Feeling interrupted—Being responsive: How online messages relate to affect at work

Abstract

Being constantly connected to others via e-mail and other online messages is increasingly typical for many employees. In this paper, we develop and test a model that specifies how interruptions by online messages relate to negative and positive affect. We hypothesize that perceived interruptions by online messages predict state negative affect via time pressure and that perceived interruptions predict state positive affect via responsiveness to these online messages and perceived task accomplishment. A daily survey study with 174 employees (a total of 811 day-level observations) provided support for our hypotheses at the between-person and within-person level. In addition, perceived interruptions showed a negative direct association with perceived task accomplishment. Our study highlights the importance of being responsive to online messages and shows that addressing only the negative effects of perceived interruptions does not suffice to understand the full impact of interruptions by online messages in modern jobs.

Full text

RESEARCH ARTICLE
Feeling interrupted—Being responsive: How online messages
relate to affect at work
Sabine Sonnentag
1
|Leonard Reinecke
2
|Jutta Mata
1
|Peter Vorderer
1
1
University of Mannheim, Mannheim,
Germany
2
University of Mainz, Mainz, Germany
Correspondence
Sabine Sonnentag, Department of Psychology,
University of Mannheim, Schloss Ehrenhof
Ost, D‐68131 Mannheim, Germany.
Email: sonnentag@uni‐mannheim.de
Summary
Being constantly connected to others via e‐mail and other online messages is increasingly typical
for many employees. In this paper, we develop and test a model that specifies how interruptions
by online messages relate to negative and positive affect. We hypothesize that perceived inter-
ruptions by online messages predict state negative affect via time pressure and that perceived
interruptions predict state positive affect via responsiveness to these online messages and per-
ceived task accomplishment. A daily survey study with 174 employees (a total of 811 day‐level
observations) provided support for our hypotheses at the between‐person and within‐person
level. In addition, perceived interruptions showed a negative direct association with perceived
task accomplishment. Our study highlights the importance of being responsive to online mes-
sages and shows that addressing only the negative effects of perceived interruptions does not
suffice to understand the full impact of interruptions by online messages in modern jobs.
KEYWORDS
affect, daily survey, e‐mail, interruptions
1|INTRODUCTION
Using e‐mail and other online messages is ubiquitous in many contem-
porary jobs. In 2015, worldwide 112.5 billion business e‐mails were
sent and received per day, corresponding to 88 e‐mails received and
34 e‐mails sent per user per day (The Radicati Group, 2015). Despite
the increase in job‐related use of social media (El Ouirdi, El Ouirdi,
Segers, & Henderickx, 2015; Skeels & Grudin, 2009) and the insight
that e‐mail might not be the best communication medium for many
purposes (Middleton & Cukier, 2006; Tan, Sutanto, Phang, & Gasimov,
2014), e‐mail is still widely used on the job. For instance, knowledge
workers spend 28% of their workday reading and responding to e‐mail
(Chui et al., 2012). Accordingly, Barley, Meyerson, and Grodal (2011, p.
887) have characterized e‐mail as a “source and symbol of stress.”
Research in the organizational and information sciences has often
looked at e‐mail and other online messages from such a stress per-
spective (Barber & Santuzzi, 2015) and has examined how these mes-
sages relate to interruptions of ongoing work (González & Mark,
2004; Jackson, Dawson, & Wilson, 2001; Mark, Iqbal, Czerwinski,
Johns, & Sano, 2016). Interruptions, in turn, have been described as
compromising task performance and eliciting negative affective states
(Mark, Gudith, & Klocke, 2008; Marulanda‐Carter & Jackson, 2012).
Wajcman and Rose (2011), however, have argued that approaches
that see e‐mail and other online messages exclusively as disruptive
may miss an important feature of many modern work settings where
“constant connectivity”(p. 941) is widespread. On the basis of quali-
tative data, Wajcman and Rose described that being available and
responsive to incoming messages is highly important to get work
done. Despite these insights, there is a paucity of research that inte-
grates this new perspective with the more traditional negative view
on interruptions. Specifically, it remains unclear how the potentially
negative and positive consequences of being interrupted are jointly
experienced on a day‐to‐day basis, how they translate into other neg-
ative and positive on‐the‐job experiences (e.g., time pressure and per-
ceived task accomplishment, respectively), and how they ultimately
relate to negative and positive affect as highly relevant outcome var-
iables in organizational behavior (Ashkanasy & Dorris, 2017). In this
study, we aim at reconciling earlier research on interruptions that
focused on negative outcomes (Mark et al., 2008) with the more
recent perspective featuring a more positive view (Wajcman & Rose,
2011). We develop and test a model that describes how interruptions
by e‐mail messages and responsiveness towards these messages are
linked to negative and positive affect at work. Specifically, we pro-
pose that interruptions are related to negative affect via time pressure
Received: 24 March 2016 Revised: 16 August 2017 Accepted: 3 September 2017
DOI: 10.1002/job.2239
J Organ Behav. 2017;1–15. Copyright © 2017 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/job 1

and to positive affect via responsiveness and perceived task
accomplishment.
Earlier research has mainly focused on overall consequences of
using e‐mail and other online messages at work (Tassabehji & Vakola,
2005) or has examined how experiences associated with e‐mail usage
differ between persons (Brown, Duck, & Jimmieson, 2014), but has
largely neglected within‐person dynamics of e‐mail usage and of inter-
ruptions by e‐mail. This is an important oversight because many orga-
nizational phenomena fluctuate within person (Beal, 2015; Sonnentag,
2015) and because research on between‐person differences does not
provide any insight into within‐person day‐level dynamics (Dalal,
Bhave, & Fiset, 2014).
To address this paucity of research, in our study we examine inter-
ruptions by e‐mail and other online messages experienced at the day
level and examine how these perceived interruptions predict day‐level
positive and negative affect at work. Affect at work is highly important
for on‐the‐job behavior and for nonwork life. First, research has shown
that positive affective states predict organizational citizenship behav-
ior (Ilies, Scott, & Judge, 2006; Spence, Brown, Keeping, & Lian,
2014) as well as proactive and creative behavior (Amabile, Barsade,
Mueller, & Staw, 2005; Fay & Sonnentag, 2012), whereas negative
affective states predict work withdrawal and other types of counter-
productive behaviors (Rodell & Judge, 2009; Scott & Barnes, 2011).
Second, due to emotional contagion processes, affect of one person
crosses over to other persons in the work environment, thereby affect-
ing the behavior of whole work groups (Barsade, 2002; Zimmermann,
Dormann, & Dollard, 2011). Finally, affect experienced at work spills
over into nonwork life and has an impact on social behavior at home
(Ilies et al., 2007; Sears, Repetti, Robles, & Reynolds, 2016) and on
nighttime sleep (Kalmbach, Pillai, Roth, & Drake, 2014). Poor nighttime
sleep, in turn, undermines desirable behaviors at work (Barnes,
Lucianetti, Bhave, & Christian, 2015). Accordingly, identifying predic-
tors of positive and negative affect at work helps to understand the
factors that ultimately stimulate favorable as well as unfavorable
behaviors and processes at work and at home.
Our paper makes several contributions to the literature. First, we
contribute to theory building by developing a model that describes
both negative and positive experiences associated with perceived
interruptions by online messages. Specifically, we examine time pres-
sure as a negative stress‐related mechanism and responsiveness and
perceived task accomplishment as a more positive beneficial mecha-
nism that link perceived interruptions by online messages to negative
and positive affect, respectively. We introduce responsiveness (i.e., a
person's timely reaction to incoming online messages) as a novel con-
cept that helps to understand how employees incorporate online
messages into their daily work and why interruptions by online mes-
sages may have affective benefits. It is important to note that time
pressure and perceived task accomplishment are not only important
as precursors of negative and positive affect. They are relevant on‐
the‐job experiences by themselves. For instance, time pressure is
associated with increased exhaustion levels (Prem, Paškvan, Kubicek,
& Korunka, 2016) and fosters risky decisions (Madan, Spetch, &
Ludvig, 2015), whereas perceived task accomplishment triggers sub-
sequent motivational processes (e.g., persistence; Seo, Bartunek, &
Barrett, 2010).
Second, although we specifically address perceived interruptions
by online messages, our study contributes to the broader literature
on interruptions at work (Jett & George, 2003). We suggest that the
currently dominant view (i.e., interruptions are detrimental for on‐task
processes) might only be one side of the coin. Interruptions at work
may in fact foster positive outcomes in an indirect way. By stimulating
responsiveness, interruptions provide an opportunity for secondary
activities such as helping coworkers (Richardson & Taylor, 2012) and
subsequently experiencing more meaning at work (Lam, Wan, &
Roussin, 2016).
Third, we contribute to research on affect at work (Beal &
Ghandour, 2011; Sonnentag & Starzyk, 2015). Research in organiza-
tional behavior described the role of information and communication
technology (ICT) in organizations (Gephart, 2002; Mazmanian, 2013),
but has rarely linked aspects of ICT use to affect at work. Although
previous research did examine how job‐related electronic communica-
tion at home predicts affective states at home (Butts, Becker, & Bos-
well, 2015), it has rarely addressed the question of how it relates to
fluctuations of affect at work on a day‐to‐day basis. This is an impor-
tant oversight because by omitting ICT use as a potential predictor of
workplace affect, the picture of what causes fluctuations in affect at
work remains incomplete, and possible starting points for interventions
stay unexplored. Our study looks at interruptions by online messages
as one specific aspect of ICT use at work. Accordingly, the study will
help to better understand interruptions as one of the many everyday
experiences at a digital workplace (Colbert, Yee, & George, 2016).
In terms of study methodology, we extend the qualitative
approach of Wajcman and Rose who focused on a sample of 18 knowl-
edge workers employed within one single company, holding mainly
managerial jobs and spending most of their working time on communi-
cation activities. In our study, we use a quantitative approach with a
much broader sample comprising different types of jobs. We move
beyond a general description of online messages in today's jobs and
aim at capturing day‐level fluctuations of perceived interruptions and
responsiveness to better understand the mechanisms through which
online messages may impact affect at work. Thus, we examine the
interplay between our study variables at the day level.
In the next sections of the paper, we describe the experiences of
feeling interrupted and being responsive as core concepts and develop
our hypotheses on how interruptions and responsiveness relate to
negative and positive affect. Figure 1 shows our conceptual model.
2|CORE CONCEPTS: FEELING
INTERRUPTED AND BEING RESPONSIVE
Broadly defined, interruptions are “incidents or occurrences that
impede or delay […] progress on work tasks”(Jett & George, 2003, p.
494). In most instances, interruptions by e‐mails and other online mes-
sages are prototypical intrusions, defined as “an unexpected encounter
initiated by another person that interrupts the flow and continuity of
an individual's work and brings that work to a temporary halt”(p.
495). Within modern information‐technology systems the initiation
“by another person”might be rather indirect because the interruption
can result from an automated preprogrammed message. Interruptions
2SONNENTAG ET AL.

