*Highlights (for review) x x x Limiting the frequency of checking email throughout the day reduced daily stress. Lower daily stress predicts greater well-being (e.g., higher positive affect). The frequency of checking email did not directly impact other well-being outcomes. *Title Page with Author details Checking Email Less Frequently Reduces Stress Kostadin Kushlev and Elizabeth W. Dunn University of British Columbia Please address correspondence to: Kostadin Kushlev, M.A. Department of Psychology University of British Columbia Vancouver, BC Phone: 1-778-866-2525 Email: kostadinpk@psych.ubc.ca *Manuscript without Author Details Click here to view linked References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Abstract Using email is one of the most common online activities in the world today. Yet, very little experimental research has examined the effect of email on well-being. Utilizing a within-subjects design, we investigated how the frequency of checking email affects well-being over a period of two weeks. During one week, 124 adults were randomly assigned to limit checking their email to three times a day; during the other week, participants could check their email an unlimited number of times per day. We found that during the limited email use week, participants experienced significantly lower daily stress than during the unlimited email use week. Lower stress, in turn, predicted higher well-being on a diverse range of well-being outcomes. These findings highlight the benefits of checking email less frequently for reducing psychological stress. Key Words: Email, subjective well-being, psychological stress, HCI 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 1. Introduction Every day, 183 billion emails are sent and received worldwide (Radicati & Levenstein, 2013). Email is among the most widespread online activities—in a 2011 survey, 92% of US adults reported using email to communicate (Pew Research Center, 2011). In addition to this ubiquity of email, people‘s inboxes play a central role in their lives: More than one-third of US adults surveyed in 2014 said that email would be ‗very hard‘ to give up—more than three times as many people who said the same about social media (Pew Research Center, 2014). And, according to one survey, about one-third of US workers report replying within 15 minutes of receiving a work email, and three-fourths reply within an hour (Kelleher, 2013). The popular press is rife with claims about the effects on well-being of this ubiquity of email in the life of today‘s information worker. Best sellers, such as the Four Hour Work Week (Ferriss, 2007), recommend a variety of approaches to reducing stress at work by, for example, checking email only twice a day. In stark contrast to this abundance of causal claims in the popular discourse, very little experimental research has explored how different approaches to dealing with email actually impact well-being. Accordingly, in the present research, we set out to conduct the first experimental field study to investigate whether the frequency with which people check email exerts a causal impact on their well-being. Correlational research has provided preliminary evidence that dealing with email may be associated with negative outcomes for well-being (for a review, see Taylor, Fieldman, & Altman, 2008). This correlational research indicates that people who handle more email experience lower job satisfaction (Merten & Gloor, 2009) and perceive email as a greater source of stress (Jerejian, Reid, & Rees, 2013; Mano & Mesch, 2010). 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Similarly, people who spend more time on email report greater work overload (e.g., feeling emotionally drained, frustrated, and stressed from work; Barley, Meyerson, & Grodal, 2011). Of course, this correlational research does not enable inferences about the causal effect of email on well-being. A busier work schedule, for example, may result in both dealing with more email and perceiving one‘s job as a greater source of stress. If email does have a causal effect on well-being, what specific aspects of dealing with a larger inbox influence well-being? One possibility is that simply thinking about the ballooning size of one‘s inbox directly causes more stress, thus compromising wellbeing. In contrast to this possibility, however, people who handle more emails at work perceive email as a way to improve work effectiveness (Mano & Mesch, 2010) and see themselves as more able to cope with stressors (Barley et al., 2011). Another popular idea is that email reduces well-being because it allows people to work longer hours, by, for example, answering emails from home (e.g., Renauld, Ramsey, & Hair, 2006). Contrary to this idea, the time spent working does not mediate the relationship between time spent on email and work overload (Barley et al., 2011). Thus, neither sheer email volume nor time spent on email seem to influence well-being directly. A third possibility is that the effect of dealing with email on well-being depends on the way people manage their large inboxes. Providing some initial support for this possibility, a training program in effective email management resulted in less self-reported workflow impairment due to email and reduced level of email strain (e.g., being annoyed by email; Soucek & Moser, 2010). One critical aspect of managing email is how frequently people attend to their inbox (e.g., Dabbish & Kraut, 2006). Faced with the constant flow of new email messages, some people respond by frequently switching between other tasks and their 3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 email (Jackson, Dawson, & Wilson, 2001, 2003; Gonzalez & Mark, 2004; Whittaker, Bellotti, & Gwizdka, 2006, Whittaker & Sidner, 1997). Employees in one British company, for example, were interrupted by email on average every five minutes, and the typical worker responded within six seconds of receiving an email (Jackson et al., 2001, 2003). Even in the absence of such frequent external interruptions, email may provide a readily available source of distraction, which is important considering that selfinterruptions account for 40% of all interruptions at work (Czerwinski, Horvitz, & Wilhite, 2004). In short, people often manage their email by attending to their inbox frequently, thus resulting in frequent interruptions and switching between tasks. In the present research, we set out to experimentally examine how the frequent interruptions and task switching due to email impact well-being. 