effects did not
exist for workers who had higher general work engagement. Furthermore, Bayesian multilevel
analyses confirmed the results. Theoretical and practical implications, limitations, and future
research directions are discussed.
Keywords: micro-breaks, job performance, positive affect, recovery, resources, work
engagement

MICRO-BREAKS AND JOB PERFORMANCE
2
Daily Micro-Breaks and Job Performance: General Work Engagement as a Cross-Level
Moderator
“All work and no play makes Jack a dull boy”
−A proverb−
The popular media warns that working individuals should take breaks or risk becoming
“worn-out” or “dulled.” The Cable News Network has recently featured on several business
cases that integrate respite activities into the workplace (e.g., foosball, yoga, snack bars, naps) as
some organizations believe that fun and relaxing activities between tasks increase employee
morale and promote productivity (Rodriguez, 2013). Conversely, some organizations and
managers may consider such activities to be counterproductive and disdain breaks as ways to
“goof off” on company time. In light of the contrasting views on respite activities within the
workplace, can organizational psychology provide any insight into whether respite activities at
work are beneficial for job performance outcomes?
Researchers have studied recovery processes that allow workers to unwind and repair
negative consequences of continual work (see, Demerouti, Bakker, Geurts, & Taris, 2009, for a
review). The recovery literature recommends micro-breaks as energy management strategies for
sustaining employee resources throughout a workday (Fritz, Lam, & Spreitzer, 2011; Hunter &
Wu, 2016; Trougakos & Hideg, 2009; Zacher, Brailsford, & Parker, 2014). A recent study also
showed that on days when office workers took frequent afternoon micro-breaks, they reported
substantially reduced effects of work demands on negative affect at the end of the workday
(Kim, Park, & Niu, 2017). Despite increasing evidence that micro-breaks have salutary effects
on well-being and strain, the literature offers little evidence and insights as to whether and how
micro-breaks promote employee job performance (Trougakos & Hideg, 2009). Moreover,

MICRO-BREAKS AND JOB PERFORMANCE
3
scholarly knowledge is lacking regarding whether micro-breaks benefit job performance for all
employees or certain employees only.
In this study, we aim to better explain the relationship between at-work recovery and its
job performance outcome. Examining the link between micro-breaks and performance will
advance recovery theory and help organizations recognize potential benefits of respite activities
at work. Figure 1 illustrates our conceptual model to examine day-level relationships between
prototypical micro-breaks—relaxation, nutrition-intake, socialization, and cognitive—and job
performance outcomes, as well as determine a mechanism underlying the relationships. Another
important aim is to test general work engagement as an individual difference factor that cross-
moderates the intermediary mechanism in predicting job performance. Our theoretical model and
empirical test contribute to the recovery literature in three key ways.
First, research has shown that recovery outside work relates to the next day’s positive
work-related outcomes, such as proactive behaviors and task performance (e.g., Binnewies,
Sonnentag, Mojza, 2009; Sonnentag, 2003; ten Brummelhuis & Bakker, 2012). However, we
have scarce evidence regarding whether at-work recovery through micro-breaks generates
similar positive work outcomes. This little attention is surprising in that employees spend most
of their time at work. Also, the ultimate purpose of recovery is to promote job performance as
well as well-being (Sonnentag & Fritz, 2007). Therefore, testing job performance as an outcome
of micro-breaks is highly warranted. In doing so, we obtained objective records of daily sales
performance to remove self-report biases and minimize common method variance concerns
(Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Also, importantly, because most recovery
research assessed its outcomes with self-report measures, it remains difficult to determine
whether recovery positively influences employees’ substantive work outcomes above and

MICRO-BREAKS AND JOB PERFORMANCE
4
beyond their perceptual outcomes (Sonnentag, Venz, & Casper, 2017, for a review).
Accordingly, we studied call center employees and collected their daily sales performance
records to test whether micro-breaks increase their important resource (i.e., state positive affect)
which then translates into tangible, core performance outcomes (i.e., daily sales in call center
business). We further confirmed our results with Bayesian multilevel analyses to provide
rigorous evidence.
Second, whereas previous research has found beneficial effects of work breaks on lower
strain symptoms and higher positive affective displays (e.g., Hunter & Wu, 2016; Trougakos,
Beal, Green, & Weiss, 2008), no study has investigated within-person mechanisms by which
micro-breaks lead to job performance outcome. It is not clear which underlying psychological
experiences explain the links between break activities and performance (Trougakos & Hideg,
2009). Specifically, combining resource-based theories with empirical research on at-work
recovery, we conceptualize micro-breaks as resource-replenishing events that aid job
performance because they increase employee resource levels, such as positive affect, especially
for employees who primarily interact with customers on a daily basis. Furthermore, given that
affective states can shape job attitudes and drive important work behaviors (Beal, Weiss, Barros,
& MacDermid, 2005; Weiss & Cropanzano, 1996), it is important to test increased positive
affect as a linking mechanism. Thus, our theoretical reasoning and empirical test of the
psychological mechanism improve understandings of exact channels through which micro-
breaks lead to desirable performance results.
Third, empirical research on moderators of work break effects is rare (for an exception,
see Trougakos, Hideg, Cheng, & Beal, 2014) and virtually nonexistent for performance
outcomes. Considering the lack of knowledge about whether or not processes of micro-breaks

MICRO-BREAKS AND JOB PERFORMANCE
5
are applicable to all workers, Trougakos and Hideg (2009) and Sonnentag et al. (2017) have
called for research on moderators that potentially influence the relationship between resources
gained from breaks and job performance. As such, we propose that employees who have a stable
characteristic of high work engagement—a general tendency to experience work in active and
energetic ways across situations (Schaufeli, Bakker, & Salanova, 2006)—have less need for
frequent micro-breaks to increase their transient positive affect, possibly because they already
have a significant and stable resource reservoir. As moderation tests are necessary for enriching
theories, our study contributes to the recovery literature and its theory development.
Theoretical Foundations and Hypotheses Development
The Concept of Micro-Breaks and Theories Explaining Their Recovery Effects
Micro-break activities are short, informal respite activities taken voluntarily between
tasks (Kim et al., 2017; Trougakos & Hideg, 2009). This concept of micro-breaks is
distinguished from other institutionalized breaks, such as lunch or formally scheduled breaks. As
individuals take micro-breaks as needed amid task activities, they are less structured compared to
formally prescheduled breaks. Although there is no established standard regarding the length of
micro-breaks, they generally last anywhere from a few seconds to several minutes. Most
important, employees pursue micro-break activities at their own discretion. If breaks are taken
otherwise (e.g., unwillingly due to others’ requests or at inopportune moments), they can be
experienced as work interruptions causing distractions, backlogs, and frustrations (Jett & George,
2003). The voluntary nature of micro-breaks also aligns with the literature on off-job recovery
indicating the importance of autonomy during respites (Sonnentag & Fritz, 2007). For example,
positive relationships between evening leisure activities and next-morning recovery state were
more pronounced among employees with greater intrinsic motivation for the activities (ten

MICRO-BREAKS AND JOB PERFORMANCE
6
Brummelhuis & Trougakos, 2014). Additionally, Trougakos et al. (2014) found that employee
autonomy over how to use lunch breaks determined whether the breaks had beneficial or harmful
effects on strain. Thus, self-initiated, voluntary micro-breaks indicate that employees choose
most optimal timing and preferred activities for their momentary respites and accommodate their
idiosyncratic recovery needs and daily rhythms (Kühnel, Zacher, de Bloom, & Bledow, 2017).
Next, we briefly introduce three theories that commonly explain recovery processes in
the literature and explain how our conceptual model of micro-breaks sits within the theoretical
frameworks: Conservation of resources theory (COR; Hobfoll, 1989), ego-depletion theory
(Baumeister, Bratslavsky, Muraven, & Tice, 1998; Baumeister, Muraven, & Tice, 2000), and
effort-recovery model (Meijman & Mulder, 1998). First, COR theory assumes that individuals
have limited resources (e.g., energy) necessary to address various demands in their life (Hobfoll,
1989). Resource depletion causes strain and poor functioning, so individuals try to conserve
resources and avoid resource losses. In the workplace, employees use their personal resources to
deal with work demands and job stressors and, as such, their resources become depleted over the
course of a workday. This resource perspective suggests that employees cannot continue work
efforts indefinitely throughout the workday but they need to replenish drained resources. In that
sense, employees can take micro-breaks and engage in their choice of respite activities to restore
depleted resources.
Similarly, ego-depletion theory is based on the resource scarcity perspective: a central,
psychological resource (ego) determines individuals’ self-regulation capacity, but it is finite and
drains easily if used continuously (Baumeister et al., 1998, 2000). Self-regulatory resources are
particularly important for service employees as emotion regulation is essential for their quality
service work (Grandey, 2000). For example, call center employees often must suppress negative

MICRO-BREAKS AND JOB PERFORMANCE
7
emotions and display positive emotions if they are to meet their customer service needs and sales
goals (Grandey, Dickter, & Sin, 2004). Thus, call center work is commonly considered a high
self-control context in which self-regulatory resource depletion can make it difficult to control
response tendencies and amplify positive emotions while concentrating on tasks at hand (e.g.,
following call scripts, answering customer questions, searching for best customer options); as a
result, job performance may deteriorate. Nevertheless, ego-depletion theory suggests that self-
regulatory resources can be renewed after sufficient rest (Muraven & Baumeister, 2000). In
addition, studies based on this theory points to positive affect as an important central resource
that counteracts ego depletion and facilitates behavioral performance. For example, Tice,
Baumeister, Shmueli, and Muraven (2007) found that self-regulatory task performance improved
when positive affect was induced by experimental conditions, such as watching a comedy video
which is similar to micro-break activity. A field study applying the theory also found that respite
activities enhance positive affect felt and displayed at work that is needed for performing service
tasks (i.e., cheerleading instruction; Trougakos et al., 2008).
Effort-recovery model (Meijman & Mulder, 1998) further highlights that taking timely
respites is important to recover diminished self-regulatory resources and even generate resource
surpluses. The model posits that individuals should rest momentarily to allow their functional
systems (e.g., emotional, cognitive) to recuperate from accumulated load reactions of continuous
working, such as fatigue; however, when respite opportunities are delayed, the load reactions
persist to the extent that it becomes more difficult to return to one’s baseline functioning. In this
regard, self-initiated micro-breaks can provide the necessary disengagement from work just in
time of needs. Accordingly, work breaks were found to be related to increased well-being and