as intrusions have been described as “external interruptions”(González
& Mark, 2004, p. 118) that have to be differentiated from self‐initiated
breaks or self‐initiated switches in task activities. When we use the
term interruptions in this paper, we refer to external interruptions
(González & Mark, 2004) or intrusions (Jett & George, 2003) that dis-
rupt the process of working on a task. Importantly, our specific focus
is on the subjective experience of being interrupted.
Typically, interruptions imply that employees discontinue the
activity they are currently working on and shift to another task
(Wajcman & Rose, 2011). Interruptions come with cognitive costs
(Eyrolle & Cellier, 2000; Oulasvirta & Saariluoma, 2004) because they
divert the attention away from the primary task (Gupta, Li, & Sharda,
2013). They often result in a fragmentation of the workday (Mark,
González, & Harris, 2005; Rose, 2014).
The concept of online responsiveness refers to short response
latencies in computer‐mediated communication (Kalman, Scissors, Gill,
& Gergle, 2013). Responsiveness is high when e‐mails and other online
messages are responded to immediately or without significant delay.
Usually, receivers of online messages appreciate short response
latencies, that is a high responsiveness of their communication
partners (Kalman & Rafaeli, 2011). Although average response time
to incoming e‐mail is typically low, responsiveness varies considerably
between individuals (Kalman & Ravid, 2015; Kalman, Ravid, Raban, &
Rafaeli, 2006) and may also fluctuate within individuals. Responsiveness
to e‐mail is predicted both by stable traits such as extraversion (Kalman
et al., 2013) as well as by situational factors such as the number of
received messages and attention paid to the inbox (Kalman & Ravid,
2015), characteristics of the sender (e.g., work‐relationship between
sender and recipient) and the perceived importance of the message
(Dabbish, Kraut, Fussell, & Kiesler, 2005) as well as the recipient's
engagement in other tasks when the message arrives (Avrahami &
Hudson, 2006). Because these situational predictors of response
latencies are likely to show daily fluctuations, so should the level of
responsiveness of individual employees. In this study, we thus refer
to responsiveness as the day‐level variation in the immediacy of
reacting to work‐related online messages.
We propose that perceived interruptions by online messages will
be positively related to responsiveness. Usually, an incoming online
message will interrupt the ongoing work process. A perceived interrup-
tion is the precondition for being responsive. Without perceiving an
interruption by an online message, it is impossible to respond to the
message. Admittedly, there might be exceptions of this general pat-
tern, for instance when one is waiting for an incoming message and
when one is already prepared to respond to it; in such a situation
one might not feel interrupted. In most instances, however, one will
first need to be interrupted by a message in order to be responsive,
contributing to a positive association between perceived interruptions
and responsiveness.
In addition, although interruptions are experienced as disruptive
and aversive (Mark et al., 2008), they may trigger responsiveness
towards further incoming online messages. This rationale is supported
by prior research that demonstrates a negative correlation between
the number of received e‐mails and response latency (Kalman & Ravid,
2015). This finding suggests that on days when employees are
frequently interrupted by online messages they tend to shift their
attention to these online messages. This shift in attention may not be
limited to the specific online message that has disrupted work on the
primary task, but will make employees aware that more online
messages might come in. Thus, the overall salience of online messages
will increase, and employees may anticipate more incoming messages
(Carton & Aiello, 2009). Accordingly, employees will be mentally
prepared for more incoming messages and will be ready to address
them rather quickly once they arrive. Moreover, on days when
employees perceive interruptions by many online messages they may
decide to respond quickly to the online messages in order to avoid that
more and more incoming online messages pile up. This quick turnaround
of incoming online messages is experienced as responsiveness. On
days, however, when no interruptions by online messages occur, there
is no need to be ready and to respond.
Hypothesis 1. Perceived interruptions by online mes-
sages are positively related to responsiveness.
3|PERCEIVED INTERRUPTIONS AND TIME
PRESSURE: POSITIVE ASSOCIATION
Jett and George (2003) described time pressure as a negative conse-
quence of interruptions. Time pressure is a job stressor that refers to
the experience of having too much to do in too little time (Parker &
DeCotis, 1983). It is often experienced as time scarcity (DeVoe &
Pfeffer, 2011) or time famine (Perlow, 1999). Being interrupted by an
online message while working on a task implies that work on this
FIGURE 1 Conceptual model
SONNENTAG ET AL.3

primary task comes to a temporary halt. While addressing the interrup-
tion, no effort can be invested into the primary task and time goes by
without making progress on the primary task. As a consequence when
later resuming work on the primary task, less time will be available for
accomplishing the primary task, resulting in the experience of time
pressure (Baethge, Rigotti, & Roe, 2015).
Moreover, switching back from the interruption to the primary
task may involve additional costs because time is needed to reorient
oneself back towards the primary task what might even result in
redundant work (Mark et al., 2005). Experimental studies have shown
that when persons are interrupted while working on a primary task,
they need longer to complete this task (Bailey & Konstan, 2006;
Eyrolle & Cellier, 2000) and, as a consequence, overall time pressure
increases.
Empirical evidence from a cross‐sectional study in a call center
(Grebner et al., 2003) and from experimental research with simulated
interruptions from supervisors (Mark et al., 2008) demonstrates that
interruptions are associated with increased time pressure. Using a
daily‐diary design in a nursing environment, Baethge and Rigotti
(2013) found that the number of perceived work‐flow interruptions
was associated with the experience of time pressure.
Hypothesis 2. Perceived interruptions by online mes-
sages are positively related to time pressure.
4|PERCEIVED INTERRUPTIONS AND
PERCEIVED TASK ACCOMPLISHMENT:
DIRECT NEGATIVE ASSOCIATION
Perceived interruptions by online messages will have a direct negative
association with perceived task accomplishment. Perceived task
accomplishment refers to the employee's subjective task performance
(Jimmieson & Terry, 1997) as his or her evaluation of how well he or
she is doing with respect to job duties and responsibilities. During
everyday work situations, it captures the subjective experience of
being effective and making progress towards one's goals (Fisher &
Noble, 2004). Perceived task accomplishment as a subjective experi-
ence overlaps with objective indicators of task performance, but is dis-
tinct from it (Harris & Schaubroeck, 1988).
Experiencing an interruption implies that the process of
accomplishing primary tasks is disrupted (Trafton & Monk, 2007) and
may be abandoned altogether (O'Conaill & Frohlich, 1995). An
interruption draws an employee's attention away from the primary
task, causing an attentional conflict between the primary task and
the interruption (Speier, Vessey, & Valacich, 2003). Switching attention
between the primary task and the interruption is associated with
switching time, and switching back from the interruption to the
primary task is associated with recall time (Gupta et al., 2013). These
switching and recall costs slow down the task‐completion process.
More specifically, to be able to continue the task‐accomplishment
process after having dealt with an interruption, an employee must
remember to resume the primary task. Thus, in the terminology of
cognitive psychology, the employee must create a prospective memory
task (Dodhia & Dismukes, 2009; Einstein, McDaniel, Williford, Pagan,
& Dismukes, 2003) and needs to keep the goal of the primary task in
his or her memory (Altmann & Trafton, 2002). Errors may occur both
during encoding as well as during retrieval of the information needed
to resume the primary task (Hodgetts & Jones, 2006; Trafton, Altmann,
Brock, & Mintz, 2003) and the resumption of the primary task may be
delayed (Monk, Trafton, & Boehm‐Davis, 2008). Errors and delays in
task resumption will compromise task performance. Accordingly,
perceiving interruptions will contribute to the feeling of not making
sufficient progress towards one's work goals.
Hypothesis 3. Perceived interruptions by online mes-
sages show a direct negative relationship with perceived
task accomplishment.
5|PERCEIVED INTERRUPTIONS AND
PERCEIVED TASK ACCOMPLISHMENT:
INDIRECT POSITIVE ASSOCIATION
In addition to the direct negative association between perceived inter-
ruptions and perceived task accomplishment, we suggest that there
will be an indirect positive association between perceived interrup-
tions and perceived task accomplishment, mediated via responsive-
ness. In Hypothesis 1 we have stated that perceived interruptions by
online messages are positively related to responsiveness. In turn, we
propose that being responsive towards online messages is positively
related to perceived task accomplishment.
In many of today's jobs, dealing with online messages is a
ubiquitous demand or—as Wajcman and Rose (2011, p. 950) put it:
“Attending to these mediated communications is a normal and
essential part of work.”Often work is done by responding to incoming
online messages and sending messages oneself. This implies that
ignoring online messages can compromise one's work performance.
Being responsive means to quickly answer incoming online mes-
sages and to engage quickly in an online conversation because one is
mentally prepared to do so. We propose that responsiveness contrib-
utes to perceived task accomplishment for three main reasons. First,
being responsive means to react quickly to incoming messages.
Because online messages often provide information essential for one's
work (Wajcman & Rose, 2011), by responding immediately one actu-
ally addresses one's work tasks. Being responsive to online messages
implies to get a portion of one's work done (Barley et al., 2011). Over
time this experience will contribute to the perception of making prog-
ress with one's tasks.
Second, responding to online message means to direct one's
attention to these messages, whereas not responding means to ignore
these messages. Also when deliberately ignoring incoming messages,
in many instances, one will be aware that messages have come in. This
awareness may trigger thoughts about these messages one is trying to
ignore, pulling attention away from the task one is actually working on
(Beal, Weiss, Barros, & E.,, & MacDermid, S. M., 2005). Consequently,
one needs to mobilize regulatory resources to ignore online messages
and to stay focused on the primary task. As a result, less resources
remain available that can be allocated to the process of accomplishing
one's primary tasks.
Third, besides these potential effects of responsiveness on goal
progress, being responsiveness most likely relates to interpersonal
4SONNENTAG ET AL.