2. Theory and relevance to basic research A wealth of basic research and theory documents the toll of task switching on cognitive resources. Classical theorizing in cognitive psychology postulates that people have limited cognitive resources (Navon & Gopher, 1979; Pashler, 1998), and basic research has shown that when two tasks require the same cognitive resource (e.g., working memory), people cannot perform these tasks simultaneously and have to instead switch between tasks (Garavan, 1998; Liefooghe, Barrouillet, Vandierendonck, Camos, 2008; Oberauer, 2003). According to the time-based resource sharing model of attention (Barrouillet, Bernardin, & Camos, 2004), the very act of switching between tasks requires deployment of attention, thus further taxing people‘s limited cognitive resources and resulting in greater cognitive load (Barrouillet, et al., 2004; Liefooghe et al., 2008). To make matters worse, according to the load theory of attention (Lavie, 2010), higher 4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 cognitive load can further increase proneness to distraction (Lavie & De Fockert, 2005; Lavie, Hirst, De Fockert, & Viding, 2004), thus potentially resulting in even more multitasking. Although relatively little research has directly examined how frequent task switching throughout the day impacts well-being, there are several reasons to believe that the cognitive tax associated with task switching may be detrimental to well-being. First, unsurprisingly, the greater cognitive load induced by frequent task switching has been postulated and shown to impair performance and speed of completing tasks that require cognitive effort (Bowman, Levine, Waite, & Gendron, 2010; Rubinstein, Meyer, & Evans, 2001). Thus, frequent multitasking may result in doing worse at work tasks, potentially increasing stress. In support of this prediction, when participants in a lab experiment were frequently interrupted by instant messages, they reported greater stress and frustration while working on another task (Mark, Gudith, & Klocke, 2008). In another study, after obtaining baseline measurements of task switching and physiological stress (as measured by heart rate variability) during three regular workdays, researchers asked a convenience sample of 13 workers to completely refrain from checking new email for five workdays (Mark, Voida, & Cardello, 2012). When they were cut off from new email, these workers both switched less between work tasks and experienced less stress as compared to baseline, suggesting a potential link between task switching and stress. Second, both psychological theory and research suggest that cognitive resources are essential for emotion regulation (Holzel et al., 2011, Posner & Rothbart, 2007), and therefore, to the extent that switching between tasks taxes cognitive resources, frequent 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 task switching may compromise emotional well-being. Indeed, experimental research has shown that increasing the frequency of interruptions during a cognitive task leads to less positive affect (Zijlstra, Roe, Leonora, & Krediet, 1999). In short, basic theory and research suggest that frequent task switching can increase cognitive load and impair performance, with potential downstream consequences for well-being. In addition, recent research has shown that people tend to check their email frequently throughout the day (e.g., Jackson et al., 2001, 2003), thus effectively making email into a source of task switching. No experimental research, however, has ever directly explored whether the frequency with which people check their emails has an impact on well-being. Thus, building on psychological theory and basic research on task switching, we set out to conduct the first experimental field investigation directly examining how the frequency of checking email affects well-being. 3. Summary of the present research Preliminary evidence has suggested a link between email and lower well-being, but most research has been correlational, preventing any causal conclusions. Furthermore, most researchers have used overall email volume to predict well-being, although evidence indicates that inbox size might matter less than the way people manage their large inboxes. A common approach to managing one‘s inbox is to check email frequently and respond to incoming messages quickly, which results in frequent task switching and task interruptions. Although some research suggests that interrupting and switching between tasks can be detrimental to well-being, no research has ever directly examined whether people experience improved well-being when they check email less 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 frequently. In the present research, we set out to experimentally examine how the frequency of checking email affects subjective well-being. 4. Method To examine whether checking email less frequently can improve well-being, we designed a two-week within-subjects study. Specifically, we randomly assigned participants to minimize the frequency of checking their email during one week and to maximize frequency during the other week. Based on previous research linking email to stress, we assessed weekly and daily stress, as well as stress during a particular important activity. Due to the dearth of research on how handling email can impact other components of well-being, we adopted an exploratory approach and assessed the effects of our manipulation on a wide range of established well-being outcomes. Specifically, given previous theorizing underscoring the importance of measuring theoretically distinct components of well-being (Biswas-Diener, Kashdan, & King, 2009; Kashdan, BiswasDiener, & King, 2008; Ryan & Huta, 2009; Ryff, 1989), we included measures of both hedonic (e.g., affect) and eudaimonic well-being (e.g., meaning in life, environmental mastery). Finally, to capture other important aspects of optimal day-to-day functioning, we examined mindfulness, perceived sleep quality, and self-reported productivity. 