ARTICLE REVIEW DUE IN 3 HOURS!!!

Sleep and Organizational Citizenship Behavior: The Mediating Role of Job Satisfaction

Christopher M. Barnes Virginia Tech

Sonia Ghumman University of Hawai’i

Brent A. Scott Michigan State University

We examine sleep as an important factor beyond the work domain that is relevant to organizational citizenship behavior. In a field study of 87 employees from a variety of organizations, an objective measure of sleep quantity predicted organizational citizenship behavior directed toward organizations but not organizational citizenship behavior directed toward individuals. Additionally, job satisfaction medi- ated this relationship. In a second field study of 85 working college students, we found that natural variation in daily sleep over the course of a work week predicted daily variance in organizational citizenship behavior directed toward both individuals and organizations, and that job satisfaction mediated these relationships. Based on these findings, we discuss theoretical and practical implications of sleep-deprived employees.

Keywords: organizational citizenship behavior, sleep, job satisfaction, work attitudes, helping behavior

Organizational citizenship behavior is defined as behavior not directly recognized by the formal reward system but that which contributes to organizational effectiveness (Smith, Organ, & Near, 1983; see also Organ, Podsakoff, & MacKenzie, 2006). Research has clearly linked organizational citizenship behavior to positive outcomes, such as organizational performance (Podsakoff, Ahearne, & MacKenzie, 1997). To date, theory and research examining organizational citizenship behavior has focused almost entirely on either individual differences such as conscientiousness or work-based antecedents such as organizational commitment (for meta-analyses, see LePine, Erez, & Johnson, 2002; Organ & Ryan, 1995). However, to date this literature has ignored the spillover of nonwork-based antecedents. This may promote the assumption that nonwork activities are not relevant in influencing organiza- tional citizenship behavior.

Sleep is a powerful determinant of human affect, cognition, and behavior (for reviews, see Barnes, 2012; Harrison & Horne, 2000; Lim & Dinges, 2010;). However, outside of a few recent excep- tions (see Barnes & Hollenbeck, 2009; Barnes & Wagner, 2009),

management and applied psychology research has generally ig- nored the effects of sleep on members of organizations. Sleep is often considered by employees to be an unnecessary luxury that motivated people go without (Kirn, 2004). Indeed, the Internet retailer zazzle.com sells coffee mugs and shirts that proclaim “Sleep is for wimps.” Even among those who acknowledge the importance of sleep for employees, a common assumption is that small amounts of lost sleep are not important. However, sleep is a fundamental physiological process required by all individuals.

Although the exact purpose of sleep is still being determined, it clearly has restorative effects (Saper, Scammell, & Liu, 2005). Sleep is a continuous variable, and recent research indicates that minor amounts of sleep loss can have surprisingly powerful ef- fects. For example, losing one night of sleep (Harrison & Horne, 1999), losing as little as five hours of nocturnal sleep (Friedman, Bigger, & Kornfeld, 1971), or restricting sleep to four to five hours per night for a week (Dinges et al., 1997) all can have detrimental consequences such as cognitive errors, performance decrements, and negative moods. In fact, even 40 minutes of lost sleep has been associated with increased work injuries (Barnes & Wagner, 2009).

Perhaps as a result of the above consequences, a small but growing body of management and applied psychology research has begun to consider the effects of smaller, more typical amounts of sleep loss on employees. Primarily, this literature has focused on cognitive (e.g., job satisfaction) and affective (e.g., positive and negative mood) outcomes of sleep (e.g., Barnes & Hollenbeck, 2009; Barnes, Schau- broeck, Huth & Ghumman, 2011; Harrison & Horne, 1999; Scott & Judge, 2006; Sonnentag & Zijlstra, 2006). Although such outcomes are important to consider, to date there is a relative dearth of research examining employee behavior as an outcome of sleep loss. This lack of research is especially disconcerting given both the powerful effects of sleep loss established in research outside of management and applied psychology as well as the prevalence of sleep loss (cf. Bonnet

This article was published Online First November 12, 2012. Christopher M. Barnes, Management Department, Pamplin College of

Business, Virginia Tech; Sonia Ghumman, Management and Industrial Relations, Shidler College of Business, University of Hawai’i at Manoa; Brent A. Scott, Management Department, Eli Broad College of Business, Michigan State University.