processes at work as well. Prior research demonstrates the presence of
responsiveness norms and expectations (Kalman et al., 2006; Tyler &
Tang, 2003). When the recipient of a message violates such respon-
siveness expectations, this may negatively affect the sender's interper-
sonal evaluation of the recipient. In contrast, high levels of
responsiveness in computer‐mediated communication are positively
associated with interpersonal trust (Kalman et al., 2013). In sum, these
findings suggest that higher levels of responsiveness should not only
increase task performance but also help to achieve interpersonal goals,
such as successful impression management or receiving positive feed-
back from colleagues or customers. These social accomplishments
should further contribute to perceived task accomplishment.
Hypothesis 4. Responsiveness to online messages is
positively related to perceived task accomplishment.
Linking Hypothesis 1 with Hypothesis 4, we propose a positive
indirect effect of perceived interruptions on perceived task accom-
plishment, via responsiveness.
Hypothesis 5. Responsiveness mediates the indirect
positive effect between perceived interruptions by online
messages and perceived task accomplishment.
6|LINKING PERCEIVED INTERRUPTIONS
AND RESPONSIVENESS TO AFFECT
We propose that experiencing interruptions by online messages is
associated with affective consequences. Potential mechanisms linking
perceived interruptions to affective states refer to time pressure,
responsiveness, and perceived task accomplishment. Specifically, we
look at state positive and negative affect as relatively transient mood
states—as opposed to more trait‐like affective dispositions (Watson,
1988). Because high‐activation states are particularly important in
the work context (Warr, Bindl, Parker, & Inceoglu, 2014), we examine
activated negative affect (e.g., feeling distressed, anxious, or angry)
and activated positive affect (e.g., feeling excited, energetic, or alert;
Watson, 1988) as two relatively independent affective states (Diener
& Emmons, 1984).
Affective events theory (AET; Weiss & Cropanzano, 1996) helps
to understand day‐to‐day fluctuations of employee affect at work.
According to AET, affective states at work are a proximal consequence
of events employees encounter and evaluate at work, particularly of
those events that have “affective significance”(p. 31). During the
two decades since the first publication on AET, researchers have
looked at a broad range of different events that have the potential to
cause changes in affect. These events comprise external events caused
by processes in the work environment such as interpersonal conflicts
and problems with tools and equipment (Ohly & Schmitt, 2015), as well
as internal events caused by employees' interpretations and behaviors,
such as task accomplishment satisfaction (Gabriel, Diefendorff, &
Erickson, 2011) or leaders' own transformational behavior (Lanaj,
Johnson, & Lee, 2016).
AET builds on appraisal theories of emotion (for an overview, cf.
Scherer, 1999) that argue that affective states result from an evalua-
tion of whether what is happening is relevant for one's goals and
well‐being (Lazarus, 1991). More specifically, events that are incom-
patible with goal progress and goal achievement elicit negative affec-
tive reactions (Lazarus, 1991). The job‐stress literature has described
job stressors as such negative events that threaten goal progress and
goal achievement and that result in negative affective states such as
anxiety (Rodell & Judge, 2009) or anger (Eatough et al., 2016). Events,
however, that are congruent with one's goals and signal progress
towards one's goals elicit positive affective states (Lazarus, 1991) and
are associated with states such as being “enthusiastic”or “happy”
(Scott, Colquitt, Paddock, & Judge, 2010).
Experiencing time pressure at work implies that one has more to
do than what one can realistically achieve during the available time
(Parker & DeCotis, 1983). Thus, although time pressure might be per-
ceived as a positive challenge (Ohly & Fritz, 2010), it is also associated
with the risk of not attaining all goals relevant at one's workplace.
Hence, time pressure implies that goal achievement is threatened or
needs a high degree of effort investment. Accordingly, in the appraisal
process time pressure is evaluated as incongruent with one's goals and
consequently elicits negative affect. Empirical studies have illustrated
that a high workload and time pressure are in fact positively related
to negative affective states between persons (van Emmerik & Jawahar,
2006) and within persons (Ilies et al., 2007; Story & Repetti, 2006;
Teuchmann, Totterdell, & Parker, 1999). Our Hypothesis 6 builds on
this earlier research:
Hypothesis 6. Time pressure is positively related to
state negative affect.
By linking Hypotheses 2 and 6, we propose an indirect effect from
interruptions by online messages and state negative affect mediated
via time pressure.
Hypothesis 7. T ime pressure mediates the indirect
effect between interruptions by online messages and
state negative affect.
Successfully accomplishing one's tasks is an important goal for
most employees. For instance, performing well is linked to pride, posi-
tive feedback, pay, and promising career opportunities, whereas failure
to perform well is associated with negative feedback and an increased
likelihood of other negative consequences. Hence, perceiving that one
is successful in accomplishing one's tasks implies to make progress
towards relevant goals. According to appraisal theories of emotions
(Lazarus, 1991; Scherer, 1999), this perception should elicit positive
affective states because successful task accomplishment is congruent
with the goal of performing well. Empirical evidence supports this
view: studies demonstrated that goal progress is associated with posi-
tive affect (Alliger & Williams, 1993; Scott et al., 2010). According to
the idea of affect symmetry (Basch & Fischer, 2000), goal progress—
as a positive event—was unrelated to negative affect in these earlier
studies (Alliger & Williams, 1993; Scott et al., 2010).
Hypothesis 8. Perceived task accomplishment is posi-
tively related to positive affect.
Because responsiveness towards online messages it positively
associated with perceived task accomplishment, and perceived task
accomplishment in turn is positively associated with positive affect,
SONNENTAG ET AL.5

responsiveness towards online messages should show a positive asso-
ciation with positive affect as well, with perceived task accomplish-
ment as the mediator.
Hypothesis 9. Perceived task accomplishment mediates
the indirect effect between responsiveness to online mes-
sages and state positive affect.
7|METHOD
7.1 |Procedure and sample
We tested our hypotheses with data from a sample recruited via a Ger-
man online panel (www.consumerfieldwork.de). The panel provider
focuses on online surveys for research purposes and uses established
procedures for ensuring data quality (e.g., identifying careless
responders). In our specific study, the online provider rigorously
screened potential participants on inclusion criteria (working full time,
having access to the Internet at the workplace, being willing to com-
plete daily surveys on 5 days during one workweek). After respondents
agreed to participate, they were directed to a website where they
entered background data (e.g., demographic data) and received daily
e‐mail invitations to complete a daily survey at the end of their work-
day, over the period of 5 days from Monday to Friday. In line with ear-
lier day‐level research (Baethge & Rigotti, 2013; Prem, Ohly, Kubicek,
& Korunka, 2017), we chose a data‐collection period of one workweek.
Daily surveys assessed perceived interruptions, responsiveness,
time pressure, perceived task accomplishment, and state negative
and state positive affect. To minimize the potential influence of the
study procedure on the results, we decided to have the links to the
daily surveys sent to participants at 4:00 p.m. (i.e., towards the end
of the workday of most study participants
1
) and refrained from sending
multiple e‐mails during the day what might have had a stronger impact
on participants' workday and what could have caused additional inter-
ruptions. We instructed participants to answer the daily surveys at the
end of their workdays and allowed them to respond to the surveys
between 4:00 p.m. and 8:30 p.m., rendering it possible for employees
who were still busy working at 4:00 p.m. to respond to the survey later
without having to suspend task accomplishment.
After having excluded careless responders, the data set included
data from a total of 176 persons who completed a total of 870 daily
surveys (out of 880 possible surveys). On a total of 59 out of the
870 days, participants reported that on the specific day they had no
possibility to be online during work time, implying that they could
not provide meaningful answers to the items assessing interruptions
and responsiveness. Accordingly, data from these days were excluded,
leaving a total of 811 days from 174 persons (54% female) for the anal-
ysis. Mean age in this sample was 42.7 years (SD = 9.5) and mean orga-
nizational tenure was 11.2 years (SD = 8.7). Among all participants,
29.9% had a high‐school degree and an additional 27.6% had a univer-
sity degree. Study participants worked in a diverse set of industries and
held a broad variety of different jobs. Specifically, the largest job
categories comprised administrative jobs (39.1%), managerial jobs
(11.5%), service jobs (10.9%), and technical jobs (10.3%).
2
About a
quarter (24.1%) of the sample held a leadership position. Mean con-
tract working time was 38.8 hr per week (SD = 4.7).
7.2 |Measures
We measured our study variables in the daily survey, at the end of the
workday. For all measures, except the affect measures, we used a 5‐
point Likert scale with 1 = I fully disagree; 5=I fully agree. All items were
in German. Table 1 shows the descriptives for the study variables at
the within‐person and the between‐person level.
We assessed perceived interruptions with three items based on
the measure developed by Ten Brummelhuis, Bakker, Hetland, and
Keulemans (2012). Specifically, we adapted the original items by asking
about “e‐mails and other online messages,”instead of “e‐mails”and
“phone calls”:“Today, incoming e‐mails and other online messages
kept me from doing my job,”“Today, e‐mails and other online mes-
sages have reached me at inconvenient moments,”“Today, e‐mails
and other online messages disturbed me in doing my work.”
Cronbach's alphas computed separately for the 5 days of data collec-
tion ranged between .70 and .90 (M= 0.84).
For measuring responsiveness, we formulated three items captur-
ing the day‐specific degree of reacting quickly to job‐related online
messages: “Today I responded immediately to online messages, even
when I was busy with other things,”“When I received an online mes-
sage today, I dropped everything else,”“When I received an online
message today, I was immediately ready to react.”Cronbach's alphas
ranged between .70 and .86 (M= 0.79).
We assessed time pressure with two items based on the scale of
Semmer (1984) and Zapf (1993), adapted for day‐specific assessment
(sample item: “When completing my tasks today, I was required to
work fast”), and one additional item (“I did not have enough time for
my tasks“). Cronbach's alphas ranged between .85 and .89 (M= 0.87).
We measured perceived task accomplishment with four items that
capture subjective goal progress and task‐related success as a task‐
related positive work event (Ohly & Schmitt, 2015): “I have accom-
plished what I had intended to do,”“I realized how well I am doing
my work,”“I completed my tasks successfully.”“I did well in
accomplishing important things.”Cronbach's alphas ranged
between .78 and .85 (M= 0.82).
We assessed state negative and state positive affect at the end of
the workday with items from the Positive and Negative Affect Sched-
ule (Watson, Clark, & Tellegen, 1988). To capture how accumulated
experiences during the workday are related to affective outcomes,
we asked the participants to answer the items with respect to their
momentary affective state at the end of the workday. In addition, by
using a different time reference than used when assessing experiences
during the day, this approach enabled us to reduce common method
bias (Podsakoff, MacKenzie, & Podsakoff, 2012). To minimize partici-
pant burden and in line with earlier research, we used a reduced set
of items (Sonnentag, Binnewies, & Mojza, 2008). For capturing nega-
tive affect, we used the six items “distressed,”“upset,”“irritable,”
1
For 79.2% of our participants, the workdays had ended by 5:00 p.m. or earlier;
and for 94.8%, the workdays had ended by 6:00 p.m. or earlier, suggesting that
the timing of our e‐mail invitations matched most participants' work schedules.
2
Controlling for job category in our analysis did not change the study findings.
6SONNENTAG ET AL.