4.1. Participants A total of 142 adults agreed to participate in this two-week study. Eighteen participants dropped out of the study before completing at least one questionnaire in each condition1, leaving a final sample of 124 participants (age: M = 30, SD = 10; sex: 67% female). Participants were predominantly Caucasian (55%) or Asian (28%). About twothirds of the sample identified as either graduate or undergraduate students (Mage = 27 7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 years). The remaining one-third of participants were community members who came from a range of occupations and industries including health care (e.g., doctor, pharmacist), academia (e.g., professor), finance (e.g., financial analyst), administration (e.g., secretary), and IT (e.g., software developer). Participants were recruited through posters in community centers, paid advertisements in local newspapers, listservs, and snowball sampling. We advertised the study as suitable for people who got a lot of email and sometimes felt overwhelmed by it. Participants only qualified for the study if they had some flexibility in how often they could check their email and were interested in experimenting with the way they managed their email. Participants received the chance to win $150 and the option to receive individualized feedback about their well-being during the study. 4.2. Design and manipulation We used a counterbalanced within-subjects design. Participants were first invited to complete an initial survey, in which they completed basic demographic questions and reported how many times they checked their email on a typical workday. On the first Sunday after this initial survey, participants received a set of instructions on how to handle their email for the following work week. The next Sunday, participants received a different set of instructions for handling their email during the second week of the study. The order of instructions was counterbalanced, such that participants were randomly assigned to spend one week in our unlimited email condition and the other week in our limited email condition. Random assignment was performed using a random number generator. 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 In the unlimited email condition, we instructed participants to check their email as often as they could, and to keep their mailbox open throughout the day; additionally, participants were asked to switch on any email notification systems that they used. By contrast, in the limited email condition, we instructed participants to check their email 3 times per day, while keeping their mailbox closed during the rest of the day and switching off any new email alerts. Although we sought to maximize the betweencondition difference in how often people checked email, we imposed a fairly moderate limit on email usage (3X/day) with the goal of enabling a diverse sample of participants to comply with the instructions. At 5 pm on each weekday during the two study weeks, we sent participants a link to complete a survey. Because we wanted to include busy professionals in our sample, we limited the time necessary to complete each daily survey to approximately 10 minutes. Thus, some measures evaluating day-to-day well-being were included only on certain days. Specifically, some scales were administered only on Monday, Wednesday, and Friday, whereas others were administered only on Tuesday and Thursday. In addition, for longer measures, we preselected items from existing scales in order to create shorter scales that could be administered more frequently throughout the study. All scales, including abbreviated scales, showed acceptable to good statistical reliability (see Table 1). All survey questions and the verbatim manipulation instructions are available online at osf.io/cx7z6. The average number of surveys participants completed per week was 4.4/5, indicating a good overall completion rate. Importantly, there were no differences in completion rate between the limited (M = 4.4) and unlimited (M = 4.4) conditions. 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Because some participants did not complete surveys on some days, degrees of freedom vary somewhat between measures. 4.3. Measures 4.3.1. Manipulation checks. We measured the successfulness of the manipulation with self-report measures of the frequency with which people checked email on particular days of the week. Although more objective estimates of the frequency of checking email can be obtained using software that tracks actual behavior, we opted for self-report measures in order to be able to recruit participants from a wide range of different professions and companies. In addition, because each survey was completed at the end of the day, we expect people‘s self-reports to be fairly accurate representations of their actual behavior (c.f., Kahneman, Krueger, Schkade, Schwarz, & Stone, 2004). Accordingly, on Monday and Friday of each week, participants reported how often they had checked their email throughout the day on a scale from 0 to 30+; participants were encouraged to report their actual email use regardless of the experimental instructions. At the initial baseline survey before participants were assigned to condition, participants also reported the number of times they normally checked their email during a workday. In addition, on Mondays and Fridays of each week of the experiment, we also collected other descriptive information about email use, including the time spent using email and the number of emails received and answered. 4.3.2. Dependent measures. 4.3.2.1. Day-level measures. Each daily survey asked participants to report how distracted they felt by email and included a series of questions broadly assessing their subjective experience during that day. Specifically, to assess well-being, we measured 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 stress, as well as hedonic and eudaimonic components of well-being, including daily affect (i.e., positive and negative affect), social connectedness, environmental mastery, nonhedonic well-being, and meaning in life. Additionally, we measured their overall state mindfulness, productivity, and sleep quality (see Table 1). 