We thank Dr. Chaudhry MS Ghumman and the Sleep and Lung Diag- nosis Center for support in collecting the data for this study.

Correspondence concerning this article should be addressed to Christo- pher M. Barnes, Management Department, Pamplin College of Business, Virginia Tech, 2106 Pamplin, Blacksburg, VA 24061. E-mail: [email protected]

Journal of Occupational Health Psychology © 2012 American Psychological Association 2013, Vol. 18, No. 1, 16 –26 1076-8998/13/$12.00 DOI: 10.1037/a0030349

16

& Arand, 1995; Kronholm et al., 2008; Luckhaupt, Tak, & Calvert, 2010; National Sleep Foundation, 2005).

Accordingly, the purpose of this study is to examine the relation- ship between sleep and daily workplace behavior. Specifically, we focus on the relationship between sleep and daily organizational citizenship behavior. We theorize that the extent to which employees go “above and beyond the call of duty” would be especially suscep- tible to a poor night of sleep. In addition, we explain part of the process by which sleep loss is associated with organizational citizen- ship behavior by examining job satisfaction as a mediator of this relationship. In the sections to follow, we provide the conceptual justification for our hypotheses.

Sleep and Organizational Citizenship Behavior in Employees

Sleep can be defined as a state of immobility with greatly reduced responsiveness (Siegel, 2005) and can be distinguished from a coma or anesthesia by its rapid reversibility. Rather than a cessation of neural activity, sleep is more accurately characterized as a reorgani- zation of neural activity (Hobson, 2005). Sleep is a biochemically regulated process, with a homeostatic function that increases the propensity to sleep in proportion to wakefulness and decreases the propensity to sleep in proportion to time spent sleeping (Borbely, 2009; Borbely & Achermann, 1999).

Although the National Sleep Foundation (2009) recommends seven to nine hours of sleep per night for adults, survey data show that many people do not adhere to this recommendation. A recent survey of over 66,000 Americans indicates that 30% get less than six hours of sleep per night (Luckhaupt et al., 2010). Kronholm and colleagues (2009) aggregated a series of seven surveys totaling over a quarter million Finnish participants and found that approximately 10% of the respon- dents slept less than six hours per night. Moreover, people tend to get less sleep during the week than on weekends (National Sleep Foun- dation, 2005), which reflects on average less sleep on work nights than on regular nights and consequently should be of concern to organizations.

As noted above, organizational citizenship behavior is defined as behavior not directly recognized by the formal reward system but that contributes to organizational effectiveness (Organ et al., 2006). Be- cause organizational citizenship behavior is not a required part of task performance and is not directly acknowledged by formal reward systems, employees have discretion in whether or not to engage in it. As Organ and colleagues (Organ et al., 2006; Smith et al., 1983) noted, some employees choose to go above and beyond the call of duty, fulfilling not only their required work tasks but also engaging in extra role behaviors that contribute to the organization’s effectiveness. Such behavior is both distinct from required task performance and critical to organizational performance (Borman & Motowidlo, 1993; Motowidlo & Van Scotter, 1994). To date, the degree to which engagement in organizational citizenship behavior is influenced by sleep has not been empirically examined. As described below, we draw from previous research to contend that sleep influences engage- ment in organizational citizenship behavior via its association with job satisfaction.