“nervous,”“jittery,”and “afraid.”Cronbach's alphas ranged between .85
and .87 (M= 0.86). For capturing positive affect, we used the six items
“active,”“interested,”“excited,”“strong,”“inspired,”and “alert.”
Cronbach's alphas ranged between .88 and .91 (M= 0.90). As response
format for all affect items, we used a 5‐point Likert scale ranging from
1=not at all to 5 = very much.
We tested the construct validity of our six study variables with a
two‐level confirmatory factor analysis, implemented in Mplus. A six‐
factor model with all items loading on their respective factors showed
an acceptable fit, χ
2
= 924.68, df = 260, CFI = 0.91, TLI = 0.89,
RMSEA = 0.057, scale correction factor (SCF) = 1.338, and fit the data
better than alternative models, including a five‐factor model with
perceived interruptions and responsiveness items loading on one
common factor, χ
2
= 1154.218, df = 265, CFI = 0.87, TLI = 0.85,
RMSEA = 0.065, SCF = 1.353, Satorra–Bentler χ
2
= 152.101, df =5,
p< .001, and a one‐factor model χ
2
= 5654.156, df = 275, CFI = 0.22,
TLI = 0.15, RMSEA = 0.156, SCF = 1.299, Satorra–Bentler
χ
2
= 9803.407, df = 15, p< .001. Overall, this confirmatory factor
analysis demonstrated that our measures capture distinct constructs.
7.3 |Statistical analysis
Because our day‐level data were nested within persons, we chose a
multilevel approach for data analysis. Specifically, we used multilevel
path modeling, implemented in Mplus. We followed the recommenda-
tions of Preacher, Zyphur, & Zhang, 2010 and specified one overall
multilevel model, including a within‐person (Level 1) and a between‐
person (Level 2) component. This approach has the advantage of
avoiding conflation of within‐person and between‐person effects and
to avoid bias when estimating indirect effects. Accordingly, we
modeled the same paths at the within‐person and the between‐person
level, and allowed correlations between time pressure and perceived
task accomplishment as well as between positive and negative affect
on both levels. At the within‐person level, we specified fixed effects.
8|RESULTS
8.1 |Variance decomposition
Decomposing the variance of our study variables into a within‐person
and a between‐person part showed that between 55% (negative
affect) and 65% (responsiveness) of the variance was attributable to
between‐person variation (cf. intraclass correlations in Table 1), and
between 35% (responsiveness) and 45% (negative affect) of the vari-
ance was attributable to within‐person variation. These figures illus-
trate that all study variables have both a stable and a more variable
component. Accordingly, multilevel modeling is appropriate for analyz-
ing the data.
9|HYPOTHESES TESTING
We tested all hypotheses simultaneously in one overall model with
paths at the between‐person and the within‐person level. In this
model, all hypothesized paths were significant. The overall model fit,
however, was relatively poor, χ
2
= 96.746, df = 14, p< .001, CFI = 0.789,
TLI = 0.548, RMSEA = 0.085. Therefore, we examined if adding a
limited number of paths to the hypothesized model improved model
fit. Specifically, we added a path from responsiveness to time pressure
because being responsive to incoming online messages will consume
time resources and as a consequence overall time pressure during
the day will increase. In addition, because time pressure might not only
be related to negative affect, but also to a low level of positive affect,
and because perceived task accomplishment might not only be related
to positive affect, but also to a low level of negative affect, we
specified also asymmetric paths, in addition to the symmetric paths
from time pressure and perceived task accomplishment to negative
and positive affect, respectively. Overall, this respecified model had a
good fit, χ
2
= 21.513, df =8,p< .01, CFI = 0.966, TLI = 0.871,
RMSEA = 0.046.
Tables 2 and 3 show the estimates of the paths included in this
model. Perceived interruptions were positively related to responsiveness
at the within‐person and the between‐person level, providing support
for Hypothesis 1. Perceived interruptions were positively related to time
pressure, again both at the within‐person and the between‐person
level. Time pressure predicted negative affect at both levels. We
tested the indirect effect from perceived interruptions to negative
affect via time pressure (Preacher et al., 2010), using the MODEL
CONSTRAINT command in Mplus. This indirect effect from perceived
interruptions to negative affect via time pressure was .020, SE = .009;
t= 2.200; p< .05, with a 95% CI of [0.002, 0.037] at the within‐person
level and .115, SE = .039; t= 2.952; p< .01, with a 95% CI of [0.039,
TABLE 1 Means, standard deviations, intraclass coefficients, and zero‐order correlations
MSDMSDICC123456
1. Perceived interruptions 1.92 0.78 1.94 0.94 .61 .54 .46 −.14 .27 −.01
2. Responsiveness 2.40 0.91 2.43 1.06 .65 .63 .14 .13 .03 .16
3. Time pressure 2.40 0.92 2.42 1.09 .62 .56 .20 −.08 .25 −.05
4. Perceived task accomplishment 3.83 0.66 3.84 0.76 .59 −.14 .15 −.06 −.34 .39
5. Negative affect 1.37 0.45 1.35 0.53 .55 .29 .06 .24 −.47 −.21
6. Positive affect 2.57 0.66 2.59 0.82 .56 .05 .21 −.01 .42 −.19
Note. Means and standard deviations at the between‐person level are displayed in Columns 1 and 2; means and standard deviations at the within‐person
level are displayed in Columns 3 and 4. Correlations below the diagonal are between‐person correlations (N= 174), with correlations of |r|≥.20 being sig-
nificant at p< .01 and correlations of |r|≥.15 being significant at p< .05. Correlations above the diagonal are within‐person correlations (811 days) with
correlations of |r|≥.10 being significant at p< .01 and correlations of |r|≥.07 being significant at p< .05. Intraclass Correlation Coefficient (ICC) = Percentage
of variance between persons (ICC = variance between persons / variance between persons + variance within).
SONNENTAG ET AL.7

0.191] at the between‐person level. Taken together, analyses provided
support for Hypotheses 2, 6, and 7 at the within‐person and the
between‐person level.
Responsiveness was positively related to perceived task accom-
plishment at the within‐person and between‐person level, and per-
ceived task accomplishment predicted positive affect at both levels.
The indirect effect from responsiveness to positive affect via task
accomplishment was .035, SE = .013; t= 2.686; p< .01, with a 95%
CI of [0.009, 0.061] at the within‐person level, and .155, SE = .047;
t= 3.313; p< .01, with a 95% CI of [0.063, 0.247] at the between‐per-
son level. Overall, Hypotheses 4, 8, and 9 received support at both
levels of analysis.
As stated in Hypothesis 3, perceived interruptions showed a
negative direct association with perceived task accomplishment,
again at the within‐person and the between‐person level. The indi-
rect effect from perceived interruptions to perceived task accom-
plishment via responsiveness was positive with .038, SE = .014;
t= 2.651; p< .01; 95% CI [0.010, 0.066] at the within‐person level
and with .250, SE = .066; t= 3.814; p< .01, 95% CI [0.122, 0.378]
at the between‐person level. This finding is in line with Hypothesis 5.
Combining the direct negative association from perceived interrup-
tions to perceived task accomplishment with the indirect positive
effect via responsiveness, resulted in a nonsignificant total
effect of −.079, SE = .042, t=−1.876, ns, with a 95% CI
of [−0.162, 0.004] at the within‐person level, and −.130, SE = .070,
t=−1.866, ns, 95% CI of [−0.266, 0.007] at the between‐
person level.
In addition to the hypotheses tested, analyses showed that
responsiveness was negatively related to time pressure between per-
sons, but not within persons. Time pressure in turn was negatively
related to positive affect at the within‐person level, but not at the
between‐person level. Perceived task accomplishment was negatively
related to negative affect at both levels.
10 |DISCUSSION
Our daily‐survey study showed that more frequent perceived
interruptions by online messages predicted higher time pressure that
TABLE 2 Unstandardized coefficients from multilevel path model predicting responsiveness, time pressure, and perceived task accomplishment
Estimate SE z Estimate SE z Estimate SE z
Responsiveness Time pressure Perceived task accomplishment
Between level
Intercept 0.932 0.152 6.114*** 1.331 0.200 6.662*** 3.777 0.159 23.756***
Perceived interruptions 0.764 0.073 10.483*** 0.955 0.109 8.762*** −0.380 0.095 −3.981***
Responsiveness −0.320 0.100 −3.199** 0.327 0.073 4.470***
Residual variance 0.412 0.062 6.681*** 0.414 0.070 5.936*** 0.298 0.040 7.364***
Within level
Perceived interruptions 0.353 0.062 5.690*** 0.260 0.064 4.077*** −0.117 0.042 −2.788**
Responsiveness −0.074 0.052 −1.404 0.108 0.037 2.932**
Residual variance 0.345 0.029 11.998*** 0.420 0.034 12.493*** 0.237 0.021 11.056***
Note. Estimates are unstandardized, resulting from one overall analysis including the prediction of responsiveness, time pressure, and perceived task
accomplishment as well as negative and positive affect in one model.
*p< .05. **p< .01. ***p< .001.
TABLE 3 Unstandardized coefficients from multilevel path model predicting negative affect and positive affect
Estimate SE z Estimate SE z
Negative affect Positive affect
Between level
Intercept 2.207 0.312 7.080*** 0.690 0.361 1.912
Time pressure 0.120 0.036 3.327** 0.028 0.058 0.484
Perceived task accomplishment −0.297 0.069 −4.271*** 0.475 0.089 5.353***
Residual variance 0.114 0.019 5.977*** 0.293 0.040 7.356***
Within level
Time pressure 0.076 0.026 2.870** −0.081 0.035 −2.279*
Perceived task accomplishment −0.157 0.038 −4.129*** 0.324 0.047 6.883***
Residual variance 0.119 0.015 7.769*** 0.263 0.017 15.463***
Note. Estimates are unstandardized, resulting from one overall analysis including the prediction of responsiveness, time pressure, and perceived task accom-
plishment as well as negative and positive affect in one model.
*p< .05. **p< .01. ***p< .001.
8SONNENTAG ET AL.

in turn predicted increased negative affect. Experiencing more
frequent interruptions was positively related to higher responsiveness
to online messages. Higher responsiveness, in turn, predicted higher
perceived task accomplishment, resulting in an indirect positive effect
of perceived interruptions on perceived task accomplishment via
responsiveness. The total effect from perceived interruptions to
perceived task accomplishment was not significant. Perceived task
accomplishment predicted increased positive affect.
These results shed a differentiated picture on the role of online
messages in employees' daily working life. On the one hand, the
finding that experiencing interruptions is positively related to time
pressure and that it shows a direct negative association with perceived
task accomplishment emphasizes the stressor perspective on online
messages. According to this perspective, feeling interrupted is stressful
and has a negative effect on the work process (Barber & Santuzzi,
2015; Marulanda‐Carter & Jackson, 2012). Experiencing interruptions
may trigger feelings of time scarcity and increased workload because
additional tasks intrude into one's daily schedule, increasing negative
affect and making it difficult to complete all preplanned tasks.
On the other hand, our study shows that feeling interrupted by
online messages might not always be detrimental for the work process.
Responding to incoming online messages contributes to perceived task
accomplishment and helps to make progress towards one's work goals.
This pattern of findings can be explained from a multiple‐goal
perspective. In many situations, dealing with online messages refers
to a multiple‐goal situation where employees have to align different
goals, prioritize them, and shield high‐priority goals from interferences
(Unsworth, Yeo, & Beck, 2014). For instance, in a situation in which the
incoming message is well aligned with the higher‐order overall goal of
an activity, the interruption by the online message—although it may
disrupt the actual process of working on another part of the overall
task—will be welcomed and responsiveness will be high, resulting in
perceived task accomplishment.
In many organizational settings, dealing with online messages is a
core aspect of an employee's job. We identified responsiveness as an
important construct pointing to the positive potential of online
messages in these settings. Being responsive contributes to the
perception of getting things done, which in turn is positively
associated with positive affect. This process might not only operate
within the person receiving and responding to the online messages.
Reacting quickly often also implies that the sender of the first
message gets a response to his or her message within reasonable time
(Simon & Pritchard, 2015), making it more likely that the original
sender reacts in a positive way, in turn fostering the perception of
good performance and positive affect.
Although it is difficult to imagine how to be responsive in
everyday working life without having been interrupted in the first
place, of course, the positive association between interruptions and
responsiveness is not a fully deterministic one. There might be days
when highly urgent primary work tasks make it difficult to respond
to incoming e‐mails. In such situations, an employee may deliberately
decide not trying to catch up with e‐mail, but to abandon responsive
behavior altogether.
Our findings also contribute to the broader research on ICT use
and affect at work. In addition to findings from earlier research that
has linked job‐related electronic communication at home to affective
processes at home (Butts et al., 2015; Ohly & Latour, 2014), our
study showed that experiences associated with e‐mail and other
online communication at work matter for employee affect, with
perceived interruptions being indirectly related to negative affect
and responsiveness being indirectly related to positive affect.
Importantly, the total indirect effect of perceived interruptions with
positive affect was not significant. Thus, perceived interruptions
seem to increase negative states such as distress and irritation, but
do not compromise feelings of interest and alertness. Of course,
experiencing e‐mail interruptions capture only a small aspect of ICT
use at work, and our focus on these interruptions cannot explain
the full range of phenomena experienced when communicating
electronically. Our study, however, demonstrates that specific
aspects of ICT use may constitute important affective events that
have implications for how employees experience their workdays.
Overall, findings at the between‐person and within‐person level
are very similar, reflecting both between‐person differences and
within‐person fluctuation. Thus, perceived interruptions and respon-
siveness can predict between‐person differences in time pressure
and perceived task accomplishment as well as between‐person
differences in affect at the end of the workday. These between‐person
findings might also reflect differences between jobs. For instance, in
some jobs “constant connectivity”(Wajcman & Rose, 2011, p. 941) is
more required than in other jobs. Importantly, even when taking these
between‐person and between‐job differences into account, the
within‐person findings demonstrate that fluctuations of perceived
interruptions and responsiveness matter for employees' everyday
work experiences. The day‐specific level of perceived interruptions
and responsiveness can explain why on some days employees
experience more time pressure and more task accomplishment than
on other days, and subsequently, why affect at the end of the workday
differs from day to day.
10.1 |Limitations and directions for future research
The contribution of our study must be seen in light of some limita-
tions. First, we included data from one measurement point per day,
implying that our findings reflect patterns of the relationships for
the whole workday. One might argue that such a day‐level approach
misses some of the complexities inherent in interruptions by online
messages because the attributes (e.g., sender, content, goal‐rele-
vance, and emotional tone) of specific messages are neglected and
that an event‐level approach might have been more informative.
Despite the potential insights an event‐level approach could offer,
the day‐level approach has some advantages over an event‐level
approach. First, some of the processes we were interested in might
not have become evident at the event level but may have evolved
over the course of the workday and were, therefore, best captured
at the day level. For instance, increased time pressure due to per-
ceived interruptions might not have become obvious at the very
moment when the interruption occurred (or shortly thereafter), but
might have unfolded over the course of the full day when
employees, for instance, fell more and more behind with their work.
An event‐level approach would have missed such lagged effects. In
SONNENTAG ET AL.9