4.3.2.2. Activity-level measures. On Wednesday of each week, participants were prompted to select one of the most important activities they did on this day. Our goal was to assess people‘s level of stress and basic need satisfaction (Deci & Ryan, 2000) during a particular activity. Specifically, we measured task tension, perceived competence, and interest/enjoyment (see Table 1). 4.3.2.3. Week-level measures. Finally, we measured participants‘ overall evaluation of their well-being over each week of the study. Specifically, on Thursday of each week, participants completed measures of stress, environmental mastery, presence of meaning in life, and perceived productivity with regards to their experience ―over the past week‖ (see Table 1). 11 1 2 3 4 5 Table 1. 6 7 8 Measures and Main Effects 9 10 Level Variable Source 11 12 13 Day Email NA 14 Distraction 15 16 Stress Perceived Stress 17 Scale (PSS, Cohen, 18 Kamarck, & 19 Mermelstein, 1983) 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Positive & PANAS (Watson, 37 negative Clark, & Tellegen, 38 affect 1988). 39 40 41 42 43 44 45 46 47 48 49 Scale Days Measured Selected Items Item Selection Rationale α‘s 0–not at all; 6–very much Mon, Tue, Wed, Thu, Fri Mon, Tue, Wed, Thu, Fri ―Overall, how distracted were you by your emails today?‖ ―1. Today, how often have you felt that you were unable to control the important things in your life?‖ ―2. Today, how often have you felt nervous and ‗stressed‘?‖ ―3. Today, how often have you found that you could not cope with all the things that you had to do?‖ ―4. Today, how often have you felt that you were on top of things?‖ (R) ―5. Today, how often have you been angered because of things that were outside of your control?‖ All 20 items + an additional item (‗happy‘) in the positive affect scale (see Aknin, Dunn, Whillans, Grant, & Norton, 2013) We created a face-valid item. 0–never 4–very often 1–very slightly or not at all; 5–extremely Tue, Thu M (SD) Unlimited Email 2.18 (1.36) Cohen ’s d NA M (SD) Limited Email 1.83 (1.18) We picked 5 items from this 10-item measure because they were adaptable to measure daily stress. .55–.85 1.46 (.55) 1.55 (.57) -.37* NA Positive .90–.91 2.87 (.65) 2.90 (.69) -.10 Negative .86–.89 1.73 (.61) 1.71 (.61) .08 12 -.51** 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 Nonhedonic well-being Environmental mastery White & Dolan (2009) Environmental mastery—short scale (EM; Ryff & Keyes, 1995). Social connectedness scale (Lee, Draper, & Lee, 2001) 0–not at all; 6–very much 1–strongly disagree; 6–strongly agree Mon, Wed, Fri Tue, Thu All items NA .86–.93 3.76 (.93) 3.71 (1.01) .15 All items. NA .60–.78 4.06 (.80) 4.10 (.88) -.08 1–strongly disagree; 6–strongly agree Mon, Tue, Wed, Thu, Fri ―1. Today, I felt distant from people.‖ (R) ―2. Today, I felt close to people.‖ .75–.85 4.10 (.87) 4.07 (.88) .06 Meaning in life Kushlev, Dunn, & Ashton-James (2012) 0–not at all; 6–very much Tue, Thu Single-item scale We chose 2 items from this 20-item scale. Item 1 was chosen because it had the highest factor loading of all other items. Item 2 was chosen because it had strong face validity. NA NA 3.47 (1.18) 3.40 (1.20) .12 State mindfulness State mindfulness scale (Brown & Ryan, 2003). NA 1–almost never; 6–almost always All items. .85–.90 2.51 (.71) 2.64 (.83) -.22 0–not at all; 6–very much Mon, Tue, Wed, Thu, Fri Mon, Tue, Wed, Thu We created Items 1 and 2 to as facevalid measures of people‘s sense of accomplishment from work. Item 3 was adapted from the Basic Need Satisfaction at Work scale. .85–.92 3.47 (1.06) 3.41 (1.14) .12 NA 0–very bad; Mon, Tue, We created a NA 3.71 (1.09) 3.79 (1.00) -.19 Social connectedness Perceived productivity Sleep quality ―1. Overall today, did you feel you got done the things at work that were most important to you?‖ ―2. Overall today, how satisfied were you with what you accomplished at work?‖ ―3. Overall today, to what extent did you feel a sense of accomplishment from working?‖ ―Overall, how would your 13 1 2 3 4 5 6 7 Activity 8 9 10 11 12 13 14 Week 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 6–very good Pressure/ tension Interest/ Enjoyment Perceived Competence Stress Environmental mastery Meaning In life Perceived productivity Ryan, Mims, & Koestner, 1983) Ryan (1982) McAuley, Duncan, &, Tammen, 1987 PSS (Cohen et al.,1983) EM (Ryff & Keyes, 1995). Meaning In Life Questionnaire— presence of meaning subscale (Steger, Frazier, Oishi, & Kaler, 2006) NA 1–not at all true 7–very true 1–not at all true 7–very true 1–not at all true 7–very true 0–never 4–very often 1–strongly disagree; 6–strongly agree 1 – absolutely untrue; 7 – absolutely true 0–not at all; 6–very much Wed, Thu, Fri Wed rate the quality of your sleep last night?‖ All items face-valid measure of sleep NA .76–.86 3.50 (1.48) 3.81 (1.34) -.45 Wed All items NA .93–.94 3.52 (1.46) 3.80 (1.53) -.39 Wed All items NA .92-.94 4.64 (1.54) 4.44 (1.46) .28 Thu All items NA .82–.85 1.68 (.63) 1.67 (.67) .04 Thu All items NA .73–.84 4.05 (1.06) 4.02 (.99) .07 Thu All items NA .92 4.69 (1.48) 4.65 (1.35) .09 Thu See daily measure See daily measure .88–.90 3.55 (1.09) 3.56 (1.19) -.01 Notes. Alpha values are calculated separately for each day the corresponding questionnaire was administered; stress was measured only on Monday, Wednesday, and Friday for some participants. † p < .10; *p < .05; **p < .01; ***p < .001 14 † † 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 5. Results 5.1. Manipulation Checks Confirming the success of our manipulation, people checked their email significantly fewer times per day in the limited email condition (M = 4.70, SD = 4.10) than in the unlimited email condition (M = 12.54, SD = 8.02; t[115] = -10.23, p < .001). Importantly, the average number of times people reported checking their email on a normal day at work was 15.48 at baseline (SD = 8.69)—similar to number of times reported in the unlimited email condition, but substantially higher than in the limited email condition. Thus, our experimental manipulation made people check their email less frequently than usual in the limited email condition, but produced trivial differences in people‘s behavior as compared to normal in the unlimited email condition. In short, our manipulation was successful in inducing differences in how people managed their email across conditions with the limited email instructions driving these differences in behavior. Intriguingly, there were no significant differences between conditions in how many emails people received (Mlimited = 16.64 vs. Munlimited = 16.04, t(114) = 1.31, p = .19) or responded to (Mlimited = 5.30 vs. Munlimited = 5.95, t(115) = -1.58, p = .12), suggesting that our manipulation primarily affected how often people checked email rather than the volume of email they managed. 