Sleep and Job Satisfaction

As noted above, a large body of research indicates that sleep loss is associated with a number of negative outcomes, one being

decrements in cognitive functioning (for reviews, see Barnes, 2012; Harrison & Horne, 2000; Lim & Dinges, 2010). Such cognitive decrements should hinder the attainment of work goals, eliciting negative feelings such as anger and frustration (Frijda, 1988; Lazarus, 1994; Totterdell, Reynolds, Parkinson, & Briner, 1994), as well as lowering feelings of well-being in general (Pilcher, Ginter, & Sadowsky, 1997; Pilcher & Ott, 1998). More- over, given that emotional regulation is more difficult for individ- uals who have been sleep deprived (Dahl & Lewin, 2002), such negative feelings are less likely to be controlled, which eventually may lead to depression (Baranski, Cian, Esquivie, Pigeau, & Raphel, 1998; Caldwell, Caldwell, Brown, & Smith, 2004; Horne, 1993).

Ultimately, to the extent that sleep loss leaves employees angry, frustrated, and less capable of successfully executing work tasks, they should be likely to appraise their jobs as dissatisfying. Research has shown that negative moods are associated with higher levels of job dissatisfaction (Ilies & Judge, 2002), even when those moods are caused by events outside of the job itself (Brief, Butcher, & Roberson, 1995). More specific to the current investigation, Scott and Judge (2006), in an experience-sampling study, found that employees were less satisfied with their jobs on days after a poor night of sleep, and feelings of hostility partially mediated this relationship. Overall then, we expect a positive relationship between an individual’s sleep quan- tity and his or her job satisfaction.

Job Satisfaction and Organizational Citizenship Behavior

As summarized in a recent review conducted by Ilies, Fulmer, Spitzmuller, and Johnson (2009), researchers have proposed that job satisfaction is positively related to organizational citizenship behavior ever since the construct of organizational citizenship behavior was introduced to the literature. The relationship between job satisfaction and organizational citizenship behavior can be explained by theories of social exchange (Blau, 1964; Emerson, 1976; for a review, see Cropanzano & Mitchell, 2005). According to these theories, individ- uals’ voluntary behaviors are motivated by obligations that arise through multiple social transactions. Guiding these transactions is the norm of reciprocity, which drives individuals to respond to others’ treatment in kind (Gouldner, 1960). From this perspective, employees view job satisfaction as a positive outcome of the social exchange relationship with their organization, and they reciprocate with positive inputs (i.e., organizational citizenship behavior).

Meta-analyses have supported the positive relationship between job satisfaction and organizational citizenship behavior (cf. LePine et al., 2002; Organ & Ryan, 1995), including both OCB-I (organizational citizenship behavior directed toward individuals) and OCB-O (orga- nizational citizenship behavior directed toward organizations [Ilies et al., 2009]). Of particular interest is a meta-analysis conducted by Riketta (2008), which examined 16 longitudinal studies to ascertain the causal relationship between job satisfaction and organizational citizenship behavior. Riketta found that extra role performance (a construct which has considerable overlap with organizational citizen- ship behavior) did not influence subsequent job satisfaction, but job satisfaction had a substantial influence of subsequent extra role per- formance, thus showing job satisfaction to be an antecedent of orga- nizational citizenship behavior. Similarly, Fassina, Jones, and Ugger- slev (2008) conducted a meta-analysis that found that job satisfaction

17SLEEP AND ORGANIZATIONAL CITIZENSHIP BEHAVIOR

relates to organizational citizenship behavior in a manner that is largely independent from perceived organizational justice. Fassina et al. note that this is consistent with a social-exchange view of job satisfaction and organizational citizenship behavior.

Given research indicating the influence of sleep on job satisfaction (Scott & Judge, 2006), and the influence of job satisfaction on organizational citizenship behavior (Riketta, 2008), we hypothesize that sleep quantity will be positively related to organizational citizen- ship behavior directed toward both individuals and organizations via its influence on job satisfaction. In particular, we expect that employ- ees who are low in sleep quantity will appraise their jobs as less satisfying, which will impede the direction of organizational citizen- ship behaviors toward their colleagues as those employees delay the reciprocation of social exchanges to a time when their well-being is better. In contrast, those with high levels of sleep will appraise their jobs as more satisfying, which will facilitate the direction of organi- zational citizenship behaviors toward their colleagues as those em- ployees’ well-being motivates them to reciprocate social exchanges with coworkers in a more immediate manner. Similarly, individuals low in sleep quantity, and thus job satisfaction, will feel less obligated to perform organizational citizenship behaviors directed toward the organization compared to individuals high in sleep quantity. Overall then, we propose the following hypotheses:

Hypothesis 1: Sleep quantity is positively associated with organizational citizenship behavior directed toward individu- als (OCB-I).