addition, assessing interruptions and their correlates at the event
level would have been disruptive in itself, rendering the data‐collec-
tion procedure highly reactive (Barta, Tennen, & Litt, 2012). This
would have been particularly troublesome in the context of this
study because interruptions are frequent events (Wajcman & Rose,
2011). Event‐level assessments, however, are better suited for cap-
turing events that occur at a lower frequency (Reis & Gable, 2000).
Nevertheless future studies might want to adopt a more fine‐
grained approach, looking at within‐day fluctuations of interruptions
and responsiveness. For instance, there might be times during the
day when the level of interruptions and responsiveness are particularly
high, being compensated by more quiet times. Moreover, one could
argue that study participants might have reconstructed their answers
in order to report a coherent picture of the workday. For instance, they
might have reported a level of responsiveness that corresponded to
their perceived stress level. Although we cannot completely rule out
this possibility, the finding that responsiveness was unrelated to time
pressure at the within‐person level speaks against this possibility.
Second, our specific measure of perceived task accomplishment
was rather broad and assessed overall task accomplishment without
differentiating between different types of tasks. For instance, it
might be that being responsive helps to accomplish some of the
tasks, but hinders goal progress with respect to other tasks. Thus,
our findings reflect the perceived “net gain”from being responsive,
but do not provide a detailed picture about which tasks benefit from
being responsive and which tasks suffer from a high level of respon-
siveness. Future studies might want to address different types of
tasks (e.g., complex and cognitively demanding tasks vs. simple and
routine tasks) and might want to examine the relevance of the
online message for the specific task. Relatedly, we used a general
measure of interruptions by online messages. Recently, Addas and
Pinsonneault (2015) have presented a taxonomy comprising various
types of interruptions. Future studies could provide more detailed
insights into the effects of interruptions when differentiating
between various types of interruptions.
Third, our interruption measure did not only ask about the occur-
rence of an interruption but also captured the appraisal of the interrup-
tion as a disturbance of the usual work process—what might have lead
to an inflation of the empirical associations with other constructs
assessed in our study. The wordings of our interruption items, how-
ever, were rather close to the definition of an intrusion as suggested
by Jett and George (2003, p. 494). Importantly, the interruption items
did not refer to the experience of time pressure, implying that the
empirical association between our interruption measure and time pres-
sure cannot be attributed to item wording.
Fourth, we assessed our data with self‐report measures. There-
fore, common‐method bias cannot be ruled out completely (Podsakoff
et al., 2012). However, some of our constructs (e.g., negative and
positive affect) and the associations between them (e.g., between
perceived task accomplishment and affect) can be best captured by
self‐reports (Haefel & Howard, 2010) and person‐level variables as
drivers of the within‐person findings can be ruled out because of the
daily‐survey design. Nevertheless, future studies should strive to
assess at least some of the construct with other methods. For instance,
they may want to assess objective data for interruptions and
responsiveness and could test the congruency between objective and
perceptual measures.
Future research should not only try to overcome the limitations of
our present study. Scholars might also want to address new research
questions emerging from our findings. First, it could be informative
to study interruptions and responsiveness at the event level. For
instance, it would be worthwhile to examine specific attributes of
every online message (e.g., sender, content of message, amount of
work required, emotional tone) and link these features to responsive-
ness and subsequent affective states. Because assessing event‐level
interruptions and responses within a field setting might be highly reac-
tive and might influence ongoing work processes, researchers might
opt for a laboratory approach, possibly using an experimental design
(cf. Zijlstra, Roe, & Leonora, 1999).
Second, future research might want to shed more light on the pre-
dictors of responsiveness to online messages. For instance, individual
(e.g., extraversion) and organizational factors (e.g., an organization's
e‐mail policy), as well as task features (e.g., importance and urgency)
and aspects of the sender–receiver relationship (e.g., status differ-
ences) might play a role here. There might also be complex mecha-
nisms operating at the day level that influence responsiveness. When
working towards a deadline, one might be selectively responsive only
to those messages that help in achieving the deadline. If task complex-
ity increases, however, there might not be sufficient cognitive
resources to decide if a message helps in achieving the deadline or not.
Third, future studies could explore the boundary conditions of our
findings and might want to examine when responsiveness is particu-
larly important and if there are situations in which responsiveness hin-
ders task accomplishment. For instance, the relationship between
responsiveness and task accomplishment may change when tasks are
more aversive. Instead of helping task performance, responsiveness
may turn into a dysfunctional form of procrastination when used as
an excuse to delay aversive primary tasks (Sirois & Pychyl, 2013). In
fact, studies suggest that media content in general and online commu-
nication in particular are frequently used “tools”for procrastination
(Reinecke & Hofmann, 2016; Vitak, Crouse, & LaRose, 2011). This sug-
gests that—depending on the task characteristics—responsiveness may
not always contribute to task accomplishment. Similarly, research
might want to address factors that have the potential to moderate
the relationship between interruptions, time pressure, and subsequent
negative affect. For instance, during a low‐workload day and when
working on easy tasks, interruptions may be less aversive than on a
high‐workload day and when working on complex tasks. Moreover,
taking interruptions into account already when planning the workday
may attenuate the association between interruptions and time
pressure. Of course, these issues are not only relevant for e‐mail, but
also for social media used at work (Leonardi, 2014; McFarland &
Ployhart, 2015).
In our study, we have examined a person's overall degree of
responsiveness on a workday, neglecting that a person may show dif-
ferent response patterns towards different online communication part-
ners. Communication accommodation theory describes that people do
not speak in the same way to everybody, but that they adjust their way
of speaking to a specific communication partner in order to match this
other person's communication behavior (Gallois & Giles, 2015).
10 SONNENTAG ET AL.

Accordingly, employees may not only adjust how they write their e‐
mails to their specific communication partner (Riordan, Markman, &
Stewart, 2012), but also the degree to which they are responsive. For
instance, employees may react more quickly to an e‐mail from a person
who immediately answers e‐mails than to an e‐mail from a slow
responder. In addition, the status of the sender might play an impor-
tant role for responsiveness to a specific message. Thus, future
research may want to develop a more differentiated perspective on
e‐mail interruptions and responsiveness at work.
Moreover, in our study we did not distinguish between different
online media. Therefore, future studies might want to pay more
attention to the characteristics of specific online media and how per-
sons customize their specific features. For instance, notifications
about incoming messages may increase e‐mail's intrusiveness and
may increase responsiveness (Hanrahan, Pérez‐Quinones, & Martin,
2016). Moreover, instant messages might be seen as more urgent
than e‐mails. Instant messaging systems often provide the sender
with feedback on when their message has been read, potentially
creating a higher pressure for the receiver to respond immediately
(Mai, Freudenthaler, Schneider, & Vorderer, 2015). Social‐media
implementations may create more communication visibility towards
third parties (Leonardi, 2014), which in turn can influence reasons
for interrupting others and being responsive.
Future studies on online messages will face particular methodo-
logical challenges. Researchers will need to find innovative ways to
separate the measurement process from the processes they are
interested in. For instance, one might argue that our e‐mail prompts
sent in the afternoon interfered with the interruptions we sought to
measure. Also, future studies may want to use other devices, such as
palm pilots or smartphones for sending the reminders and assessing
the data.
10.2 |Practical implications
Although more research is needed about the factors that might act as
moderators of the association between interruptions and time
pressure and between responsiveness and perceived task accomplish-
ment, our study offers suggestions about how to deal with online
messages within organizations. First, because perceived interruptions
show an indirect relationship with negative affect and because nega-
tive affect can have adverse consequences at work (Rodell & Judge,
2009; Scott & Barnes, 2011), it is important that perceived interrup-
tions by e‐mails and other online messages are limited. To reduce the
experience of being interrupted, employees can schedule some time
periods during the day when they will not be disturbed, for instance
by closing specific programs or by setting mobile devices to airplane
mode. Such steps, however, have their downsides because they com-
promise online collaborations. On the basis of research on job control
(Parker, 2014), it is possibly most important that employees have the
choice to disconnect from online devices from time to time in order
to be able to work on other tasks. There might be instances when
experiencing this control is even more important than actually
disconnecting (Glass, Singer, Leonard, Krantz, & Cohen, 1973).
Research suggests that “batching e‐mail”(i.e., processing e‐mail only
at specific times of the day) is associated with perceived productivity,
although physiological strain indicators seem not to reflect any benefit
of this type of e‐mail‐management behavior (Mark et al., 2016). In
addition, technical solutions for decreasing perceived interruptions,
for instance, by automatically prioritizing incoming e‐mails can be an
additional useful step to minimizing experienced disruptions of the
actual work process (Kobayashi, Tanaka, Aoki, & Fujita, 2015).
Second, when being online and after having been interrupted by
an online message, a high responsiveness should be encouraged.
Because leaving tasks unfinished can have undesirable consequences
(Syrek & Antoni, 2014; Zeigarnik, 1927), it seems particularly important
that employees do not only read incoming online messages, but that
they decide quickly if an answer is needed, how they will respond to
each specific message, and actually prepare the answer. Immediately
focusing on incoming e‐mails, however, postpones work on other
tasks, potentially leading to switching costs with respect to other
duties. Thus, careful prioritizing of different ongoing tasks is important.
In addition, there will always be some messages that need deep delib-
erate processing of the information, and responding without reflection
can have detrimental consequences. The point here is that being
responsive when it is appropriate will foster the experience of
accomplishing one's tasks and making progress towards one's goals
at work. Being responsive to e‐mails that are related to important work
goals may even be experienced as “small wins”(Amabile & Kramer,
2011) that in turn energize employees during their day at work.
Third, building on the affective gains of being responsive,
employees could be encouraged to capitalize on their responsiveness
as a positive self‐generated event. For instance, employees could
deliberately appreciate and positively reflect on having been respon-
sive to e‐mails that are linked to important work goals (Bono, Glomb,
Shen, Kim, & Koch, 2013). Such an approach may help to keep positive
affect high, even during a long day at work.
11 |CONCLUSION
Taken together, our study contributes to a new perspective on
interruptions at work where “constant connectivity”(Wajcman & Rose,
2011, p. 941) and engagement in multiple conversations at the same
time (Cameron & Webster, 2013) become more and more widespread.
Accordingly, for many employees frequent interruptions are a recog-
nized part of typical workdays. Our findings demonstrate that interrup-
tions are still associated with increased strain levels, but they also
suggest that being responsive to interruptions stimulates the feeling
of getting work done. Thus, the impact of e‐mail and online technology
does not only influence communication processes itself (Skovholt,
Grønning, & Kankaanranta, 2014; Sproull & Kiesler, 1986), but seems
to have a profound impact on employees' “lived‐through experience”
(Weiss & Rupp, 2011, p. 83) during their entire day at work.
ORCID
Sabine Sonnentag http://orcid.org/0000-0002-9464-4653
REFERENCES
Addas, S., & Pinsonneault, A. (2015). The many faces of information tech-
nology interruptions: A taxonomy and preliminary investigation of
SONNENTAG ET AL.11