5.2. Direct Effects Our goal was to explore whether manipulating how often people checked email would affect their subjective experience. First, we ran a series of ANOVAS comparing people‘s experiences in each of the two conditions as assessed by all activity, day, and week level measures. In order to minimize the effect of individual day variation, we 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 calculated weekly composites for all constructs that were assessed on more than one day of each week. We found that participants felt less daily stress in the limited as compared to the unlimited email condition, F(1, 121) = 4.18, p =.04, Cohen’s d = .37 (for descriptive statistics on all measures, see Table 1; d-scores were calculated using the paired-samples F-test conversion tool of the ESCI software, as recommended by Cumming, 2012). Consistent with this difference in day-to-day stress, when engaged in a specific important activity, people felt less tense in the limited as compared to the unlimited email condition, F(1, 96) = 3.84, p =.05, Cohen’s d = .45. Interestingly, while limiting the frequency of checking email influenced people‘s daily stress and the tension they felt during a particular activity, the manipulation did not affect their memory of how stressful the week had been overall, F(1, 91) = .04, p =.838, Cohen’s d = .04. In addition to the main effects on stress, we also found that people felt less distracted by their email in the limited as compared to the unlimited email condition, F(1, 123) = 8.04, p =.01, Cohen’s d = .51. No other significant main effects emerged, although people reported marginally greater enjoyment during a particular important activity in the unlimited vs. limited email condition, F(1, 96) = 3.71, p =.06, Cohen’s d = .39. To examine whether our manipulation produced different effects for students versus community members, we ran a series of mixed ANOVAs with condition as a within-subjects factor and status (student vs. community member) as a between-subjects factor. We found that student status did not moderate the effect of condition on tension, F(1, 94) = .65, p =.42, or daily stress, F(1, 119) = 1.07, p =.30. Student status, however, moderated the effect of condition on distraction by email, F(1, 121) = 5.27, p =.02, although the main effect of condition remained significant, F(1, 121) = 4.29, p =.04. 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Post-hoc analyses indicated that while students were significantly less distracted by their email in the limited email condition than in the unlimited email condition (p = .001), community members were not (p = .89). In short, stress was the only outcome variable that was consistently and directly influenced by our manipulation. Because stress can have a wide range of downstream consequences for well-being (Bolger, DeLongis, Kessler, & Schilling,1989; Daniels & Guppy, 1994; DeLongis, Folkman, & Lazarus, 1988; Dua, 1994; Lazarus, 2006), reducing stress by checking email less often may have broader implications for wellbeing. Accordingly, we next examine whether the differences in stress between conditions predicted other measures of well-being. 5.3. Indirect Effects Through Stress To examine the indirect effect of our manipulation on well-being through stress, we followed recommendations by Judd, Kenny, and McClelland (2001) for conducting mediation analyses with repeated measures. As show in Equation 1 below, in each case, we predicted the difference scores in the outcome variables (Y) from sum and difference scores of stress (X). The difference scores were calculated by subtracting the unlimited email scores from their corresponding scores in the limited email week. The regression coefficient of the difference score of stress controlling for its sum score is the measure of indirect effect of condition on well-being through stress (Judd et al., 2001). If the difference score of stress significantly predicts the difference score of other well-being measures, this will provide initial evidence that by influencing stress, checking email less frequently may have broader implications for well-being. (1) Ydiff = β1Xsum + β2Xdiff + ε, where 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 (a) Ydiff = Ylimited – Yunlimited; (b) Xdiff = Xlimited – Xunlimited (c) Xsum = Xlimited + Xunlimited Using Equation 1, we found that lower daily stress in the limited email condition was associated with significantly better subjective experiences across almost all daily measures (see Figure 1). That is, stress was associated with significantly higher negative affect (β = .37, p < .001) and marginally lower positive affect (β = -.16, p = .10). Stress was also negatively associated with state mindfulness (β = -.43, p < .001), nonhedonic well-being (β = -.42, p < .001), environmental mastery (β = -.40, p < .001), meaning in life (β = -.26, p = .01), social connectedness (β = -.24, p = .01), self-reported productivity (β = -.23, p = .01), and sleep quality (β = -.22, p = .02; see Figure 1). Finally, we examined whether daily stress predicted people‘s reports of their overall weekly well-being. Unsurprisingly, daily stress was predictive of weekly stress (β = .50, p < .001). Additionally, daily stress was related to weekly environmental mastery (β = -.37, p < .001). People who experienced more day-to-day stress also reported somewhat lower productivity (β = -.19, p = .07) and slightly less meaning in life (β = .12, p = .26) during the week, although these effects did not reach statistical significance. Taken together, this pattern of indirect effects points to the conclusion that checking email less frequently might have broader downstream consequences for wellbeing by reducing stress. Because indirect effect analyses are inherently correlational, however, the present research only provides direct causal evidence for the impact of our manipulation on stress. 