MICRO-BREAKS AND JOB PERFORMANCE
8
decreased strain (Hunter & Wu, 2016; Kim et al., 2017; Kühnel et al., 2017; Trougakos et al.,
2008; Zacher et al., 2014).
Building on these theoretical frameworks and empirical findings, we contend that micro-
break activities will be positively linked to job performance through increased positive affect as a
resource. We test daily fluctuating positive affect—feeling active, confident, interested,
concentrating, enthusiastic, and happy—as a linking mechanism in that positive affect is an
integral part of self-regulatory resources and task accomplishment for emotional laborers (Beal et
al., 2005). In particular, we investigate four prototypical micro-breaks (i.e., relaxation, nutrition-
intake, socialization, and cognitive activities), considering that recovery effects may hinge on the
exact nature of the activities employees engage in during breaks. In the next section, we first
hypothesize how each category of micro-breaks increases positive affect. Then, drawing on the
episodic model of affective influences on job performance by Beal et al. (2005), we theorize the
intermediary role of positive affect in connecting micro-breaks and job performance. Last, we
turn to general work engagement as a moderator in the indirect effect of micro-breaks on job
performance.
Relaxation Activities
Relaxation is to momentarily relieve psychological and physical tension from continuous
work and further prevent its short-term accumulations throughout a workday. Common
relaxation activities include taking a short nap or walk, meditating, daydreaming, and stretching,
all of which are characterized by low effort or effortless activities (e.g., Sianoja, Syrek, de
Bloom, Korpela, & Kinnunen, 2017; Trougakos & Hideg, 2009). Effort-recovery model explains
that relaxation activities help individuals’ physical and psychological systems return to pre-stress
levels (Meijman & Mulder, 1998). As such, relaxation is considered one of the core recovery

MICRO-BREAKS AND JOB PERFORMANCE
9
experiences for retreating physically and psychologically to restore depleted resources
(Sonnentag & Fritz, 2007). Likewise, previous studies found that short relaxation at work (e.g.,
stretching, napping) is associated with lower physical and mental fatigue and higher positive
emotions (Henning, Jacques, Kissel, Sullivan, & Alteras-Webb, 1997; Trougakos et al., 2008).
Also, on days when employees had more relaxation activities, they reported much lower impacts
of work demands on affective distress (Kim et al., 2017) and higher vitality, such as feeling
energetic and cheerful, at work (Zacher et al., 2014). Furthermore, an experimental study found
that individuals who appraised their micro-breaks as relaxing reported more feelings of vigor
(Bennett, 2015). Moreover, short relaxation activities (i.e., park walks, soundscape of nature,
relaxation exercises) not only reduced employees’ daily fatigue but also led to more positive
states, such as feelings of vigor, enjoyment, and concentration (Sianoja et al., 2017; Steidle,
Gonzalez-Morales, Hoppe, Michel, & O’shea, 2017). In summary, these combined findings
suggest that relaxation micro-breaks provide optimal conditions for releasing any negative load
reactions of continuous working and resource recovery, thereby increasing positive and pleasant
states for the next task episodes.
Hypothesis 1a: Daily relaxation micro-breaks will be positively related to increased
positive affect at work.
Nutrition-Intake Activities
Nutrition-intake activities refer to snacking and drinking at work. Human physiology runs
on nutrients like a fuel (e.g., water, minerals, protein, glucose), so getting adequate supplies of
nutrients is essential for human energy and daily functioning (Renner, Sproesser, Strohbach, &
Schupp, 2012). For example, glucose (blood sugar) is one of the major sources of nutritional
energy and, as such, individuals who have higher glucose levels tend to show less negative

MICRO-BREAKS AND JOB PERFORMANCE
10
emotions and more helping behaviors because of high self-regulatory resources (Gailliot et al.,
2007). In addition, some nonessential nutrients from foods and beverages can influence
emotional and mental states. For example, caffeine is a mild stimulant boosting feelings of
alertness, activeness, and energy (Hausser, Schlemmer, Kaiser, Kalis, & Mojzisch, 2014). Recent
studies have discovered beneficial effects of caffeine intake, given the common coffee and tea
breaks at work. Specifically, caffeine consumption attenuated sleep deprivation effects on
feelings of energy depletion (Welsh, Ellis, Christian, & Mai, 2014) and minimized adverse
effects of daily work demands on end-of-work negative affect (Kim et al., 2017). Although most
employees take regular lunches, they may still feel hungry or thirsty as they continuously use
nutritional energy to exert self-control efforts for job tasks throughout the work day. Indeed,
employees tend to eat more snacks on days when they want to reduce feelings of frustration and
fatigue (i.e., high affect-regulation motive) and thereby boost their energy at work (Sonnentag,
Pundt, & Venz, 2016). In short, brief breaks for snacks and beverages between tasks may help
employees avoid discomforting physiological experiences and further increase mental alertness
and energy. Thus, we expect that nutrition-intake breaks will increase positive affect at work
(e.g., feeling active and concentrating, etc.).
Hypothesis 1b: Daily nutrition-intake micro-breaks will be positively related to increased
positive affect at work.
Socialization Activities
Off-job recovery research suggests that one way to assist recovery is through social
activities in which employees interact with others, garner social support, and psychologically
detach from work-related thoughts (Sonnentag, 2001; ten Brummelhuis & Bakker, 2012). In this
study, we contend that nonwork-related social contacts and interactions at work can boost

MICRO-BREAKS AND JOB PERFORMANCE
11
employees’ positive affect. The concept of relational energy (Owens, Baker, Sumpter, &
Cameron, 2016) explains that interpersonal interactions at work give employees a heightened
level of psychological resourcefulness that can enhance their capacity to do work. That is,
employees can be energized by and even proactively seek social interactions to increase their
energy at work. Moreover, employees who draw more relational energy from social interactions
tend to be highly engaged in their job and more productive (Owens et al., 2016). Trougakos et al.
(2014) also found that daily lunch break socialization reduced end-of-work fatigue when
employees had high autonomy for their lunch break (i.e., deciding what they want to do during
the break). Similarly, Kim et al. (2017) demonstrated that voluntary, nonwork social activities
weakened the effects of daily work demands on subsequent negative affect at work. Moreover,
Zacher et al. (2014) showed that daily micro-breaks including social contacts and interactions
(i.e., talking about common interests like sports or hobbies, checking in with a friend and family
member) increased feelings of vitality at work, whereas work-related break activities (e.g.,
seeking feedback) failed to do so. Taken together, we argue that social interactions and contacts
should be free from work-related aspects to ensure optimal recovery effects of micro-breaks.
Thus, we expect that voluntary, nonwork-related social activities during micro-breaks will
increase positive affect.
Hypothesis 1c: Daily socialization micro-breaks will be positively related to increased
positive affect at work.
Cognitive Activities
Cognitive micro-break activities refer to any activities that facilitate a mental break from
work demands although they may still require some cognitive attention and effort. Examples
include casual reading and browsing the Internet for entertainment or personal learning. Most

MICRO-BREAKS AND JOB PERFORMANCE
12
important is that cognitive distraction momentarily shifts employees’ attention from high self-
control demands toward chosen activities for entertainment or casual learning. The off-job
recovery literature has suggested that psychological detachment from work is critical for
replenishing energy and affective resources (see Sonnentag & Fritz, 2015, for a review). In
addition, doing preferred activities during breaks increases personal resources at work such as
energy, motivation, and concentration (Hunter & Wu, 2016). However, contrary to the
hypothesized recovery effects of cognitive activities at work, Kim et al. (2017) found that
voluntarily chosen cognitive breaks aggravated employee distress. As a post-hoc explanation,
they attributed the unexpected finding to their failure in distinguishing pleasurable activities from
personal chores and duties (e.g., online banking). Accordingly, we expect that pleasurable or
personally meaningful cognitive breaks will provide temporary escape from work demands and
momentarily improve affective state, as reflected in our cognitive activity items. Thus, we
hypothesize the following.
Hypothesis 1d: Daily cognitive micro-breaks will be positively related to increased
positive affect at work.
Indirect Effects of Micro-Break Activities on Job Performance via Positive Affective
Resource
Broaden-and-build theory (Fredrickson, 1998, 2001) posits that individuals’ positive
emotions from positive episodes can widen momentary thought−action repertoires, thereby
increasing performance in various contexts. That is, momentary positive experiences generate
positive affect which, then, leads individuals to seek underexplored paths of thoughts and actions
rather than typical, automatic behavioral options. This proposition is congruent with the episodic
model of affective influences on performance (Beal et al., 2005) as the model theorizes that

MICRO-BREAKS AND JOB PERFORMANCE
13
affective state influences cognition and behavioral styles that are conducive to effective task
achievement. In other words, the model posits that affective states can not only influence one’s
attentional regulation for task conducts but also directly impact on task approaches and
momentary response tendencies in doing job tasks (Beal et al., 2005). This combined theoretical
perspective suggests that positive affect accruing from micro breaks will influence job
performance.
More specifically, in the context of call center jobs, we argue that positive affective state
from micro-breaks may lead to better sales performance for the following reasons. First, call
center employees interact daily with multiple customers. Their task episodes tend to begin and
end with each call they make. In emotionally draining jobs, especially when employees faced
with difficulties and frustrations, they tend to perceive more difficulty in displaying positive
emotions because of greater emotional dissonance between their true feelings and outward
emotional expressions (Beal, Trougakos, Weiss, & Green, 2006). However, shortly after taking
micro-breaks, the resultant positive affect may help employees reappraise and reframe task
situations more confidently and positively, thereby increasing work motivation. Indeed, lab
experiments showed that positive affect improved task performance through enhanced beliefs
that task effort would result in good performance (Erez & Isen, 2002). Similarly, a field study
using insurance sales agents found that positive affect predicted task performance by boosting
motivational components such as task self-efficacy and persistency (Tsai, Chen, & Liu, 2007).
Thus, we expect that positive affect gained from between-task micro-breaks (e.g., feeling
confident, enthusiastic, active, etc.) will shape more positive work attitudes and increase task
motivation momentarily, facilitating job performance.