their performance effects. Information Systems Journal,25, 231–273.
https://doi.org/10.1111/isj.12064
Alliger, G. M., & Williams, K. J. (1993). Using signal‐contingent experience
sampling methodology to study work in the field: A discussion and illus-
tration examining task perceptions and mood. Personnel Psychology,46,
525–549. https://doi.org/10.1111/j.1744‐6570.1993.tb00883.x
Altmann, E. M., & Trafton, J. G. (2002). Memory for goals: An activation‐
based model. Cognitive Science,26,39–83. https://doi.org/10.1016/
S0364‐0213(01)00058‐1
Amabile, T. M., Barsade, S. G., Mueller, J. S., & Staw, B. M. (2005). Affect
and creativity at work. Administrative Science Quarterly,50, 367–403.
https://doi.org/10.2189/asqu.2005.50.3.367
Amabile, T. M., & Kramer, S. J. (2011). The power of small wins. Harvard
Business Review,89(5), 70–80.
Ashkanasy, N. M., & Dorris, A. D. (2017). Emotions in the workplace. Annual
Review of Organizational Psychology and Organizational Behavior,4,
67–90. https://doi.org/10.1146/annurev‐orgpsych‐032516‐113231
Avrahami, D., & Hudson, S. E. (2006). Responsiveness in instant messaging:
Predictive models supporting inter‐personal communication. In
Proceedings of the SIGCHI conference on human factors in computing
systems (CHI ‘06) (pp. 731–740). New York: ACM Press.
Baethge, A., & Rigotti, T. (2013). Interruptions to workflow: Their relation-
ship with irritation and satisfaction with performance, and the
mediating roles of time pressure and mental demands. Work & Stress,
27,43–63. https://doi.org/10.1080/02678373.2013.761783
Baethge, A., Rigotti, T., & Roe, R. A. (2015). Just more of the same, or differ-
ent? An integrative theoretical framework for the study of cumulative
interruptions at work. European Journal of Work and Organizational
Psychology,24, 308–323. https://doi.org/10.1080/1359432X.2014.
897943
Bailey, B. P., & Konstan, J. A. (2006). On the need for attention‐aware sys-
tems: Measuring effects of interruption on task performance, error rate,
and affective state. Computers in Human Behavior,22, 685–708.
https://doi.org/10.1016/j.chb.2005.12.009
Barber, L. K., & Santuzzi, A. M. (2015). Please respond ASAP: Workplace
telepressure and emplyoee recovery. Journal of Occupational Health
Psychology,20, 172–189. https://doi.org/10.1037/a0038278
Barley, S. R., Meyerson, D. E., & Grodal, S. (2011). E‐mail as a source and
symbol of stress. Organization Science,22, 887–906. https://doi.org/
10.1287/orsc.1100.0573
Barnes, C. M., Lucianetti, L., Bhave, D. P., & Christian, M. S. (2015). You
wouldn't like me when I'm sleepy: Leaders' sleep, daily abusive supervi-
sion, and work unit engagement. Academy of Management Journal,58,
1419–1437. https://doi.org/10.5465/amj.2013.1063
Barsade, S. (2002). The ripple effects: Emotional contagion and its influence
on group behavior. Administrative Science Quarterly,47, 644–675.
https://doi.org/10.2307/3094912
Barta, W. D., Tennen, H., & Litt, M. D. (2012). Measurement reactivity in
diary research. In M. R. Mehl, & T. S. Conner (Eds.), Handbook of
research methods for studying daily life (pp. 108–123). New York:
Guilford.
Basch, J., & Fischer, C. D. (2000). Affective events‐emotion matrix: A clas-
sification of work events and associated emotions. In N. M. Ashkanasy,
C. E. J. Härtel, & W. J. Zerbe (Eds.), Emotions in the workplace: Research,
theory and practice (pp. 36–48). Westport, CT: Quorum.
Beal, D. J. (2015). ESM 2.0: State of the art and future potential of
experience sampling methods in organizational research. Annual Review
of Organizational Psychology and Organizational Behavior,2, 383–407.
https://doi.org/10.1146/annurev‐orgpsych‐032414‐111335
Beal, D. J., & Ghandour, L. (2011). Stability, change, and the stability of
change in daily workplace affect. Journal of Organizational Behavior,
32, 526–546. https://doi.org/10.1002/job.713
Beal, D., Weiss, J., Barros, H. M., & E., & MacDermid, S. M. (2005). An epi-
sodic process model of affective influences on performance. Journal of
Applied Psychology,90, 1054–1068. https://doi.org/10.1037/0021‐
9010.90.6.1054
Bono, J. E., Glomb, T. M., Shen, W., Kim, E., & Koch, A. J. (2013). Building
positive resources: Effects of positive events and positive reflection
on work‐stress and health. Academy of Management Journal,56,
1601–1627. https://doi.org/10.5465/amj.2011.0272
Brown, R., Duck, J., & Jimmieson, N. (2014). E‐mail in the workplace: The
role of stress appraisals and normative response pressure in the rela-
tionship between E‐mail stressors and employee strain. International
Journal of Stress Management,21, 325–347. https://doi.org/10.1037/
a0037464
Butts, M. M., Becker, W. J., & Boswell, W. R. (2015). Hot buttons and time
sinks: The effects of electronic communication during nonwork time on
emotions and work‐nonwork conflict. Academy of Management Journal,
58, 763–788. https://doi.org/10.5465/amj.2014.0170
Cameron, A.‐F., & Webster, J. (2013). Multicommunicating: Juggling multi-
ple conversations in the workplace. Information Systems Research,24,
352–371. https://doi.org/10.1287/isre.1120.0446
Carton, A. M., & Aiello, J. R. (2009). Control and anticipation of social inter-
ruptions: Reduced stress and improved task performance. Journal of
Applied Social Psychology,39, 169–185. https://doi.org/10.1111/
j.1559‐1816.2008.00434.x
Chui, M., Manyika, J., Bughin, J., Dobbs, R., Roxburgh, C., Sarrazin, H., …
Westergren, W. (2012). The social economy: Unlocking value and productivity
through social technologies. Seoul: McKinsey Global Institute.
Colbert, A. E., Yee, N., & George, G. (2016). The digital workforce and the
workplace of the future. Academy of Management Journal,59,
371–379. https://doi.org/10.5465/amj.2016.4003
Dabbish, L. A., Kraut, R. E., Fussell, S., & Kiesler, S. (2005). Understanding
email use: Predicting action on a message. In Proceedings of the SIGCHI
conference on human factors in computing systems (CHI ‘05) (pp.
691–700). New York: ACM Press.
Dalal, R. S., Bhave, D. P., & Fiset, J. (2014). Within‐person variability in job
performance: A theoretical review and research agenda. Journal of
Management,40, 1396–1436. https://doi.org/10.1177/0149206314
532691
DeVoe, S. E., & Pfeffer, J. (2011). Time is tight: How higher economic value
of time increases feelings of time pressure. Journal of Applied Psychol-
ogy,96, 665–676. https://doi.org/10.1037/a0022148
Diener, E., & Emmons, R. A. (1984). The independence of positive and neg-
ative affect. Journal of Personality and Social Psychology,47,
1105–1117. https://doi.org/10.1037/0022‐3514.47.5.1105
Dodhia, R. M., & Dismukes, R. K. (2009). Interruptions create prospective
memory tasks. Applied Cognitive Psychology,23,73–89. https://doi.
org/10.1002/acp.1441
Eatough, E. M., Meier, L. L., Igic, I., Elfering, A., Spector, P. E., & Semmer, N.
K. (2016). You want me to do what? Two daily diary studies of illegiti-
mate tasks and employee well‐being. Journal of Organizational
Behavior,37, 108–127. https://doi.org/10.1002/job.2032
Einstein, G. O., McDaniel, M. A., Williford, C. L., Pagan, J. L., & Dismukes, R.
K. (2003). Forgetting of intentions in demanding situations is rapid.
Journal of Experimental Psychology: Applied,9, 147–162. https://doi.
org/10.1037/1076‐898X.9.3.147
El Ouirdi, A., El Ouirdi, M., Segers, J., & Henderickx, E. (2015). Employees'
use of social media technologies: A methodological and thematic
review. Behaviour & Information Technology,34, 454–464. https://doi.
org/10.1080/0144929X.2015.1004647
van Emmerik, I. H., & Jawahar, I. M. (2006). The independent relationships
of objective and subjective workload with couples' mood. Human Rela-
tions,59, 1371–1392. https://doi.org/10.1177/0018726706071649
Eyrolle, H., & Cellier, J.‐M. (2000). The effects of interruptions in work
activity: Field and laboratory results. Applied Ergonomics,31, 537–543.
https://doi.org/10.1016/S0003‐6870(00)00019‐3
Fay, D., & Sonnentag, S. (2012). Within‐person fluctuations of proactive
behavior: How affect and experienced competence regulate work
12 SONNENTAG ET AL.