Hypothesis 2: Sleep quantity is positively associated with organizational citizenship behavior directed toward organiza- tions (OCB-O).

Hypothesis 3: Job satisfaction mediates the relationship be- tween sleep quantity and organizational citizenship behavior directed toward individuals (OCB-I).

Hypothesis 4: Job satisfaction mediates the relationship be- tween sleep quantity and organizational citizenship behavior directed toward organizations (OCB-O).

Overview of Studies

We conducted two field studies to examine the influence of sleep quantity on organizational citizenship behavior and job satisfaction. Study 1 measured sleep quantity objectively in a sleep clinic setting and used self-reported ratings of organizational citizenship behavior and job satisfaction. Study 2 used a within-persons, experience sam- pling method in which self-reported measures of sleep quantity were measured before a given work shift and organizational citizenship behavior and job satisfaction were measured at the end of a given work shift over a course of a work week.

Study 1: Method

Participants

Participants were 87 individuals (29 female) who voluntarily sought treatment at a sleep clinic located in the East coast of the United States of America. The sleep clinic collects data from individ- uals who seek treatment for difficulties sleeping in the process of

diagnosing the reasons for patients’ sleep difficulties and treating their sleep disorders. The mean age of the respondent was 47.32 years (SD � 10.68 years). This clinic routinely collects objective data on sleep, providing a unique opportunity to both measure sleep objec- tively and to conduct surveys on work behavior. A requirement of study participation was holding full-time employment.

Procedure

We collected data from two independent sources. First, laboratory data from the sleep clinic on participants’ sleep quantities were collected by experienced sleep clinicians who were unfamiliar with the study’s hypotheses. Second, a questionnaire assessing job satis- faction and organizational citizenship behavior was completed by participants after their work shift following their visit to the sleep clinic.

In the process of diagnosing sleep difficulties, the sleep clinicians collected objective data on sleep during a laboratory study in which participants slept overnight at the clinic. Patients were instructed to arrive at the sleep clinic on the night of the sleep clinic lab study no later than 9 p.m. After reviewing the protocol of the clinic laboratory procedures, they were placed in a bedroom in the sleep clinic. Par- ticipants were then observed throughout the remainder of the night and until 7:00 a.m. the following morning. Participants were not provided any of their sleep data until a scheduled appointment that occurred after the questionnaire.

After their night in the sleep laboratory, participants were provided with the self-report questionnaire measuring job satisfaction and or- ganizational citizenship behavior. Only participants who were sched- uled to work a shift the day after their night in the sleep laboratory were invited to participate. Participants were instructed to complete the measure after the end of their work shift that same work day, and they were provided with a stamped and addressed return envelope. As an incentive, potential participants were informed that those who participated would be entered into a drawing for $500. Of the 400 questionnaires that were handed out, 87 were completed and returned, for a response rate of 21.8%.

Measures

Sleep quantity1. The sleep literature indicates a multitude of ways to measure sleep (Barnes, 2012). Subjective measures are con- venient and correlate well with objective measures (Barnes et al., 2011). However, objective sleep measurement is still considered to be the most accurate and precise way to measure sleep. In Study 1, we used an objective polysomographic measure of sleep, focused on measurement of brain electrical activity.

As a part of the visit to the sleep clinic, participants wore an Electroencephalogram (EEG) device which was linked to a computer in a separate room where the sleep technicians monitored the EEG data being recorded. EEG devices record brain electrical activity through electrodes that are attached to the individual’s scalp. EEG recordings are displayed as waves on the computer, which are used to define whether a person is awake or asleep and the different stages of sleep (Horne, 1988; Pressman, 2002). Beta and Alpha waves are

1 Sleep quality was measured, but when controlling for sleep quality in a separate analysis, it did not influence results. Sleep quality also did not have clear and consistent effects on the variables of interest in this study.