MICRO-BREAKS AND JOB PERFORMANCE
14
Second, the performance model of affective influences (Beal et al., 2005) views that
performance during an episode is a joint function of resource level and resource allocation. That
is, job performance is a result of a dynamic process in which individuals control and allocate
various resources to task activities. The model further suggests that state positive affect may
allow employees to better address emotional labor while simultaneously staying focused on
cognitive processing. For example, even during difficult calls, employees may use their state
positive affect to please dissatisfied customers while cognitively searching for the best services
and products. That is, employees in positive affective state may easily alter emotion-regulation
strategies without being distracted from finding the best customer options. In other words, this
example indicates that positive affect helps employees allocate attentional resources to
information-processing as well as address emotional tasks during the sales calls. Accordingly,
when call center workers were in a pleasant and positive affective state, they solved customer
problems more efficiently during calls (Miner & Glomb, 2010) and showed more verbal fluency
in unscripted interactions with customers (“no dead air time”), an important call-quality metric
(Rothbard & Wilk, 2011). Also, bank tellers’ positive emotion displays were associated with
customers’ positive perception of received services (Pugh, 2001). Likewise, positive emotion in
sales clerks was found to relate to their customers’ actual product purchase (Tsai, 2001).
Taken all together, we argue that positive affect derived from micro-breaks will function
as a self-regulatory resource allowing employees to maintain their work motivation, better
handle difficult calls, and display positive emotions in order to meet their service and sales goals.
Hypothesis 2: There will be indirect relationships between micro-breaks—(a) relaxation,
(b) nutrition-intake, (c) socialization, and (d) cognitive activities—and job performance
through increased positive affect, such that on days when employees take more micro-

MICRO-BREAKS AND JOB PERFORMANCE
15
breaks, they are more likely to gain positive affect and thereby achieve higher
performance.
General Work-Engagement as a Cross-Level Moderator
Work engagement is defined as one’s inclination to view work in a positive, fulfilling
way that is characterized by vigor, dedication, and absorption (Schaufeli et al., 2006). The
concept essentially captures the extent that individuals experience their work as stimulating and
energetic (vigor), meaningful and significant (dedication), and interesting and captivating
(absorption). Although work engagement levels can fluctuate daily, some individuals are more
engaged than others across situations, a tendency known as general work engagement (Breevaart,
Bakker, Demerouti, & Hetland, 2012)—hereinafter work engagement in short. This construct is
about dedicating the “full self” and being highly motivated for work, and thus differs from other
traditional job attitudes such as job satisfaction and organizational commitment (see, Christian,
Garza, & Slaughter, 2011, for a review).
A corollary in COR theory (Hobfoll, 2001) is that: “those with greater resources are less
vulnerable to resource loss and more capable of orchestrating resource gain. Conversely, those
with fewer resources are more vulnerable to resource loss and less capable to resource gain” (pp.
349). Although work engagement has not been examined as a cross-level moderator of the
affect−performance link within individuals, the past literature suggests that work engagement is
an important motivational resource for employee performance. In a cross-sectional study of hotel
workers and their customers (Salanova, Agut, & Peiro, 2005), employees’ general work
engagement was related to higher service climate which was, in turn, linked to employee
performance and then customer loyalty. Also, employees’ work engagement predicted
subsequent job performance as assessed by their supervisors and coworkers (Halbesleben &

MICRO-BREAKS AND JOB PERFORMANCE
16
Wheeler, 2008). A meta-analytic study also confirmed that work engagement explains
incremental variance in task performance above and beyond other job attitudes (Christian, Garza,
& Slaughter, 2011). Thus, work engagement represents a relatively stable resource that may
reduce one’s sensitivity to the effects of fluctuating positive affect on job performance.
More specifically, task performance during an episode is a joint function of resource level
and resource allocation (Beal et al., 2005). As such, employees with high work engagement may
be able to offset temporary losses of affective resources by drawing from larger resource
reservoirs. That is, even with dwindling transient affective resources, they still remain motivated
to pursue persistent task efforts and effectively allocate other necessary resources. In addition,
highly engaged employees tend to go beyond their roles to help achieve goals of coworkers and
their organization, further suggesting that highly engaged individuals are better able to “free up”
their personal resources (Christian et al., 2011). Moreover, general work engagement is driven
by both job resources (e.g., social support, feedback) and personal resources (e.g., resiliency,
positive self-evaluations) (Bakker, 2011; Christian et al., 2011). This may indicate that
individuals with high work engagement have a “caravan of resources” for sustaining work
motivation and enhancing performance; therefore, they are less affected by transient, volatile
resources (Hobfoll, 2001). Translating the conceptual and theoretical reasoning into the current
context, we hypothesize that positive affect gained from micro-breaks will have a weaker
relation with job performance especially for those who have high work engagement but will have
a stronger relation for those who have low work engagement.
Hypothesis 3: Work engagement at the between-person level will moderate (reduce) the
day-level relationship between positive affect and sales performance, such that

MICRO-BREAKS AND JOB PERFORMANCE
17
employees with high (low) work engagement will be less (more) likely to be influenced
by positive affect during the workday.
Further, our combined hypotheses (H1−H3) imply that the strength of the indirect
relationships between micro-breaks and performance outcomes via positive affect may differ by
the levels of work engagement. That is, high work engagement makes employees less sensitive
and vulnerable to state positive affect from micro-breaks for their performance benefits, such that
the indirect effects of micro-breaks on job performance will be weaker for employees with high
work engagement but will be stronger for those with low work engagement.
Hypothesis 4: The day-level indirect effects of (a) relaxation, (b) nutrition-intake, (c)
socialization, and (d) cognitive micro-breaks on performance outcomes via positive affect
will be weaker (stronger) for employees who have higher (lower) levels of work
engagement.
Method
Sample and Procedure
The current study is the first publication based on data collected as part of larger study
entitled, “Employees’ Work and Nonwork Experiences in Telemarketing Job.” IRB approval
was granted by Kansas State University for protocol number 7120. Participants included 71
telemarketers at call centers in Korea. Their primary job is to call customers and persuade them
to purchase services or products (O*Net, 2015). The participants sold insurance, credit cards,
cable and Internet, and landline and mobile phone services. Consequently, they often engage in
emotional labor: they must suppress their true feelings, display positive emotions, answer
customers’ questions, and solicit sales (Grandey, 2003). Industries represented were finance
(49%) or telecommunications (51%). The two groups showed no significant differences

MICRO-BREAKS AND JOB PERFORMANCE
18
regarding age, job tenure, weekly work hours, and work engagement (p-values = .29 − .82). Most
participants were women (79%), reflecting the female-dominated call center field in Korea
(Korea Labor and Information Service, 2012). The final sample averaged 37.03 years-old (SD =
7.58), organizational tenure of 4.15 years (SD = 3.19), and 39.45 hours worked per week (SD =
6.42).
Participants were recruited through several online community websites for call center
workers in Korea. With permission from the websites’ administrators, the recruitment message
was posted for about 4 weeks, explaining the two-phase online study procedure (an initial survey
and two daily surveys for 10 workdays), compensation for participation (a $30 value mobile gift
card), and eligibility for participation. To participate, they had to work full-time at call centers
with a regular work schedule (9 a.m. to 6 p.m.) in accordance with regular working hours in
Korea, and fixed lunch hours (no shift and telecommuting workers). The initial survey link was
distributed to 165 participants who expressed interest via email. Completion of the initial survey
indicated their consent to participate in daily surveys and provide their objective performance
report retrieved from their organizations’ sales database (i.e., daily sales performance records
during the participation period). Their organizations used archival sales records to calculate
incentive payments so employees were allowed to track and retrieve their sales records freely on
their work computer1. A total of 105 out of 165 individuals completed the initial survey (64%),
reporting demographic information and their work engagement. They also self-generated ID
codes and used throughout the study to allow us to match responses across measurement
occasions.
1 Before this study, we conducted a pilot focus group interview with call center workers in Korea. We learned that
they would be able to retrieve their individual sales performance report from their work computer which was
connected to their company’s sales management system.

MICRO-BREAKS AND JOB PERFORMANCE
19
About one week after the initial survey, they started answering two short daily surveys
for 10 consecutive workdays: one in the morning (Time 1) to assess their morning positive affect
(control variable) and another at the end of each workday before leaving their call center (Time
2) to measure their day-level workload (control variable), engagement in micro-break activities,
and positive affect at work. As expected in daily diary studies, 34 participants (n = 7) skipped
some of the daily surveys (e.g., completing only morning or end-of-work survey), or failed to
complete any daily surveys after the first phase (n = 27). They were removed, leaving 71 of 105
individuals (68%). No significant differences were observed between the initial sample and the
final sample regarding age, sex, education level, organizational and job tenure, average work
hours per week, and work engagement (p-values = .22 − .73). On average, they completed the
morning survey at 9:22 a.m. (SD = 0.34), and the end-of-work survey at 6:21 p.m. (SD = 0.78).
In addition, each Friday, participants emailed us the week’s sales report along with their ID code.
The sales reports included successful sales transactions and amount of gross sales in Korean
Won for each day during the entire week. By matching these performance records with daily
survey responses, we obtained 632 day-level data points of 710 points possible (71 participants x
10 workdays), yielding a response rate of 89%.
Day-Level Measures
The measures were provided in Korean. To ensure accurate translation of English-based
measures, we used a translation-back translation procedure with two independent bilingual
translators (Brislin, 1970). All scales were slightly adapted to suit the daily measurement context,
except for the micro-break activities measure.
Micro-break activities (Time 2). Table 1 displays the nine items of micro-break
activities and their categories. We measured micro-break activities by slightly adapting the items

MICRO-BREAKS AND JOB PERFORMANCE
20
of common respite activities that Kim et al. (2017) developed and fully tested according to the
literature on micro-breaks at work (Fritz et al., 2011; Trougakos & Hideg, 2009). Specifically,
we added a small phrase (“for entertainment”) at the end of their cognitive activity items to
clearly indicate the pleasurable nature of the activities. This adapted measure asked respondents
to recall their short, informal respites taken voluntarily during their work day and then rate how
often they engaged in those break activities (1 = never to 5 = very frequently): two items each for
relaxation, nutrition-intake, and cognitive activities, and three items for social activities. We then
computed an average score within each of the categories to produce four micro-break variables.
We did not calculate the coefficient alphas as these activities are not caused by a latent construct,
and each category of break activities is defined by the combination of its measures (i.e.,
formative measure; Diamantopoulos & Siguaw, 2006).
Positive affect (Time 2). Positive affect was assessed with positive emotion scales from
the Positive and Negative Affective Schedule (PANAS; Watson, Clark, & Tellegen, 1988). To
minimize participants’ burden, we used six shortened descriptors (i.e., happy, enthusiastic,
active, concentrating, confident, interested) as common in diary studies. Participants indicated
how extensively they had felt the positive emotions during their workday (1 = none to 5 = to a
great extent). The average Cronbach’s alpha across observations was .91.
Job performance. To operationalize job performance, we used daily sales performance
records in Korean Won which indicate gross sales for each work day. Considering that the call
center employees’ gross sales greatly varied across the industries due to the different unit prices
of the products and services they sold, we standardized daily sales performance scores, using the
individuals’ respective average gross sales of the two weeks during the study period2. Therefore,
2 We also used individuals’ average gross sales for the month to calculate standardized performance scores but the
significance of our results did not differ.