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Figure 1. Relationships between daily stress and daily well-being ordered by effect size (β). Effect sizes represent the effect of the difference in stress between the limited and unlimited email conditions on the difference in the outcomes measures between the two conditions (see Equation 1 for details of analyses). * p < .05; **p < .01; ***p < .001. 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 6. Discussion In the first experimental field study examining the effect of checking email less frequently, people experienced reduced stress when they were assigned to limit the number of times they checked their email. Specifically, limiting the number of times people checked their email per day lessened tension during a particular important activity and lowered overall day-to-day stress. In turn, lower daily stress was associated with higher well-being, as assessed by a range of outcomes including hedonic (e.g., affect) and eudaimonic outcomes (e.g., meaning in life, environmental mastery, social connectedness). Furthermore, lower stress was associated with other positive outcomes including higher mindfulness, self-perceived productivity, and sleep quality. These findings provide causal evidence that checking email less frequently can directly decrease stress, with potential downstream benefits for well-being. 6.1. Implications and Limitations In line with recent recommendations to assess multiple specific components of well-being (Kashdan, et al., 2008), we included a broad array of measures in our study. Given this exploratory approach, it is possible that the significant effects we observed on stress are simply an artifact of the large number of statistical tests we conducted. The present study, therefore, should be seen as laying the groundwork for future confirmatory research. That said, previous correlational research has also shown that the way people handle email is related to stress rather than other components of well-being (e.g., Jerejian, et al, 2013; Mano & Mesch, 2010). The present findings dovetail with this existing work in suggesting that checking email less often primarily affects stress, rather than other components of well-being, such as people‘s sense of meaning in life. In short, our pattern 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 of findings suggest that while checking email less frequently may help to alleviate stress, changing how frequently people check email is by no means a panacea for improving well-being. Over time, however, it is conceivable that reduced levels of stress could eventually produce consequences for well-being more broadly. Indeed, a meta-analysis of forty-eight experimental studies (n = 3736) showed that stress reduction interventions have an impact on a range of outcomes including anxiety, symptoms of depression, and overall perceived quality of work life (van der Klink, Blonk, Schene, & van Dijk, 2001). Consistent with this research, we found that stress was associated with an overall poorer well-being in the course of our experiment. Thus, given that checking email less frequently can reduce stress in the course of a week, the benefits for other aspects of well-being might emerge over time. The broader benefits of reducing the frequency of checking email on well-being might also be more likely to materialize if changes were made at the organizational level, rather than just the individual level. In our study, we manipulated participants‘ behavior, but had no control over the expectations of those around them. Indeed, recent research suggests that some people feel stressed by email in part because others expect them to reply quickly (e.g., Gillespie, Walsh, Winefields, Dua, & Stough, 2001). Organizations might be able to maximize workers’ well-being by introducing interventions at a company-wide or team-wide level, thereby altering co-workers‘ expectations. Another potential limitation of the present research is that we did not include a control condition in which participants completed our measures without being asked to alter their email usage patterns. At baseline, however, participants in our study reported 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 checking email roughly the same number of times (~15) as people in the unlimited email condition (~13), but significantly more times than people in the limited email condition (~5). Thus, being instructed to check email as frequently as possible did not increase the number of times people checked email as compared to baseline, whereas being instructed to limit checking email reduced the number of times people checked email as compared to baseline. Our findings suggest, therefore, that checking email less frequently than normal reduces stress rather than that checking email more frequently than normal increases stress. Of course, because our measures of frequency were based on self-reports, the particular values participants reported should be interpreted with caution. For the purposes of the present experimental research, however, we were not interested in estimating the exact number of times people checked their email, but rather in inducing an overall measurable difference in behavior across the two experimental conditions. For this purpose, our measures indicate a clear reduction of the number of times people checked their email in the limited email condition as compared to baseline and the unlimited email condition. More broadly, although the effects we observed did not depend on whether participants were students or community members, our reliance on a convenience sample raises important issues of generalizability. In particular, given that we intentionally recruited heavy email users who had some flexibility in the way they managed email, our intervention might be unlikely to reduce stress among individuals who receive little email or have no choice about how frequently they check email. In some professions, for example, workers rely on constant updates to successfully do their job (e.g., stock 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 brokers), such that attempting to check email less often might be more stressful. Thus, future research with larger representative samples should explore when and for whom limiting email checking is beneficial vs. detrimental for well-being. 