behavior. Human Performance,25,72–93. https://doi.org/10.1080/
08959285.2011.631647
Fisher, C. D., & Noble, C. S. (2004). A within‐person examination of corre-
lates of performance and emotions while working. Human Performance,
17, 145–168. https://doi.org/10.1207/s15327043hup1702_2
Gabriel, A. S., Diefendorff, J. M., & Erickson, R. J. (2011). The relations of
daily task accomplishment satisfaction with changes in affect: A
multilevel study in nurses. Journal of Applied Psychology,96,1095–1104.
https://doi.org/10.1037/a0023937
Gallois, C., & Giles, H. (2015). Communication accommodation theory. The
International Encyclopedia of Language and Social Interaction,1–18.
Gephart, R. P. J. (2002). Introduction to the brave new workplace: Organi-
zational behavior in the electronic age. Journal of Organizational
Behavior,23, 327–344. https://doi.org/10.1002/job.143
Glass, D. C., Singer, J. E., Leonard, H. S., Krantz, D., & Cohen, S. (1973).
Perceived control of aversive stimulation and the reduction of stress
response. Journal of Personality,41, 577–595. https://doi.org/
10.1111/j.1467‐6494.1973.tb00112.x
González, V. M., & Mark, G. (2004). "Constant, constant, multi‐tasking cra-
ziness": Managing multiple working sheres. In CHI 2004 (pp. 113–120).
Vienna, Austria: CHI Letters.
Grebner, S., Semmer, N. K., Lo Faso, L., Gut, S., Kälin, W., & Elfering, A.
(2003). Working conditions, well‐being, and job‐related attitudes
among call centre agents. European Journal of Work and Organizational
Psychology,12, 341–365. https://doi.org/10.1080/13594320344
000192
Gupta, A., Li, H., & Sharda, R. (2013). Should I send this message? Under-
standing the impact of interruptions, social hierarchy and perceived
task complexity on user performance and perceived workload. Decision
Support Systems,55, 135–145. https://doi.org/10.1016/j.dss.
2012.12.035
Haefel, G. J., & Howard, G. S. (2010). Self‐report: Psychology's four‐letter
word. American Journal of Psychology,123, 181–188. https://doi.org/
10.5406/amerjpsyc.123.2.0181
Hanrahan, B. V., Pérez‐Quinones, M. A., & Martin, D. (2016). Attending to
email. Interacting with Computers,28, 253–272. https://doi.org/
10.1093/iwc/iwu048
Harris, M. M., & Schaubroeck, J. (1988). A meta‐analysis of self‐supervisor,
self‐peer, and peer‐supervisor ratings. Personnel Psychology,41,43–62.
https://doi.org/10.1111/j.1744‐6570.1988.tb00631.x
Hodgetts, H. M., & Jones, D. M. (2006). Contextual cues aid recovery from
interruption: The role of associative activation. Journal of Experimental
Psychology: Learning, Memory, and Cognition,32, 1120–1132. https://
doi.org/10.1037/0278‐7393.32.5.1120
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 behavior
at work. Journal of Applied Psychology,92, 1368–1379. https://doi.org/
10.1037/0021‐9010.92.5.1368
Ilies, R., Scott, B. A., & Judge, T. A. (2006). The interactive effects of per-
sonal traits and experienced states on intraindividual patterns of
citizenship behavior. Academy of Management Journal,49, 561–575.
https://doi.org/10.5465/AMJ.2006.21794672
Jackson, T., Dawson, R., & Wilson, D. (2001). The cost of email interruption.
Journal of Systems and Information Technology,5,81–92.
Jett, Q. R., & George, J. M. (2003). Work interrupted: A closer look at the
role of interruptions in organizational life. Academy of Management
Journal,28, 494–507. https://doi.org/10.5465/AMR.2003.10196791
Jimmieson, N. L., & Terry, D. J. (1997). Responses to an in‐basket activity:
The role of work stress, behavioral control, and informational control.
Journal of Occupational Health Psychology,2,72–83. https://doi.org/
10.1037/1076‐8998.2.1.72
Kalman, Y. M., & Rafaeli, S. (2011). Online pauses and silence: Chronemic
expectancy violations in written computer‐mediated communication.
Communication Research,38,54–69. https://doi.org/10.1177/
0093650210378229
Kalman, Y. M., & Ravid, G. (2015). Filing, piling, and everything in between:
The dynamics of E‐mail inbox management. Journal of the Association
for Information Science and Technology,66, 2540–2552. https://doi.
org/10.1002/asi.23337
Kalman, Y. M., Ravid, G., Raban, D. R., & Rafaeli, S. (2006). Pauses and
response latencies: A chronemic analysis of asynchronous CMC. Journal
of Computer‐Mediated Communication,12(1), 1–23. https://doi.org/
10.1111/j.1083‐6101.2006.00312.x
Kalman, Y. M., Scissors, L. E., Gill, A. J., & Gergle, D. (2013). Online
chronemics convey social information. Computers in Human Behavior,
29, 1260–1269. https://doi.org/10.1016/j.chb.2012.12.036
Kalmbach, D. A., Pillai, V., Roth, T., & Drake, C. L. (2014). The interplay
between daily affect and sleep: A 2‐week study of young women. Jour-
nal of Sleep Research,23, 636–645. https://doi.org/10.1111/jsr.12190
Kobayashi, Y., Tanaka, T., Aoki, K., & Fujita, K. (2015). E‐mail delivery medi-
ation system based on user interruptibility. In M. Kurosu (Ed.), Human–
computer interaction: Design and evaluation: 17th International Confer-
ence, HCI International 2015, Los Angeles, CA, USA, August 2–7, 2015,
Proceedings, Part I (pp. 370–380). Cham: Springer International
Publishing.
Lam, C. F., Wan, W. H., & Roussin, C. J. (2016). Going the extra mile and
feeling energized: An enrichment perspective of organizational citizen-
ship behaviors. Journal of Applied Psychology,101, 379–391. https://
doi.org/10.1037/apl0000071
Lanaj, K., Johnson, R. E., & Lee, S. M. (2016). Benefits of transformational
behaviors for leaders: A daily investigation of leader behaviors and
need fulfillment. Journal of Applied Psychology,101, 237–251. https://
doi.org/10.1037/apl0000052
Lazarus, R. S. (1991). Emotion and adaptation. New York, NY: Oxford Uni-
versity Press.
Leonardi, P. M. (2014). Social media, knowledge sharing, and innovation:
Toward a theory of communication visibility. Informations Systems
Research,25, 796–816. https://doi.org/10.1287/isre.2014.0536
Madan, C. R., Spetch, M. L., & Ludvig, E. A. (2015). Rapid makes risky: Time
pressure increases risk seeking in decisions from experience. Journal of
Cognitive Psychology,27, 921–928. https://doi.org/10.1080/
20445911.2015.1055274
Mai, L. M., Freudenthaler, R., Schneider, F. M., & Vorderer, P. (2015). "I
know you've seen it!" Individual and social factors for users' chatting
behavior on Facebook. Computers in Human Behavior,49, 296–302.
https://doi.org/10.1016/j.chb.2015.01.074
Mark, G., González, V. M., & Harris, J. (2005). No task left behind? Examin-
ing the nature of fragmented work. In CHI 2005 (pp. 321–330). Oregon:
Portland.
Mark, G., Gudith, D., & Klocke, U. (2008). The cost of interrupted work:
More speed and stress. CHI '08. Proceedings of the SIGCHI Conference
on Human Factors in Computing Systems, 107–110.
Mark, G., Iqbal, S. T., Czerwinski, M., Johns, P., & Sano, A. (2016). Email
duration, batching and self‐interruptions: Patterns of email use on pro-
ductivity and stress. CHI'16, May 07–12, 2016, San Jose, CA, USA.
https://doi.org/10.1145/2858036.2858262
Marulanda‐Carter, L., & Jackson, T. W. (2012). Effects of e‐mail addiction
and interruptions on employees. Journal of Systems and Informaiton
Technology,14,82–94. https://doi.org/10.1108/13287261211221146
Mazmanian, M. (2013). Avoiding the trap of constant connectivity: When
congruent frames allow for heterogeneous practices. Academy of
Management Journal,56, 1225–1250. https://doi.org/10.5465/
amj.2010.0787
McFarland, L. A., & Ployhart, R. E. (2015). Social media: A conceptual frame-
work to guide research and practice. Journal of Applied Psychology,100,
1653–1677. https://doi.org/10.1037/a0039244
SONNENTAG ET AL.13