MICRO-BREAKS AND JOB PERFORMANCE
21
negative values represent below-average performance in two working weeks, while positive
values indicate above-average performance.
Control variables. We measured morning positive affect as a control variable (Time 1)
because baseline affective state could influence subsequent affect and work approaches during
the workday. We used the same PANAS affective descriptors and response options to ask
participants to rate how extensively they felt the emotions in the morning. The average
Cronbach’s alpha across observations was .90. In addition, we measured day-specific workload
as another control variable because it could influence affective state at work (Ilies et al., 2007)
and opportunities to interact with customers on the phone. This was assessed at the end of the
workday with a short three-item Quantitative Workload Inventory (Spector & Jex, 1998), on a 5-
point rating format (1 = strongly disagree to 5 = strongly agree). Example item was “Today, I
had a lot of work to do.” The average Cronbach’s alpha across observations was .89.
Person-Level Measure
General work engagement. This was assessed in the initial survey with the short nine-
item Utrecht Work Engagement Scale (Schaufeli et al., 2006) on a 5-point rating format (1 =
strongly disagree to 5 = strongly agree). Participants rated how well each item described them in
general at work. Example items included: “At my work, I feel bursting with energy in general”
and “I am enthusiastic about my job, in general.” The Cronbach’s alpha was .89.
Analytic Approach
We used multilevel path analysis to test the hypothesized model shown in Figure 1 with
Mplus 6.12 (Muthén & Muthén, 2007). With this approach, we were able to accommodate the
multilevel structure of the data (i.e., daily responses nested in individuals) and simultaneously
estimate the path coefficients for the hypothesized relationships. In our model, as control

MICRO-BREAKS AND JOB PERFORMANCE
22
variables, morning positive affect and workload were specified to have fixed effects on positive
affect (Time 2) as well as on sales performance. We specified the Level-1 (i.e., intraindividual
level) fixed effects of four micro-break activities on positive affect, and the Level-1 random
effect of positive affect on sales performance. In addition, we specified the direct fixed effects of
the micro-break activities on sales performance. To facilitate the interpretation of the findings,
all Level-1 predictors (i.e., workload, morning positive affect, micro-break activities) were
person-mean centered to obtain unbiased estimates of the intraindividual-level relationships
(Enders & Tofighi, 2007). The Level-2 variable, work engagement, was grand-mean centered.
Variance partitioning showed that within-person fluctuations explained a significant amount of
the variance in the mediator and outcome variable: 84.1% for positive affect during the workday
and 73.4% for sales performance. Also, substantial variability was due to within-person
fluctuations in micro-breaks (75.3% for relaxation, 74% for nutrition-intake, 86.7% for
socialization, and 55.1% for cognitive activities). Thus, the multilevel-modeling approach was
appropriate to test the hypotheses. Mediation hypotheses were tested via Monte Carlo simulation
procedures using the open-source software R, found at
http://www.quantpsy.org/medmc/medmc111.htm (Bauer, Preacher, & Gil, 2006; Preacher &
Selig, 2010).
Furthermore, we ran Bayesian multilevel analysis to confirm our findings and respond to
recent calls to deal with traditional null hypotheses significance testing issues (Andraszewicz et
al., 2015; McKee & Miller, 2015). Critics have objected to the traditional significance testing
because p-values and statistical significance can be misinterpreted, and model comparison
processes are lacking in regression analyses (e.g., Johnson, 2013; Sellke, Bayarri, & Berger,
2001). Bayesian analysis is an alternative to overcome these issues because it depends less on p-

MICRO-BREAKS AND JOB PERFORMANCE
23
values and significance and provides more flexibility for modeling complex error structures
(Andraszewicz et al., 2015; Gelman, 2003).
Results
Preliminary Analysis
Table 2 presents means, standard deviations, and intercorrelations of the study variables.
As expected, at the within-person level, the four micro-breaks were positively related with
positive affect at Time 2 (rs = .17 to .24, ps < .001). Positive affect at Time 2 was also positively
related with sales performance (r = .29, p < .001). Four types of micro-breaks were also
moderately correlated with each other (rs = .19 to .57, ps < .01), except the relationship between
relaxation and cognitive activities (r = .08, p = .055) and relaxation and nutrition-intake activities
(r = -.02, p = .664). In addition, we checked whether there were any time trends in our mediator
and outcome variables because they were measured repeatedly over two weeks. We entered the
linear time trend variable in each regression model for positive affect and performance outcomes
but found no significant linear time trends in positive affect (γ = .01, p = .999) and sales
performance (γ = .001, p = .999).
Hypothesis Testing
Table 3 presents the results from the multilevel path analysis that estimated all the path
coefficients simultaneously. Relaxation (γ = .13, p < .001), social (γ = .14, p < .001), and
cognitive micro-break activities (γ = .13, p < .01) were positively related to increased positive
affect when morning baseline affect and workload were controlled for; thus, H1a, H1c, and H1d
were supported, respectively. However, the effect of nutrition-intake activities on positive affect
was not significant (γ = .04, p = .333), failing to support H1b. In Table 3, positive affect was

MICRO-BREAKS AND JOB PERFORMANCE
24
positively related to sales job performance (γ = .41, p < .001), after controlling for morning
affect and workload.
Increased positive affect was further hypothesized to connect the relationship between
employees’ break activities and job performance (H2a – H2d). Testing a series of indirect effects
based on 20,000 Monte Carlo replications showed that the indirect effect of relaxation activities
on sales performance via positive affect was .05 with 95% bias-corrected bootstrap confidence
interval (CI) from 0.02 to 0.09. As the CI did not include zero, H2a was supported. However,
nutrition-intake activities did not show a significant indirect effect on sales performance as the
CI included zero (95% CI [-0.03, 0.07]). Thus, H2b was not supported. The indirect effect of
social activities on sales performance via positive affect was .06 (95% CI [0.02, 0.10]),
supporting H2c. Cognitive activities indirectly affected sales performance via positive affect (.05,
95% CI [0.02, 0.08]), so H2d was supported. Note that there were no significant direct
relationships between the micro-breaks and sales performance. Nonetheless, the effects of
relaxation, socialization, and cognitive micro-breaks on performance were fully mediated by
positive affect.
We tested a cross-level moderation effect of work engagement on the within-person
relationship between positive affect and sales performance. In Table 3, results showed that work
engagement was negatively associated with the random slope between positive affect and sales
amount (γ = -.15, p < .05). Following Preacher, Curran, and Bauer (2006), we conducted simple
slope tests in multi-level modeling to confirm the nature of the interaction effects. As Figure 2
shows, the interaction pattern was observed for sales performance outcome: the positive within-
person link between positive affect and sales performance existed only for those who had low

MICRO-BREAKS AND JOB PERFORMANCE
25
levels of work engagement (γ = .61, SE = .07, p < .001) but not for those who had high levels of
work engagement (γ = .31, SE = .16, p = .06). Thus, H3 was supported.
Last, we tested whether estimated indirect effects of micro-breaks on sales performance
via positive affect differed at the lower (-1 SD) and higher (+1 SD) level of work engagement.
Relaxation activities had a significant indirect effect of .17 (p < .05) under the low work
engagement level, versus .01 (p = .743) under the high work engagement level. Moreover, the
indirect effects were significantly different between the two conditions: -.15 (95% CI [-.302,
-.002]). We did not test the moderated indirect effect of nutrition-intake activities as H1b and
H2b were not supported. Social activities had a significant indirect effect of .16 (p < .001) under
the low work engagement level, versus .01 (p = .801) under the high work engagement level. The
two conditions had significantly different indirect effects: -.15 (95% CI [-.276, -.015]). Last,
cognitive activities also had a significant indirect effect of .18 (p < .01) under the low work
engagement level but not under the high work engagement level (.04, p = .460). The indirect
effects between the two conditions were significantly different: -.14 (95% CI [-.273, -.014]).
Therefore, work engagement significantly moderated the indirect relationships between the three
micro-breaks and job performance via positive affect, supporting H4a, H4c, and H4d,
respectively. Following suggestions by Preacher and Kelley (2011), we calculated kappa-
squared (P
2), the proportion of the maximum possible indirect effects, to evaluate effect sizes of
indirect effects. The P
2 values were .04, .043, and .037, respectively for relaxation, social, and
cognitive activities’ indirect effect on performance through positive affect. The range of these
effect sizes are between small (.01) and medium (.09) size (Cohen, 1988; Preacher & Kelley,
2011). However, the P
2 value for nutrition-intake’s indirect effect was .009 in accordance with
its nonsignificant indirect effect.