6.3. Coda In conclusion, we employed careful experimental design to demonstrate that a simple change in how people approach email may reduce overall levels of stress on a typical day. Thus, by applying psychological theory and extending basic research on task switching, we provided evidence for the potential toll on well-being of frequent checking of email—one of the most common sources of task switching for the modern information worker. 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Footnotes 1 Of the 18 people who dropped out before completing at least one survey per week, 7 did not complete any surveys during both weeks and the remaining 11 completed at least one survey during the first week, but none in the second week. For those 11, the drop out rate from each condition was virtually the same: 8% dropped out when checking email was minimized and when 8% when checking email was maximized. 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 References Aknin, L. B., Dunn, E. W., Whillans, A. V., Grant, A. M., & Norton, M. I. (2013). Making a difference matters: Impact unlocks the emotional benefits of prosocial spending. Journal of Economic Behavior and Organization, 88, 90–95. doi:10.1016/j.jebo.2013.01.008 Barley, S. R., Meyerson, D. E., & Grodal, S. (2011). E-mail as a source and symbol of stress. Organization Science, 22(4), 887–906. doi:10.1287/orsc.1100.0573 Barrouillet, P., Bernardin, S., & Camos, V. (2004). Time constraints and resource sharing in adults‘ working memory spans. Journal of Experimental Psychology: General, 133, 83–100. Biswas-Diener, R., Kashdan, T. B., & King, L. A. (2009). Two traditions of happiness research, not two distinct types of happiness. The Journal of Positive Psychology, 4(3), 208–211. doi:10.1080/17439760902844400 Bolger, N., DeLongis, A., Kessler, R. C., & Schilling, E. A. (1989). The emotional effects of daily stress. Journal of Personality and Social Psychology, 57, 808-818. Bowman, L. L., Levine, L. E., Waite, B. M., & Gendron, M. (2010). Can students really multitask? An experimental study of instant messaging while reading, Computers and Educaiton, 51, 927–931. Brown, K. W., & Ryan, R. M. (2003). The benefits of being present: Mindfulness and its role in well-being. Journal of Personality and Social Psychology, 84(4), 822–848. doi:10.1037/0022-3514.84.4.822 Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396. Czerwinski, M., Horvitz, E. & Wilhite, S. A diary study of task switching and interruptions. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI04, 175–182. doi:10.1145/985692.985715 Cumming, G. (2012). Understanding the new statistics: Effect sizes, confidence intervals, and meta-analysis. New York: Routledge. Dabbish, L. A., & Kraut, R. E. (2006). Email overload at work: An analysis of factors associated with email strain. In Proceedings of the 2006 20th anniversary conference on Computer supported cooperative work, Banff, Alberta, Canada (pp. 431–440). ACM: New York. Daniels, K., & Guppy A (1994). Occupational stress, social support, job control, and well-being. Human Relations, 47, 1523–1544, doi:10.1177/001872679404701205 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 DeLongis, A., Folkman, S., & Lazarus, R.S. (1988). The impact of daily stress on health and mood: Psychological and social resources as mediators. Journal of Personality and Social Psychology, 54, 454-460. Dua, J. K. (1994). Job stressors and their effects on physical health, emotional health and job satisfaction in a university. Journal of Educational Administration, 32(1), 59–78. doi:10.1108/09578239410051853 Ferriss, Timothy (2007). The 4-hour workweek: Escape 9-5, live anywhere, and join the new rich. New York, NY: Crown Publishing Group. Gillespie, N. A., Walsh, M., Winefields, A. H., Dua, J., & Stough, C. (2001). Occupational stress in universities: Staff perceptions of the causes, consequences and moderators of stress. Work & Stress, 15(1), 53–72. Gonzáles, V. & Mark, G. (2004). Constant, constant, multi- tasking craziness: managing multiple working spheres. In Proceedings of CHI ‘04, Volume 6 (pp. 113-120). ACM Press: New York Holzel, B. K., Lazar, S. W., Gard, T., Schuman-Olivier, Z., Vago, D. R., & Ott, U. (2011). How does mindfulness meditation work? Proposing mechanisms of action from a conceptual and neural perspective. Perspectives on Psychological Science, 6(6), 537–559. doi:10.1177/1745691611419671 Jackson, T. W., Dawson, R. J., & Wilson, D. (2001). The cost of email interruption. Journal of Systems and Information Technology, 5, 81–92. Jackson, T., Dawson, R., & Wilson, D. (2003). Understanding e-mail interaction increases organizational productivity. Communications of the ACM, 48(6), 80–84. Jerejian, A. C. M., Reid, C., & Rees, C. S. (2013). The contribution of email volume, email management strategies and propensity to worry in predicting email stress among academics. Computers in Human Behavior, 29(3), 991–996. doi:10.1016/j.chb.2012.12.037 Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone, A. A. (2004). A survey method for characterizing daily life experience: The Day Reconstruction Method. Science, 306, 1776–1780. doi:10.1126/science.1103572 Kelleher, D. (2013). Confessions of an email user. Retrieved from: http://www.gfi.com/blog/survey-81-of-u-s-employees-check-their-work-mail-outsidework-hours/ Kushlev, K., Dunn, E. W., & Ashton-James, C. E. (2012). Does affluence impoverish the experience of parenting? Journal of Experimental Social Psychology, 48(6), 1381–1384. 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 doi:10.1016/j.jesp.2012.06.001 Lavie, N. (2010). Attention, distraction, and cognitive control under load. Current Directions in Psychological Science, 19(3), 143–148. doi:10.1177/0963721410370295 Lavie, N., Hirst, A., De Fockert, J.W., & Viding, E. (2004). Load theory of selective attention and cognitive control. Journal of Experi- mental Psychology: General, 133, 339–354. Lavie, N., & De Fockert, J.W. (2005). The role of working memory in attentional capture. Psychonomic Bulletin & Review, 12, 669–674. Lazarus, R. S. (2006). Stress and emotion: A new synthesis. Springer, New York. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer, New York. Lee, R. M., Draper, M., & Lee, S (2001). Social connectedness, dysfunctional interpersonal behaviors, and psychological distress: testing a mediator model. Journal of Counseling Psychology, 48, 310–318. Liefooghe, B., Barrouillet, P., Vandierendonck, A., & Camos, V. (2008). Working memory costs of task switching. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34(3), 478–494. doi:10.1037/0278-7393.34.3.478 Mano, R. S., & Mesch, G. S. (2010). E-mail characteristics, work performance and distress. Computers In Human Behavior, 26(1), 61–69. doi:10.1016/j.chb.2009.08.005 Mark, G. J., Gudith, D., & Klocke, U (2008). The cost of interrupted work: More speed and stress. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI08, 107-110. doi:10.1145/1357054.1357072 Mark, G., J., Voida, S., & Cordello, A. V. (2012). A pace not dictated by electrons: An empirical study of work without email. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI12, 555–564. doi:10.1145/2207676.2207754 McAuley, E., Duncan, T., & Tammen, V. V. (1987). Psychometric properties of the Intrinsic Motivation Inventory in a competitive sport setting: A confirmatory factor analysis. Research Quarterly for Exercise and Sport, 60, 48-58. Merten, F., & Gloor, P. (2010). Too much e-mail decreases job satisfaction. Procedia – Social and Behavioral Science, 2(4), 6457–6465. doi:10.1016/j.sbspro.2010.04.055 Pashler, H. (1998). The psychology of attention. Cambridge, MA: MIT Press. 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Pennebaker, J. (2014, February). Expanding social and personality psychology in departments, universities, and within SPSP. Presidential address presented at the Annual Meeting of the Society for Personality and Social Psychology, Austin, TX. Pew Research Center (2011). Internet and American Life Project. Search and Email Still Top the List of Most Popular Activities: Two Activities Nearly Universal Among Adult Internet Users. Available at: http://pewinternet.org/Reports/2011/Search-and-email.aspx Pew Research Center (2014). The Web at 25. Available at: http://www.pewinternet.org/2014/02/25/the-web-at-25-in-the-u-s Posner, M. I., & Rothbart, M. K. (2007). Research on attention networks as a model for the integration of psychological science. Annual Review of Psychology, 58(1), 1–23. doi:10.1146/annurev.psych.58.110405.085516 Radicati, S. & Levenstein, J (2013). Email Statistics Report, 2013-2017. Available at: http://www.radicati.com/wp/wp-content/uploads/2013/04/Email-Statistics-Report-20132017-Executive-Summary.pdf Renaud, K., Ramsay, J., & Hair, M. (2006). ―You‘ve got e- mail!‖ ... shall I deal with it now? Electronic mail from the recipient's perspective. International Journal Of Human– Computer Interaction, 21 (3), 313–332. Robinson, M. D., & Clore, G. L. (2002). Belief and feeling: Evidence for an accessibility model of emotional self-report. Psychological Bulletin, 128(6), 934-960. doi: 10.1037/0033-2909.128.6.934 Rozin, P. (2001). Social psychology and science: Some lessons from Solomon Asch. Personality and Social Psychology Review, 5, 2–14. Rubinstein, J. S., Meyer, D. E., & Evans, J. E. (2001). Executive control of cognitive processes in task switching. Journal of Experimental Psychology: Human Perception and Performance, 27(4), 763–797. doi:10.1037//0096-1523.27.4.763 Ryan, R. M. (1982). Control and information in the intrapersonal sphere: An extension of cognitive evaluation theory. Journal of Personality and Social Psychology, 43, 450461. Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68–78. doi:10.1037/0003-066X.55.1.68 Ryan, R.M., & Huta, V. (2009). Wellness as healthy functioning or wellness as happiness: The importance of eudaimonic thinking. Journal of Positive Psychology, 4, 202–204. 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Ryan, R. M., Mims, V., & Koestner, R. (1983). Relation of reward contingency and interpersonal context to intrinsic motivation: A review and test using cognitive evaluation theory. Journal of Personality and Social Psychology, 45, 736-750. Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of well-being. Journal of Personality and Social Psychology, 57, 1069–1081. doi:10.1037/0022-3514.57.6.1069 Ryff, C., & Keys C. L. M. (1995). The structure of well-being revisited. Journal of Personality and Social Psychology, 69(4), 719–727. Soucek, R., & Moser, K. (2010). Coping with information overload in email communication: Evaluation of a training intervention. Computers in Human Behavior, 26(6), 1458–1466. doi:10.1016/j.chb.2010.04.024 Steger, M. F., Frazier, P., Oishi, S., & Kaler, M. (2006). The meaning in life questionnaire: Assessing the presence of and search for meaning in life. Journal of Counseling Psychology, 53(1), 80–93. doi:10.1037/0022-0167.53.1.80 Taylor, H., Fieldman, G., & Altman, Y (2008). E-mail at work: A cause for concern? The implications of the new communication technologies for health, wellbeing and productivity at work. Journal of Organizational Transformation and Social Change, 5(2), 159–173. van der Klink, J. J. L., Blonk, R. W. B., Schene, A. H., & van Dijk, F. J. H. (2001). The benefits of interventions for work-related stress. American Journal of Public Health, 91, 270–276. 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. White, M. P., & Dolan, P. (2009). Accounting for the Richness of Daily Activities. Psychological Science, 20(8), 1000–1008. doi:10.1111/j.1467-9280.2009.02392.x Whittaker, S., Bellotti, V., & Gwizdka, J. (2006). Email and PIM: Problems and possibilities. Communications of the ACM, 49, 68–73. Whittaker, S., Bellotti, V., & Gwizdka, J. (2006). Email and PIM: Problems and possibilities. Communications of the ACM, 49, 68–73. Whittaker, S., & Sidner, C. (1997). Email overload: Exploring personal information management of email. In S. Kiesler (Ed.), Culture of the Internet (pp. 277–295). Mahwah, NJ: Lawrence Erlbaum. Zijlstra, F. R. H., Roe R. A., Leonora A. B., Krediet I. (1999). Temporal factors in mental work: Effects on interrupt activities. Journal of Occupational and Organizational Psychology, 72, 163–185. 29 Figure Click here to download high resolution image