Middleton, C. A., & Cukier, W. (2006). Is mobile email functional or dysfunc-
tional? Two perspectives on mobile email usage. European Journal of
Information Systems,15, 252–260.
Monk, C. A., Trafton, J. G., & Boehm‐Davis, D. A. (2008). The effect of
interruption duration and demand on resuming suspended goals. Jour-
nal of Experimental Psychology: Applied,14, 299–313. https://doi.org/
10.1037/a0014402
O'Conaill, B., & Frohlich, D. (1995). Timespace in the workplace: Dealing
with interruptions. In CHI (pp. 262–263). CO: Denver.
Ohly, S., & Fritz, C. (2010). Work characteristics, challenge appraisal, crea-
tivity and proactive behavior: A multi‐level study. Journal of
Organizational Behavior,31, 543–565. https://doi.org/10.1002/job.633
Ohly, S., & Latour, A. (2014). Use of smartphones for work and well-being
in the evening: The role of autonomous and controlled motivation.
Journal of Personnel Psychology,13, 174–183. https://doi.org/
10.1027/1866‐5888/a000114
Ohly, S., & Schmitt, A. (2015). What makes us enthustiastic, angry, feeling
at rest or worried? Development and validation of an affective work
events taxonomy using concept mapping methodology. Journal of Busi-
ness and Psychology,30,15–35. https://doi.org/10.1007/s10869‐013‐
9328‐3
Oulasvirta, A., & Saariluoma, P. (2004). Long‐term working memory and
interrupting messages in human‐computer interaction. Behavior & Infor-
mation Technology,23,53–64. https://doi.org/10.1080/0144929031
0001644859
Parker, S. K. (2014). Beyond motivation: Job and work design for develop-
ment, health, ambidexterity, and more. Annual Review of Psychology,65,
661–691. https://doi.org/10.1146/annurev‐psych‐010213‐115208
Parker, D. F., & DeCotis, T. A. (1983). Organizational determinants of job
stress. Organizational Behavior and Human Performance,32, 160–177.
https://doi.org/10.1016/0030‐5073(83)90145‐9
Perlow, L. A. (1999). The time famine: Toward a sociology of work time.
Administrative Science Quarterly,44,57–81. https://doi.org/10.2307/
2667031
Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of
method bias in social science research and recommendations on how
to control it. Annual Review of Psychology,63, 539–569. https://doi.
org/10.1146/annurev‐psych‐120710‐100452
Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM
framework for assessing multilevel mediation. Psychological Methods,
15, 209–233. https://doi.org/10.1037/a0020141
Prem, R., Ohly, S., Kubicek, B., & Korunka, C. (2017). Thriving on challenge
stressors. Journal of Organizational Behavior,38, 108–123. https://doi.
org/10.1002/job.2115
Prem, R., Paškvan, M., Kubicek, B., & Korunka, C. (2016). Exploring the
ambivalence of time pressure in daily working life. International Journal
of Stress Management.. https://doi.org/10.1037/str0000044
Reinecke, L., & Hofmann, W. (2016). Slacking off or winding down? An
experience sampling study on the drivers and consequences of media
use for recovery versus procrastination. Human Communication
Research,42, 441–461. https://doi.org/10.1111/hcre.12082
Reis, H. T., & Gable, S. L. (2000). Event‐sampling and other methods for
studying everyday experience. In T. H. Reis, & M. C. Judd (Eds.), Hand-
book of research methods in social and personality psychology (pp.
190–222). New York: Cambridge University Press.
Richardson, H. A., & Taylor, S. G. (2012). Understanding input events: A
model of employees' responses to requests for their input. Academy
of Management Review,37, 471–491. https://doi.org/10.5465/
amr.2010.0327
Riordan, M. A., Markman, K. M., & Stewart, C. O. (2012). Communication
accommodation in instant messaging. Journal of Language and Social
Psychology,32,84–95. https://doi.org/10.1177/0261927X12462695
Rodell, J. B., & Judge, T. A. (2009). Can "good" stressors spark "bad" behav-
iors? The mediating role of emotions in links of challenge and hindrance
stressors with citizenship and counterproductive behaviors. Journal of
Applied Psychology,94, 1438–1451. https://doi.org/10.1037/
a0016752
Rose, E. (2014). Who's controlling who? Personal communication devices
and work. Sociology Compass,8, 1004–1017. https://doi.org/
10.1111/soc4.12194
Scherer, K. R. (1999). Appraisal theory. In T. Dalgleish, & M. J. Power (Eds.),
Handbook of cognition and emotion (pp. 637–663). Chichester, UK:
Wiley.
Scott, B. A., & Barnes, C. M. (2011). A multilevel field investigation of emo-
tional labor, affect, work withdrawal, and gender. Academy of
Management Journal,54, 116–136. https://doi.org/10.5465/
AMJ.2011.59215086
Scott, B. A., Colquitt, J. A., Paddock, E. L., & Judge, T. A. (2010). A daily
investigation of the role of manager empathy on employee well‐being.
Organizational Behavior and Human Decision Processes,113, 127–140.
https://doi.org/10.1016/j.obhdp.2010.08.001
Sears, M. S., Repetti, R. L., Robles, T. F., & Reynolds, B. M. (2016). I just
want to be left alone: Daily overload and marital behavior. Journal of
Family Psychology,30, 569–579. https://doi.org/10.1037/fam0000197
Semmer, N. (1984). Streßbezogene Tätigkeitsanalyse [Stress‐oriented task‐
analysis]. Beltz: Weinheim.
Seo, M.‐G., Bartunek, J. M., & Barrett, L. F. (2010). The role of affective
experience in work motivation: Test of a conceptual model. Journal of
Organizational Behavior,31, 951–968. https://doi.org/10.1002/job.655
Simon, G., & Pritchard, K. (2015). Performing the responsive and committed
employee through the sociomaterial mangle of connection. Organization
Studies,36, 241–263. https://doi.org/10.1177/0170840614556914
Sirois, F., & Pychyl, T. (2013). Procrastination and the priority of short‐term
mood regulation: Consequences for future self. Social and Personality
Psychology Compass,7, 115–127. https://doi.org/10.1111/spc3.12011
Skeels, M. M., & Grudin, J. (2009). When social networks cross boundaries:
A case study of workplace use of facebook and linkedin. GROUP '09.
Proceedings of the ACM 2009 international conference on Supporting
group work. New York: ACM,95–104. https://doi.org/10.1145/
1531674.1531689
Skovholt, K., Grønning, A., & Kankaanranta, A. (2014). The communicative
functions of emoticons in workplace e‐mails : :‐). Journal of Computer‐
Mediated Communication,19, 780–797. https://doi.org/10.1111/
jcc4.12063
Sonnentag, S. (2015). Dynamics of well‐being. Annual Review of Organiza-
tional Psychology and Organizational Behavior,2, 261–293. https://doi.
org/10.1146/annurev‐orgpsych‐032414‐111347
Sonnentag, S., Binnewies, C., & Mojza, E. J. (2008). "Did you have a nice
evening?" A day‐level study on recovery experiences, sleep, and affect.
Journal of Applied Psychology,93, 674–684. https://doi.org/10.1037/
0021‐9010.93.3.674
Sonnentag, S., & Starzyk, A. (2015). Perceived prosocial impact, perceived
situational constraints, and proactive work behavior: Looking at two
distinct affective pathways. Journal of Organizational Behavior,36,
806–824. https://doi.org/10.1002/job.2005
Speier, C., Vessey, I., & Valacich, J. S. (2003). The effects of interruptions,
task complexity, and information presentation on computer‐supported
decision‐making performance. Decision Sciences,34, 771–797.
https://doi.org/10.1111/j.1540‐5414.2003.02292.x
Spence, J. R., Brown, D. J., Keeping, L. M., & Lian, H. (2014). Helpful today,
but not tomorrow? Feeling grateful as a predictor of daily organizational
citizenship behaviors. Personnel Psychology,67, 705–738. https://doi.
org/10.1111/peps.12051
Sproull, L., & Kiesler, S. (1986). Reducing social context cues: Electronic mail
in organizational communication. Management Science,32, 1492–1512.
https://doi.org/10.1287/mnsc.32.11.1492
Story, L. B., & Repetti, R. (2006). Daily occupational stressors and marital
behavior. Journal of Family Psychology,20, 690–700. https://doi.org/
10.1037/0893‐3200.20.4.690
14 SONNENTAG ET AL.

Syrek, C., & Antoni, C. H. (2014). Unfinished tasks foster rumination and
impair sleeping—particularly if leaders have high performance expecta-
tions. Journal of Occupational Health Psychology,19, 490–499. https://
doi.org/10.1037/a0037127
Tan, C.‐H., Sutanto, J., Phang, C. W., & Gasimov, A. (2014). Using personal
communication technologies for commercial communications: A cross‐
country investigation of email and SMS. Information Systems Research,
25, 307–327. https://doi.org/10.1287/isre.2014.0519
Tassabehji, R., & Vakola, M. (2005). Business email: The kller impact. Com-
munications of the ACM,48(11), 64–70.
Ten Brummelhuis, L. L., Bakker, A. B., Hetland, J., & Keulemans, L. (2012).
Do new ways of working foster work engagement? Psicothema,24,
113–120.
Teuchmann, K., Totterdell, P., & Parker, S. K. (1999). Rushed, unhappy, and
drained: An experience sampling study of relations between time pres-
sure, perceived control, mood, and emotional exhaustion in a group of
accountants. Journal of Occupational Health Psychology,4,37–54.
https://doi.org/10.1037/1076‐8998.4.1.37
The Radicati Group (2015). Email statistics report 2015‐2019. London.
http://www.radicati.com/wp/wp‐content/uploads/2015/02/Email‐
Statistics‐Report‐2015‐2019‐Executive‐Summary.pdf
Trafton, J. G., Altmann, E. M., Brock, D. P., & Mintz, F. E. (2003). Preparing
to resume an interrupted task: Effects of prospective goal encoding and
retrospective rehearsal. International Journal of Human‐Computer
Studies,58, 583–603. https://doi.org/10.1016/S1071‐5819(03)00
023‐5
Trafton, J. G., & Monk, C. A. (2007). Task interruptions. Reviews of Human
Factors and Economics,3, 111–126. https://doi.org/10.1518/
155723408X299852
Tyler, J. R., & Tang, J. C. (2003). When can I expect an email response? A
study of rhythms in email usage. In Proceedings of the Eighth European
Conference on Computer‐Supported Cooperative Work. Helsinki, Finland:
Kluwer Academic Publishers.
Unsworth, K., Yeo, G., & Beck, J. (2014). Multiple goals: A review and
derivation of general principles. Journal of Organizational Behavior,35,
1064–1078. https://doi.org/10.1002/job.1963
Vitak, J., Crouse, J., & LaRose, R. (2011). Personal internet use at work:
Understanding cyberslacking. Computers in Human Behavior,27,
1751–1759. https://doi.org/10.1016/j.chb.2011.03.002
Wajcman, J., & Rose, E. (2011). Constant connectivity: Rethinking interrup-
tions at work. Organization Science,32, 941–961. https://doi.org/
10.1177/0170840611410829
Warr, P., Bindl, U. K., Parker, S. K., & Inceoglu, I. (2014). Four‐quadrant
investigation of job‐related affects and behaviors. European Journal of
Work and Organizational Psychology,23, 342–363. https://doi.org/
10.1080/1359432X.2012.744449
Watson, D. (1988). Intraindividual and interindividual analyses of positive
and negative affect: Their relation to health complaints, perceived
stress, and daily activities. Journal of Personality and Social Psychology,
54, 1020–1030. https://doi.org/10.1037/0022‐3514.54.6.1020
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. https://
doi.org/10.1037/0022‐3514.54.6.1063
Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoret-
ical 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). Stamford, CT: JAI Press.
Weiss, H. M., & Rupp, D. E. (2011). Experiencing work: An essay on a per-
son‐centric work psychology. Industrial and Organizational Psychology,
4,83–97. https://doi.org/10.1111/j.1754‐9434.2010.01302.x
Zapf, D. (1993). Stress‐oriented analysis of computerized office work.
European Work and Organizational Psychologist,3,85–100. https://doi.
org/10.1080/09602009308408580
Zeigarnik, B. (1927). Das Behalten erledigter und unerledigter Handlungen
[Remembering of finished and unfinished tasks]. Psychologische
Forschung,9,1–85.
Zijlstra, F. R. J., Roe, R. A., & Leonora, A. B. (1999). Temporal factors in
mental work: Effects of interrupted activities. Journal of Occupational
and Organizational Psychology,72, 163–185. https://doi.org/10.1348/
096317999166581
Zimmermann, B. K., Dormann, C., & Dollard, M. F. (2011). On the positive
aspects of customers: Customer‐initiated support and affective
crossover in employee‐customer dyads. Journal of Occupational and
Organizational Psychology,84,31–57. https://doi.org/10.1111/j.2044‐
8325.2010.02011.x
Sabine Sonnentag is a full professor of Work and Organizational
Psychology at the University of Mannheim, Germany. Her research
addresses the question of how individuals can stay healthy, ener-
getic, and productive at work. She studies job stress and job‐stress
recovery, health behavior (eating, physical exercise), proactive
behavior, and self‐regulation.
Leonard Reinecke is an associate professor of Media Psychology
at the Department of Communication at Johannes Gutenberg Uni-
versity Mainz, Germany. His research addresses media uses and
effects, media entertainment, and online communication with a
special focus on the interplay of media use and psychological
well‐being.
Jutta Mata is a full professor of Health Psychology at the Univer-
sity of Mannheim, Germany. Her research focuses on determi-
nants and consequences of different health behaviors. She also
studies how digital media can promote and impair health behaviors
and psychological well‐being.
Peter Vorderer is a full professor of Media and Communication
Studies at the University of Mannheim, Germany. His research
focuses on the uses and effects of media and new technology, par-
ticularly on the question what the permanent availability of online
communication does to its users.
How to cite this article: Sonnentag S, Reinecke L, Mata J,
Vorderer P. Feeling interrupted—Being responsive: How online
messages relate to affect at work. J Organ Behav. 2017;1–15.
https://doi.org/10.1002/job.2239
SONNENTAG ET AL.15