MICRO-BREAKS AND JOB PERFORMANCE
26
Bayesian Multilevel Analyses to Confirm the Results
For Bayesian multilevel analysis, we used uninformative (objective) prior information
(Kruschke, Aguinis, & Joo, 2012) with the current data to produce a posterior distribution of
parameter estimates. For the Bayesian inference, we used Markov Chain Monte Carlo simulation
using the open-source software R. This method iteratively draw samples from a set of
conditional distributions to create the posterior density of interest (Martin, Quinn, & Park, 2011).
Following Yuan and MacKinnon (2009), we ran the analysis with two chains iterating 10,000
times with 500 burn-in iterations, and both chains converged for all estimated parameters. Table
4 details the results for indirect effects of the four micro-breaks on sales performance via positive
affect. Specifically, relaxation activities had an indirect effect of .15 (95% CI [0.08, 0.21]) for
sales job performance. Socialization activities had an indirect effect of .12 (95% CI [0.07, 0.20])
while cognitive activities had an indirect effect of .13 (95% CI [0.05, 0.21]). However, nutrition-
intake activities did not show a significant indirect effect on job performance as shown in the
previous analyses. General work engagement was also confirmed to moderate the link between
positive affect and job performance: -.15 (95% CI [-0.24, -0.08]). In addition, our Bayesian
multilevel analysis results found the moderated indirect effect of relaxation (-.02, 95% CI [-0.03,
-0.01]), social (-.02, 95% CI [-0.04, -0.01]), and cognitive (-.02, 95% CI [-0.03, -0.01]) activities
on sales performance through positive affect. However, the moderated indirect effect of
nutrition-intake activities was not supported (.001, 95% CI [-0.02, 0.01]). These Bayesian results
indicate that the indirect effects of relaxation, social, and cognitive activities will be reduced
by .02 when the score of general work engagement increases by one unit. Thus, the results from
Bayesian analyses indicated robustness in our initial results.
Supplementary Analyses

MICRO-BREAKS AND JOB PERFORMANCE
27
While a priori theory of affective influences on job performance (Beal et al., 2005) and
its empirical literature suggest a path direction from state affect to job performance (Shockley,
Ispas, Rossi, & Levine, 2012), it may be possible that preceding episodes of successful sales
increase subsequent positive affect during the workday. Thus, we further explored our partial
data given that 85% of our final sample (n = 60 out of 71 participants) provided time information
of their sales transactions, so we were able to separate their sales performance in the morning
(before lunch hour) and afternoon (after lunch hour). This allowed us to test a possible role of
morning performance in increasing positive affect during the workday (536 day-level data points
out of a possible total of 600). The multilevel path analysis showed that morning sales
performance did not predict increased positive affect during the workday (γ = .12, p = .221),
controlling for morning positive affect (γ = 0.58, p = .255) and day-specific workload (γ = -.075,
p = .314). Also, without the control variables, the path from morning performance to positive
affect still remained nonsignificant.
Discussion
The purpose of this study was to determine whether and how micro-breaks are positively
linked to daily job performance outcomes within call center employees. Another important goal
was to test whether general work engagement moderates the daily affective process of micro-
breaks involving job performance. Our results from both traditional and Bayesian multilevel
analyses show that relaxation, socialization, and cognitive activities were positively related to
increased positive affect leading to greater sales performance, after controlling for daily
workload and baseline morning affect. We also find that positive affect fully mediates the day-
level relationships between the three micro-breaks and sales performance. Micro-breaks for
nutrition-intake, however, do not predict positive affect and job performance. Furthermore,

MICRO-BREAKS AND JOB PERFORMANCE
28
micro-breaks have varying indirect effects depending on the levels of work engagement between
persons. Specifically, for employees who have high work engagement, micro-breaks have no
significant day-level indirect effects on sales performance via positive affect, but employees who
have low work engagement show significant indirect effects.
Theoretical Implications
Our findings offer important implications to the work recovery literature, which has
predominantly investigated off-job recovery experiences and their positive outcomes, but has
underexplored on-the-job recovery phenomena and work-relevant outcomes (Trougakos &
Hideg, 2009). Considering that employees spend a significant amount of time in their work
places, we examined micro-breaks as a way to recover momentary affective resources for call
center employees’ job performance. In particular, to maximize more prescriptive knowledge
about on-the-job recovery, we considered the specific content of the breaks. That is, when
employees voluntarily engage in respite activities that facilitate relaxation and socialization, they
can relieve themselves from work demands and increase their affective resources. Also,
importantly, our results showed that cognitive activities can still boost positive affect when the
activities had personal entertainment and learning purposes. Thus, our findings support the
theorized recovery effect of specific micro-breaks as providing important resources (Baumeister
et al., 1998; Hobfoll, 1989; Meijman & Mulder, 1998).
On the other hand, nutrition-intake activities did not have a significant indirect effect on
job performance via positive affect. As a post-hoc explanation, it may be plausible that
employees enjoy snacks and coffee while interacting with others or watching a fun video
footage. Indeed, the within-person correlations showed that on days when individuals had more
nutrition-intake activities, they were also likely to have socialization (r = .32, p < .001) and

MICRO-BREAKS AND JOB PERFORMANCE
29
cognitive activities at work (r = .57, p < .001). In addition, when we ran another path model
including nutrition-intake breaks as a sole predictor, nutrition-intake significantly increased
positive affect (γ = .14, p < .001) which, in turn, led to higher sales performance (γ = .50, p
< .001). Furthermore, considering the short-term effects of caffeine in boosting one’s energy (cf.
Kim et al., 2017; Welsh et al., 2014), we differentiated consumption of caffeine vs. snacks and
noncaffeinated beverage as separate predictors in exploratory analyses. Caffeine-intake seemed
to provide a quick boost in positive affect (γ = .10, p = .041), whereas snacking and having
noncaffeine drinks did not show the effects (γ = .04, p = .309). Nevertheless, the significant
effect of caffeine-intake disappeared when all types of micro-breaks were included in the
analysis. These patterns suggest that the effects of nutrition-intake might have been masked by
other break activities due to their concurrent occurrences. Thus, it may be premature to draw a
definitive conclusion that nutrition-intake activities do not have recovery effects.
Moreover, the notion of limited self-regulatory resources has been central to the recovery
literature, but no study has actually tested specific resources as a mechanism involved in daily
micro-breaks and job performance outcomes (Trougakos & Hideg, 2009). Although Hunter and
Wu (2016) tested combined personal resources (energy, motivation, concentration) as a linking
mechanism between preferred break activities and somatic symptoms (e.g., headaches, eye
strain), it remained unclear whether micro-breaks can be connected to performance outcomes via
increased resources. Building on the major theoretical frameworks in recovery, we further drew
upon the performance model of affective influences (Beal et al., 2005) to theorize why state
positive affect from micro-break activities translates into substantive job performance outcomes.
This supported the theoretical notion that positive affect is a key self-regulatory resource
especially for employees who often need emotion regulation to achieve service and sales goals.

MICRO-BREAKS AND JOB PERFORMANCE
30
Of important note, both our study and Hunter and Wu (2016) found full mediation effects in
which break activities were not directly linked to the outcome variables but only through
increased personal resources. The consistent indirect effects in our study further corroborate the
theoretically driven role of micro-breaks in temporarily halting resource expenditure and
replenishing important resources for positive work outcomes. Thus, we advance theoretical and
empirical knowledge by showing that recovery activities (i.e., micro-breaks) influence
performance through an enhanced resource. Moreover, our findings of the objective performance
outcomes expand a recent study’s result that short, self-initiated breaks at work were related to
employees’ positive approaches to their job with more vigor, dedication, and enthusiasm (Kühnel
et al., 2017).
Last, it may be too simplistic to assume that work breaks aid momentary recovery for all
employees without considering individual or situational differences (Trougakos et al., 2014;
Sonnetag et al., 2017). When it comes to job performance, some individuals may be more
motivated to perform well, regardless of their fluctuating resource levels (Trougakos & Hideg,
2009). In accordance, we found that work engagement moderated the indirect effects of micro-
breaks on performance via positive affect, supporting the contention that not all individuals are
sensitized to transient affective resource for their job performance. That is, relaxation,
socialization, and cognitive micro-breaks are most beneficial for employees in low work
engagement in that they depend more on state positive affect due to their lack of work motivation
in general. In contrast, fluctuating positive affect has much less influence on employees in high
work engagement as they can remain highly engaged in their job tasks, regardless of situations.
In other words, those who have high work engagement may not need frequent micro-breaks to
replenish affective resources for job performance. Therefore, our findings add to the performance

MICRO-BREAKS AND JOB PERFORMANCE
31
model of affective influences (Beal et al., 2005) by showing that the within-person variability of
transient affect−performance linkage is contingent upon the important individual difference
factor, particularly for tasks that require high self-regulations. Additionally, our findings support
COR theory (Hobfoll, 2002): a stable resource reservoir (general work engagement) increases
resistance to the ebb and flow of transient resources (state positive affect). In short, our study
contributed to identifying general work engagement as a boundary condition for the performance
benefits of micro-breaks via positive affect.
Limitations and Future Research Directions
Overall, we used more rigorous methodological approaches: two daily surveys for ten
consecutive days, objective performance records, and confirmation through Bayesian multilevel
analysis that lessens concerns for traditional null hypothesis testing issues (Andraszewicz et al.,
2015; McKee & Miller, 2015). However, we must acknowledge a few limitations for future
research directions. First, the same measurement points for the predictor and mediator limited
causal inferences in our study, although we controlled for morning affect and workload to rule
out the possibility that they drive the hypothesized relationships. Nevertheless, for better causal
inferences, future studies should separate the time points for the predictor and mediator when
possible. In addition, although objective sales records were used to show that micro-breaks
matters for tangible job outcomes beyond the perceptual outcomes in the literature, our study
cannot determine the exact causal direction from work-day positive affect and performance
outcomes due to the inherent limitations of correlational design and daily aggregated sales data.
Thus, to empirically determine the causal directions, we strongly recommend that future studies
should confirm our findings using laboratory or field experiments in which researchers can

MICRO-BREAKS AND JOB PERFORMANCE
32
manipulate micro-breaks and induce positive affect and then examine various aspects of job
performance including quality of handling service and sales calls beyond quantitative metrics.
Third, although we found relaxation, socialization, and cognitive micro-breaks to have
positive effects independent of each other, nutrition-intake activities did not significantly
increase positive affect. Our current design cannot disentangle possible patterns of co-occurring
activities during micro-break episodes. Thus, future research should measure micro-break
activities at an episodic level, using alternative methods, such as event-contingent diary in which
participants record any event of micro-breaks in detail every time it occurs (Wheeler & Reis,
1991). Also, the event-level approach may better reflect post-break affective changes, their
influences on performance-related behaviors, and more nuanced break activities, such as habitual
versus purposeful snacking at work (cf. Sonnentag et al., 2016).
Next, our sample included only call center employees, perhaps restricting generalizability
of the current findings. Future studies should examine whether micro-breaks bring similar
performance benefits across different jobs and industries. In particular, given the fully mediated
effects via positive affect in our study, future research should confirm whether the indirect
effects also replicate in different job settings other than customer service and sales jobs. For
example, the impact of positive affective state will be beneficial only to the extent that the affect
matches the task requirements, in that positive affect might not be so helpful when job tasks
require a narrower set of response tendencies and attentional focus (Beal et al., 2005).
In addition, whereas we investigated only nonwork-related social activities as breaks in
this study, employees in other job contexts might enjoy work-related social activities. For
example, a researcher may take a small break from his or her manuscript-writing task and have a
casual conversation with colleagues about new research ideas. This break activity may still boost

MICRO-BREAKS AND JOB PERFORMANCE
33
the person’s positive affect. Perhaps, some specific job or performance contexts determine the
potential, salutary effects of work-related activities during breaks. Thus, future research needs to
explore when and why the context of work provides energizing experiences during breaks.
Another important note is that we tested general work engagement as the only cross-level
moderator, so future research should search for other situational or contextual factors that may
moderate micro-breaks’ recovery effects or their indirect effects on performance outcomes. For
example, some organizations and work groups may emphasize performance and competition too
much, underestimating values of well-being and social interactions (e.g., Hammer, Saksvik,
Nytro, Torvatn, & Bayazit, 2004). Under such workplace norms, employees might experience
adverse effects of micro-breaks as they may feel guilty about nonwork respite activities at work.
In fact, an organizational survey has shown that about 20% of respondent employees said they
did not step away from their workspaces due to guilt (Staples Advantage, 2014).
Future research should go beyond our study to explore micro-break timing. For example,
micro-breaks later in the day were found to be less effective than breaks before work shifts for
replenishing resources (Hunter & Wu, 2016), whereas another study found that afternoon micro-
breaks boosted daily work engagement while breaks in the morning failed to do so (Kühnel et al.,
2017). Although micro-breaks’ exact timings, durations, and specific activities may be difficult
to track in real time, researchers might try a day reconstruction method (see Diener & Tay, 2014,
for a review) in which participants systematically parse their previous day into major activities
and then rate their experiences and time spent on each activity (e.g., Oerlemans & Bakker, 2014).
In addition, micro-breaks might have an optimal duration for the best performance outcomes; too
much might be counterproductive. Thus, future research should explore micro-breaks’ durations,

MICRO-BREAKS AND JOB PERFORMANCE
34
timings, and their recovery effects on various outcomes, both over the course of a workday and
greater periods of time.
Last, we recommend that future studies investigate antecedents to micro-breaks and their
specific content. For example, Bowling, Beehr, and Swader (2005) found that extraverted
individuals engage in more frequent and positive social interactions at work. Also, a few studies
using a latent profile analysis approach found that work and personal factors (e.g., role
ambiguity, job control, supervisor support) are associated with certain profiles of post-work
recovery experiences (Bennett, Gabriel, Calderwood, Dabling, & Trougakos, 2016; Kinnunen et
al., 2016). Other studies have recently suggested that organizational climates (e.g., eating and
exercise climate) may influence employees’ snacking at work and health behaviors in general
(Sonnentag et al., 2016; Sonnentag & Pundt, 2016). Therefore, future research may find it
fruitful to examine various personality, work, and organizational factors as potential predictors of
micro-breaks and specific activities chosen.
Practical Implications
Our findings have important implications for organizational and managerial practices.
Organizations should educate their employees and managers in the values of micro-breaks
between task episodes for enhancing job performance, so that self-initiated micro-breaks are not
frowned upon. Considering that employees spend most of their lifetime working, they should
strategically construct their workday activities to allow brief moments for recovery, possibly
using a time-tracking app. In addition, organizations may redesign their call centers to be more
conducive to micro-break activities. For example, areas might be designated for specific
relaxation activities such as meditation or listening to music. Indeed, an intervention study found
that call center employees reported reduced physical and psychological strain after using a silent

MICRO-BREAKS AND JOB PERFORMANCE
35
break room with daybeds for practicing progressive muscle relaxation (Krajewski, Wieland, &
Sauerland, 2010). Organizational wellness training might include instructions for effective desk
stretching between calls and encouraging social interactions during breaks. In call centers,
employees are often separated by partitions which might constrain positive social interactions, so
easily accessible areas for breaks would help. For cognitive activities, we recommend giving
employees free access to the Internet, books, magazines, and newspapers, so that employees can
take a mental break from work demands.
As our results showed that micro-breaks have conditional indirect effects on performance
via positive affect, managers should be aware that employees may vary in their need for micro-
breaks depending on their general work engagement levels. Especially after difficult calls, those
who have low work engagement may need small breaks to boost positive affect and prevent
further loss of affective resources. Relatedly, we found that general work engagement positively
predicted an average level of job performance across days (i.e., direct effect of general work
engagement on the intercept of daily job performance). Thus, in addition to encouraging micro-
breaks on a need basis, we recommend that managers should help employees increase work
engagement to improve their performance, recognizing that day-to-day engagement experiences
may develop into work engagement over time (Bakker, 2011). Also, considering that both job
and personal resources drive the general work engagement levels, managers may try to provide
workers with necessary job resources (e.g., autonomy, performance feedback, social support,
coaching), as well as training to build personal resources such as self-efficacy and optimism
(Bakker, 2011).

MICRO-BREAKS AND JOB PERFORMANCE
36
References
Andraszewicz, S., Scheibehenne, B., Rieskamp, J., Grasman, R., Verhagen, J., & Wagenmakers,
E. J. (2015). An introduction to Bayesian hypothesis testing for management research.
Journal of Management, 41, 521-543. doi: 10.1177/0149206314560412
Bakker, A. B. (2011). An evidence-based model of work engagement. Current Directions in
Psychological Science, 20, 265-269. doi: 10.1177/0963721411414534
Bauer, D. J., Preacher, K. J., & Gil, K. M. (2006). Conceptualizing and testing random indirect
effects and moderated mediation in multilevel models: New procedures and
recommendations. Psychological Methods, 11, 142-163. doi: 10.1037/1082-
989X.11.2.142
Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: is the
active self a limited resource? Journal of Personality and Social Psychology, 74, 1252-
1265. doi: 10.1037/0022-3514.74.5.1252
Baumeister, R. F., Muraven, M., & Tice, D. M. (2000). Ego depletion: A resource model of
volition, self-regulation, and controlled processing. Social Cognition, 18, 130-150. doi:
10.1521/soco.2000.18.2.130
Beal, D. J., Trougakos, J. P., Weiss, H. M., & Green, S. G. (2006). Episodic processes in
emotional labor: perceptions of affective delivery and regulation strategies. Journal of
Applied Psychology, 91, 1053-1065. doi: 10.1037/0021-9010.91.5.1053
Beal, D. J., Weiss, H. M., Barros, E., & MacDermid, S. M. (2005). An episodic process model of
affective influences on performance. Journal of Applied Psychology, 90, 1054-1068. doi:
10.1037/0021-9010.90.6.1054

MICRO-BREAKS AND JOB PERFORMANCE
37
Bennett, A. A. (2015). Take five? Examining the impact of microbreak duration, activities, and
appraisals on human energy and performance. (Doctoral dissertation). Retrieved from
ProQuest Dissertations and Theses Database (UNI No. 3727013)
Bennett, A. A., Gabriel, A. S., Calderwood, C., Dabling, J. J., & Trougakos, J. P. (2016). Better
together? Examining profiles or employee recovery experiences. Journal of Applied
Psychology, 101, 1635-1654. doi: 10.1037/apl0000157
Binnewies, C., Sonnentag, S., & Mojza, E. J. (2009). Daily performance at work: Feeling
recovered in the morning as a predictor of day-level job performance. Journal of
Organizational Behavior, 30, 67-93. doi: 10.1002/job.541
Bowling, N. A., Beehr, T. A., & Swader, W. M. (2005). Giving and receiving social support at
work: The roles of personality and reciprocity. Journal of Vocational Behavior, 67, 476-
489. doi: 10.1016/j.jvb.2004.08.004
Breevaart, K., Bakker, A. B., Demerouti, E., & Hetland, J. (2012). The measurement of state
work engagement. European Journal of Psychological Assessment, 28, 305-312. doi:
10.1027/1015-5759/a000111
Brislin, R. (1970). Back-translation for cross-cultural research. Journal of Cross-Cultural
Psychology, 1, 186-216. doi: 10.1177/135910457000100301
Christian, M. S., Garza, A. S., & Slaughter, J. E. (2011). Work engagement: A quantitative
review and test of its relations with task and contextual performance. Personnel
Psychology, 64, 89-136. doi: 10.1111/j.1744-6570.2010.01203.x
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New York:
Academic Press.
Demerouti, E., Bakker, A. B., Geurts, S. A. E., & Taris, T. W. (2009). Daily recovery from

MICRO-BREAKS AND JOB PERFORMANCE
38
work-related effort during non-work time. In P. L. Perrewé, D. C. Ganster, & S.
Sonnentag (Eds.), Research in organizational stress and well-being (Vol. 7, pp. 85–123).
Oxford, England: Emerald Publishing Group.
Diamantopoulos, A., & Siguaw, J. A. (2006). Formative versus reflective indicators in
organizational measure development: A comparison and empirical illustration. British
Journal of Management, 17, 263-282. doi: 10.1111/j.1467-8551.2006.00500.x
Diener, E., & Tay, L. (2014). Review of the Day Reconstruction Method (DRM). Social
Indicators Research, 116, 255-267. doi: 10.1007/s11205-013-0279-x
Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel
models: A new look at an old issue. Psychological Methods, 12, 121–138. doi:
10.1037/1082-989X.12.2.121
Erez, A., & Isen, A. M. (2002). The influence of positive affect on the components of expectancy
motivation. Journal of Applied Psychology, 87, 1055-1067. doi: 10.1037/0021-
9010.87.6.1055
Fredrickson, B. L. (1998). What good are positive emotions? Review of General Psychology, 2,
300–319. doi: 10.1037/1089-2680.2.3.300
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden-
and-build theory of positive emotions. American Psychologist, 56, 218-226. doi:
10.1037/1089-2680.2.3.300
Fritz, C., Lam, C. F., & Spreitzer, G. M. (2011). It’s the little things that matter: An examination
of knowledge workers’ energy management. Academy of Management Perspectives, 25,
28-39.
Gailliot, M. T., Baumeister, R. F., DeWall, C. N., Maner, J. K., Plant, E. A., Tice, D. M., …

MICRO-BREAKS AND JOB PERFORMANCE
39
Schmiechel, B.J. (2007). Self-control relies on glucose as a limited energy source:
Willpower is more than a metaphor. Journal of Personality and Social Psychology, 92,
325-336. doi: 10.1037/0022- 3514.92.2.325
Gelman, A. (2003). A Bayesian formulation of exploratory data analysis and goodness-of-fit
testing. International Statistical Review, 71, 369-382.doi: 10.1111/j.1751-
5823.2003.tb00203.x
Grandey, A. A. (2000). Emotion regulation in the workplace: A new way to conceptualize
emotional labor. Journal of Occupational Health Psychology, 5, 59–100.
doi:10.1037/1076-8998.5.1.95
Grandey, A. A. (2003). When “the show must go on:” Surface acting and deep acting as
determinants of emotional exhaustion and peer-rated service delivery. Academy of
Management Journal, 46, 86-96. doi: 10.2307/30040678
Grandey, A. A., Dickter, D. N., & Sin, H.P. (2004). The customer is not always right: Customer
aggression and emotion regulation of service employees. Journal of Organizational
Behavior, 25, 397−418. doi: 10.1002/job.252
Halbesleben, J. R., & Wheeler, A. R. (2008). The relative roles of engagement and
embeddedness in predicting job performance and intention to leave. Work & Stress, 22,
242-256. doi: 10.1080/02678370802383962
Hammer, T. H., Saksvik, P. Ø., Nytrø, K., Torvatn, H., & Bayazit, M. (2004). Expanding the
psychosocial work environment: Workplace norms and work-family conflict as correlates
of stress and health. Journal of Occupational Health Psychology, 9, 83-97. doi:
10.1037/1076-8998.9.1.83
Hausser, J. A., Schlemmer, A., Kaiser, S., Kalis, A., & Mojzisch, A. (2014). The effects of

MICRO-BREAKS AND JOB PERFORMANCE
40
caffeine on option generation and subsequent choice. Psychopharmacology, 231, 3719-
3727. doi: 10.1007/s00213-014-3506-5
Henning, R. A., Jacques, P., Kissel, G. V., Sullivan, A. B., & Alteras-Webb, S. M. (1997).
Frequent short rest breaks from computer work: Effects on productivity and well-being at
two field sites. Ergonomics, 40, 78–91. doi: 10.1080/001401397188396
Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress.
American Psychologist, 44, 513–524. doi:10.1037/0003-066X.44.3.513
Hobfoll, S.E. (2001). The influence of culture, community, and the nested-self in the stress
process: advancing conservation of resources theory. Applied Psychology: An
International Review, 50, 337-370. doi: 10.1111/1464-0597.00062
Hobfoll, S. E. (2002). Social and psychological resources and adaptation. Review of General
Psychology, 6, 307-324. doi: 10.1037/1089-2680.6.4.307
Hunter, E. M., & Wu, C. (2016). Give me a better break: Choosing workday break activities to
maximize resource recovery. Journal of Applied Psychology, 101, 302–311. doi:
10.1037/apl0000045
Ilies, R., Schwind, K. M., Wagner, D. T., Johnson, M. D., DeRue, D. S., & Ilgen, D. R. (2007).
When can employees have a family life? The effects of daily workload and affect on
work-family conflict and social behaviors at home. Journal of Applied Psychology, 92,
1368–1379. doi:10.1037/0021-9010.92.5.1368
Jett, Q. R., & George, J. M. (2003). Work interrupted: A closer look at the role of interruptions in
organizational life. Academy of Management Review, 28, 494-507. doi:
10.5465/AMR.2003.10196791
Johnson, V. E. (2013). Revised standards for statistical evidence. Proceedings of the National

MICRO-BREAKS AND JOB PERFORMANCE
41
Academy of Sciences, 110, 19313-19317.
Kim, S., Park, Y., & Niu, Q. (2017). Micro-break activities in the workplace to recover from
daily work demands. Journal of Organizational Behavior, 38, 28-44. doi:
10.1002/job.2109
Kinnunen, U., Feldt, T., de Bloom, J., Sianoja, M., Korpela, K., & Geurts, S. (2016). Linking
boundary crossing from work to nonwork to work-related rumination across time: A
variable-and person-oriented approach. Journal of Occupational Health Psychology.
Advanced online publication. doi: 10.1037/ocp0000037
Korea Labor and Information Service. (2012). Labor Industry Statistics. Korea Labor and
Information Service. Retrieved from http://statistics.keis.or.kr/stats/index.do
Krajewski, J., Wieland, R., & Sauerland, M. (2010). Regulating strain states by using the
recovery potential of lunch breaks. Journal of Occupational Health Psychology, 15, 131-
139. doi: 10.1037/a0018830
Kruschke, J. K., Aguinis, H., & Joo, H. (2012). The time has come: Bayesian methods for data
analysis in the organizational sciences. Organizational Research Methods, 15, 722-752.
doi: 10.1177/1094428112457829
Kühnel, J., Zacher, H., de Bloom, J., & Bledow, R. (2017). Take a break! Benefits of sleep and
short breaks for daily work engagement. European Journal of Work and Organizational
Psychology, 26, 481-491. http://dx.doi.org/10.1080/1359432X.2016.1269750
Martin, A. D., Quinn, K. M., & Park, J. H. (2011). MCMCpack: Markov chain Monte Carlo in
R. Journal of Statistical Software, 42, 1–21. doi: 10.18637/jss.v042.i08
McKee, R. A., & Miller, C. C. (2015). Institutionalizing Bayesianism within the organizational
sciences: A practical guide featuring comments from eminent scholars. Journal of

MICRO-BREAKS AND JOB PERFORMANCE
42
Management, 41, 471-490. doi: 10.1177/0149206314546750
Meijman, T. F., & Mulder, G. (1998). Psychological aspects of workload. In P. J. D. Drenth, H.
Thierry & C. J. de Wolff (Eds.), Handbook of Work and Organizational Psychology,
Work Psychology, (Vol. 2, pp. 5-33). Hove, England: Psychology Press.
Miner, A. G., & Glomb, T. M. (2010). State mood, task performance, and behavior at work: A
within-persons approach. Organizational Behavior and Human Decision Processes, 112,
43-57. doi: 10.1016/j.obhdp.2009.11.009
Muthén, L. K., and Muthén, B. O. (2007). Mplus User’s Guide, Fifth Edition. Los Angeles, CA:
Muthén & Muthén.
Muraven, M., & Baumeister, R. F. (2000). Self-regulation and depletion of limited resources:
Does self-control resemble a muscle? Psychological bulletin, 126, 247-259.
doi: 10.1037/0033-2909.126.2.247
O*NET. (2015). Summary report for: 41-9041.00-telemarketers. O*NET online. Retrieved
from http://www.onetonline.org/link/summary/41-9041.00
Oerlemans, W. G., & Bakker, A. B. (2014). Burnout and daily recovery: A day reconstruction
study. Journal of Occupational Health Psychology, 19, 303-314. doi: 10.1037/a0036904
Owens, B. P., Baker, W. E., Sumpter, D. M., & Cameron, K. S. (2016). Relational energy at
work: Implications for job engagement and job performance. Journal of Applied
Psychology, 101, 35-49. doi: 10.1037/apl0000032
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. doi: 10.1037/0021-9010.88.5.879
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing interactions

MICRO-BREAKS AND JOB PERFORMANCE
43
in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of
Educational and Behavioral Statistics, 31, 437-448. doi: 10.3102/10769986031004437
Preacher, K. J., & Kelley, K. (2011). Effect size measures for mediation models: Quantitative
strategies for communicating indirect effects. Psychological Methods, 16, 93–115.
http://dx.doi.org/10.1037/a0022658
Preacher, K. J., & Selig, J. P. (2010). Monte Carlo method for assessing multilevel mediation:
An interactive tool for creating confidence intervals for indirect effects in 1-1-1
multilevel models [Computer software]. Retrieved from http://quantpsy.org/
Pugh, S. D. (2001). Service with a smile: Emotional contagion in the service encounter. Academy
of Management Journal, 44, 1018-1027. doi: 10.2307/3069445
Renner, B., Sproesser, G., Strohbach, S., & Schupp, H. T. (2012). Why we eat what we eat: The
Eating Motivation Survey (TEMS). Appetite, 59, 117-128. doi:
10.1016/j.appet.2012.04.004
Rodriguez, R. (2013, March 29). Goofing off company time? Go for it. CNN.com. Retrieved
from http://www.cnn.com/2013/03/29/living/play-at-work-irpt/
Rothbard, N. P., & Wilk, S. L. (2011). Waking up on the right or wrong side of the bed: Start-of-
workday mood, work events, employee affect, and performance. Academy of
Management Journal, 54, 959-980. doi: 10.5465/amj.2007.0056
Salanova, M., Agut, S., & Peiró, J. M. (2005). Linking organizational resources and work
engagement to employee performance and customer loyalty: The mediation of service
climate. Journal of Applied Psychology, 90, 1217-1227. doi: 10.1037/0021-
9010.90.6.1217
Schaufeli, W. B., Bakker, A. B., & Salanova, M. (2006). The measurement of work engagement

MICRO-BREAKS AND JOB PERFORMANCE
44
with a short questionnaire a cross-national study. Educational and Psychological
Measurement, 66, 701-716. doi: 10.1177/0013164405282471
Sellke, T., Bayarri, M. J., & Berger, J. O. (2001). Calibration of p values for testing precise null
hypotheses. The American Statistician, 55, 62-71. doi: 10.1198/000313001300339950
Shockley, K. M., Ispas, D., Rossi, M. E., & Levine, E. L. (2012). A meta-analytic investigation
of the relationship between state affect, discrete emotions, and job performance. Human
Performance, 25, 377-411. doi: 10.1080/08959285.2012.721832
Sianoja, M, Syrek, C. J., de Bloom, J., Korpela, K., & Kinnunen, U. (2017). Enhancing daily
well-being at work through lunchtime park walks and relaxation exercises: Recovery
experiences as mediators. Journal of Occupational Health Psychology, Advanced online
publication. doi: 10.1037/ocp0000083
Sonnentag, S. (2001). Work, recovery activities, and individual well-being: A diary study.
Journal of Occupational Health Psychology, 6, 196–210. doi: 10.1037/1076-
8998.6.3.196
Sonnentag, S. (2003). Recovery, work engagement, and proactive behavior: A new look at the
interface between nonwork and work. Journal of Applied Psychology, 88, 518-528. doi:
10.1037/0021-9010.88.3.518
Sonnentag, S., & Fritz, C. (2007). The recovery experience questionnaire: Development and
validation of a measure for assessing recuperation and unwinding from work. Journal of
Occupational Health Psychology, 12(3), 204-221. doi: 10.1037/1076-8998.12.3.204
Sonnentag, S., & Fritz, C. (2015). Recovery from job stress: The stressor‐detachment model as
an integrative framework. Journal of Organizational Behavior, 36, S72-S103.
doi: 10.1002/job.1924

MICRO-BREAKS AND JOB PERFORMANCE
45
Sonnentag, S. & Pundt, A. (2016). Organizational health behavior climate: Organizations can
encourage healthy eating and physical exercise. Applied Psychology: An International
Review, 65, 259-286. doi: 10.1111/apps.12059
Sonnentag, S., Pundt, A., & Venz, L. (2016). Distal and proximal predictors of snacking at work:
A daily-survey study. Journal of Applied Psychology, 102, 151-162. doi:
10.1037/apl0000162
Sonnentag, S., Venz, L., & Casper, A. (2017). Advances in recovery research: What have we
learned? What should be done next? Journal of Occupational Health Psychology, 22,
365-380. doi: 10.1037/ocp0000079
Spector, P. E., & Jex, S. M. (1998). Development of four self-report measures of job stressors
and strain: Interpersonal conflict at work scale, organizational constraints scale,
quantitative workload inventory, and physical symptoms inventory. Journal of
Occupational Health Psychology, 3, 356–367. doi:10.1037/1076-8998.3.4.356
Stapes Advantage (2014). News release: Staple survey reveals many employees feel too guilty to
take breaks, despite spending longer hours at work. Retrieved from
http://investor.staples.com/phoenix.zhtml?c=96244&p=irol-
newsArticle&ID=1928035&highlight=
Steidle, A., Gonzalez-Morales, M. G., Hoppe, A., Michel, A., & O’shea, D. (2017). Energizing
respites from work: A randomized controlled study on respite interventions. European
Journal of Work and Organizational Psychology, 26, 650-662. doi:
10.1080/1359432X.2017.1348348
ten Brummelhuis, L. L., & Bakker, A. B. (2012). Staying engaged during the week: The effect of
off-job activities on next day work engagement. Journal of Occupational Health

MICRO-BREAKS AND JOB PERFORMANCE
46
Psychology, 17, 445-455. doi: 10.1037/a0029213
ten Brummelhuis, L. L., & Trougakos, J. P. (2014). The recovery potential of intrinsically versus
extrinsically motivated off‐job activities. Journal of Occupational and Organizational
Psychology, 87, 177-199. doi: 10.1111/joop.12050
Trougakos, J. P., Beal, D. J., Green, S. G., & Weiss, H. M. (2008). Making the break count: An
episodic examination of recovery activities, emotional experiences, and positive affective
displays. Academy of Management Journal, 51, 131–146. doi:
10.5465/AMJ.2008.30764063
Trougakos, J. P., & Hideg, I. (2009). Momentary work recovery: The role of within day work
breaks. In S. Sonnentag, P. L. Perrrewé, & D. C. Ganster (Eds.), Research in
Occupational Stress and Well-Being, (Vol. 7, pp. 37-84). Oxford, UK: JAI Press. doi:
10.1108/S1479-3555(2009)0000007005
Trougakos, J. P., Hideg, I., Cheng, B., & Beal, D. (2014). Lunch breaks unpacked: The role of
autonomy as a moderator of recovery during lunch. Academy of Management Journal,
57, 405-421. doi: 10.5465/amj.2011.1072
Tsai, W. C. (2001). Determinants and consequences of employee displayed positive
emotions. Journal of Management, 27, 497-512. doi: 10.1177/014920630102700406
Tsai, W. C., Chen, C. C., & Liu, H. L. (2007). Test of a model linking employee positive moods
and task performance. Journal of Applied Psychology, 92, 1570-1583. doi: 10.1037/0021-
9010.92.6.1570
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: the PANAS scales. Journal of Personality and Social
Psychology, 54, 1063-1070. doi: 10.1037/0022-3514.54.6.1063

MICRO-BREAKS AND JOB PERFORMANCE
47
Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion of the
structure, causes and consequences of affective experiences at work. In B. M. Staw & L.
L. Cummings (Eds.), Research in Organizational Behavior, (Vol. 18, pp. 1-74).
Greenwich, CT: JAI Press.
Welsh, D. T., Ellis, A. P., Christian, M. S., & Mai, K. M. (2014). Building a self-regulatory
model of sleep deprivation and deception: The role of caffeine and social influence.
Journal of Applied Psychology, 99, 1268-1277. doi: 10.1037/a0036202
Wheeler, L., & Reis, H. T. (1991). Self‐recording of everyday life events: Origins, types, and
uses. Journal of Personality, 59, 339-354. doi: 10.1111/j.1467-6494.1991.tb00252.x
Yuan, Y., & MacKinnon, D. P. (2009). Bayesian mediation analysis. Psychological Methods, 14,
301-322. doi: 10.1037/a0016972
Zacher, H., Brailsford, H. A., & Parker, S. L. (2014). Micro-breaks matter: A diary study on the
effects of energy management strategies on occupational well-being. Journal of
Vocational Behavior, 85, 287-297. doi: 10.1016/j.jvb.2014.08.005

MICRO-BREAKS AND JOB PERFORMANCE
48
Table 1
Common Micro-Break Activities
Category
Examples
Relaxation activities
• Stretching, walking around the office, or relaxing briefly
• Daydreaming, gazing out the office windows, taking a quick nap, or any other psychological relaxation
Nutrition-intake activities
• Drinking caffeinated beverages (e.g., energy drinks, coffee, black or green tea)
• Snacking (e.g., cookies) or drinking non-caffeinated beverages (e.g., juice, water, vitamin water)
Social activities
• Chatting with coworkers on non-work related topics
• Texting, using instant messenger, or calling to friends or family members
• Checking personal SNS (e.g., Facebook, Twitter, or personal blogs)
Cognitive activities
• Reading books, newspapers, or magazines for personal learning or entertainment.
• Surfing the Web for entertainment (e.g., watching short video clips, playing a game)

MICRO-BREAKS AND JOB PERFORMANCE
49
Table 2
Means, Standard Deviations, and Intercorrelations among Study Variables
Variable M SDa SDb 1 2 3 4 5 6 7 8 9
1. General work engagement (Level 2) 3.16 − 1.00 .18 .07 -.08 .08 .14 .02 .08 .16
2. Morning positive affect (Control) 2.84 0.72 0.30 − -.19 -.03 .19 .03 .22 .24* .16
3. Workload (Control) 3.00 0.68 0.29 − -.04 .02 -.11 .06 -.15 -.02 .05
4. Relaxation activities 2.58 0.88 0.50 − -.04 -.07 -.11 .31** .14 .24* -.17
5. Nutrition-intake activities 2.70 0.86 0.50 − .05 -.05 -.02 .40** .68*** .19 .02
6. Social activities 2.68 0.92 0.44 − -.06 -.06 .19*** .32*** .33** .18 .06
7. Cognitive activities 2.88 0.82 0.59 − .04 -.07 .08 .57*** .29** .32** -.10
8. Positive affect 2.88 0.66 0.34 − .12** -.03 .21*** .17*** .24*** .23*** .22
9. Sales performance 0.00 1.07 0.51 − .22*** .10** -.05 .07 .07 .01 .29***
Note. All variables are within persons, except general work engagement. Correlations below the diagonal represent within-person correlations (n =
632). Correlations above the diagonal represent between-person correlations (n = 71). To calculate between-person correlations, we averaged within-
person scores across days. a = within-person, b = between-person. Sales performance was standardized based on individual mean scores during the
two weeks. *p < .05, **p < .01; ***p < .001.

MICRO-BREAKS AND JOB PERFORMANCE
50
Table 3
Unstandardized Coefficients of the Multilevel Model
Positive Affect
Sales Performance
Estimate
S.E.
Est./SE
Estimate
S.E.
Est./SE
Intercept
2.88
0.04
72.57
***
-1.18
0.21
-5.63
***
General work engagement
a
0.52
0.18
2.93
**
Control: Morning positive affect
b
0.11
0.04
3.24
**
0.37
0.06
6.59
***
Control: Workload
b
0.002
0.04
0.06
0.21
0.06
3.49
***
Relaxation activities
b
0.13
0.03
4.04
***
-0.02
0.05
-0.34
Nutrition-intake activities
b
0.04
0.04
0.97
0.04
0.06
0.62
Socialization activities
b
0.14
0.03
4.46
***
0.04
0.05
0.84
Cognitive activities
b
0.13
0.05
2.77
**
0.08
0.07
1.11
Positive affect (PA)
b
0.41
0.07
5.88
***
PA x General work engagementc - 0.15 0.06 -2.59*
Level 1 n = 632, Level 2 n = 71. a between-person variable; b within-persons variables; c cross-level interaction term;
S.E. = standard error. All results came from a path model that included all variables: controls, predictors, mediator,
moderator, and interaction term. *p < .05, **p < .01; ***p < .001.

MICRO-BREAKS AND JOB PERFORMANCE
51
Table 4
Results of Bayesian Multilevel Analysis
Sales Performance
Estimate
S.E.
2.5% of CI
97.5% of CI
Indirect effects of relaxation activities
Indirect effects of nutrition-intake activities
Indirect effects of social activities
Indirect effects of cognitive activities
Positive affect (PA)
General work engagement
PA x Work engagement
Conditional indirect effects of relaxation activities
Conditional indirect effects of nutrition-intake activities
Conditional indirect effects of social activities
Conditional indirect effects of cognitive activities
0.15
0.01
0.08
0.21
0.04
0.01
-0.03
0.08
0.12
0.01
0.07
0.20
0.13
0.01
0.05
0.21
0.89
0.09
0.69
1.24
0.40
0.06
0.20
0.68
-0.15
0.02
-0.24
-0.08
-0.02
0.01
-0.03
-0.01
0.001
0.01
-0.02
0.01
-0.02
0.01
-0.04
-0.01
-0.02
0.01
-0.03
-0.01
Note. Level 1 n = 632, Level 2 n = 71. SE = standard error. PA = positive affect. The conditional indirect effects indicate that
as the score of moderator, general work engagement, increases by one unit, each indirect effect decreases, except nutrition-intake breaks.

MICRO-BREAKS AND JOB PERFORMANCE
52
Between-Individual (Level 2)
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Within-Individual (Level 1)
Figure 1. Conceptual Model
Note. Control variables’ paths were not shown in the figure for simplicity but were included in the analysis (i.e., day-specific morning positive affect
and workload predicting the mediator and outcome variable).
Day-specific relaxation activities
Day-specific nutrition-intake activities
Day-specific social activities
Day-specific cognitive activities
Day-specific positive affect
+
+
+
+
-
General work engagement
Day-specific sales performance
+

MICRO-BREAKS AND JOB PERFORMANCE
53
Figure 2. Cross-Level Moderation Effect of General Work Engagement in Predicting Daily Sales Performance
-3
-2
-1
0
1
2
Sales Performance
Low General Work
Engagement
High General Work
Engagement
Low Positive Affect High Positive Affective