18 BARNES, GHUMMAN, AND SCOTT

present in the awakened state and drowsy state, respectively. Sleep is characterized by Theta and Delta waves. The EEG computer software calculated these different types of brain waves and used them to differentiate between wakefulness and sleep states. This was calcu- lated in minutes.

Job satisfaction. A five-item scale was used to measure job satisfaction (Scott & Judge, 2006, adapted from Brayfield & Rothe, 1951). Participants used a five-point scale (1 � strongly disagree, 5 � strongly agree) to indicate how satisfied they felt with their job. Participants were instructed to rate their job satisfaction from that specific workday (the day immediately after their night in the sleep laboratory). Sample items include “I feel fairly satisfied with my present job” and “I consider my job rather unpleasant.” The coeffi- cient alpha for this scale was 0.78.

Organizational citizenship behavior. Organizational citizen- ship behavior was measured by the 16-item scale developed by Lee and Allen (2002); eight items each reflected organizational citizenship behavior directed toward individuals (OCB-I) and organizational cit- izenship behavior directed toward the organization (OCB-O). Partic- ipants were instructed to rate their behavior from that specific work- day (the day immediately after their night in the sleep laboratory). Respondents used a five-point scale (1 � to a very little extent, 5 � to a very great extent) to indicate how often they engaged in these behaviors during the day following their night in the laboratory. The coefficient alphas for these scales were 0.93 and 0.93, respectively. Sample items for organizational citizenship behavior directed toward individuals were “Gave up time to help others who had work or nonwork problems” and “Assisted others with their duties.” Sample items for organizational citizenship behavior directed toward organi- zations include “Offered ideas to improve the functioning of the organization” and “Took actions to protect the organization from potential problems.”

Caffeine intake. Because research indicates that caffeine can alter the effects of sleep quantity on a variety of outcomes (Lieber- man, Tharion, Shukitt-Hale, Speckman, & Tully, 2002), we included caffeine intake as a control variable. At the beginning of the labora- tory study, participants were asked to report whether or not they had consumed any caffeine in the previous 24 hours.

Age. Because research indicates that people at different ages have different needs for sleep (Kramer, Kerkhof, & Hofman, 1999), we included age as an additional control variable.

Study 1: Results

Descriptive Statistics and Correlations

Table 1 shows the means, standard deviations, and correlations among the measured variables in Study 1. Participants in this study slept an average of five and a half hours, with a standard deviation of a little less than an hour and a half. This mean was approxi- mately two hours less than the population estimate, with a standard deviation approximately 20 minutes less than the population esti- mate (Bureau of Labor Statistics, 2010)—a point we return to in the discussion. As indicated by Table 1, the zero order correlations between sleep and OCBI (�.02) and OCBO (.17) were not statis- tically significant.

Tests of Hypotheses

As indicated by Tables 2 and 3, regression analyses indicate that sleep had a positive relationship with subsequent job satisfaction (� � .29, p � .01), and job satisfaction had a positive relationship with OCB-I (� � .42, p � .01) and OCB-O (� � .57, p � .01). These relationships are not only consistent with previous research, but they also point in the direction of mediation (formally tested below).

To formally test our hypotheses, we used ordinary least squares multiple regression analyses. Hypothesis 1 indicated that we ex- pected a positive relationship between sleep quantity and OCB-I. As shown in Table 2, the standardized regression coefficient for the relationship between sleep quantity and OCB-I was .06 (p � .05), which was in the direction hypothesized but not significant. Thus, Hypothesis 1 was not supported, in that there was not a significant direct effect of sleep on OCB-I. Hypothesis 2 indicated that we expected a positive relationship between sleep quantity and OCB-O. As shown in Table 2, the standardized regression coeffi- cient for the relationship between sleep quantity and OCB-O was .24 (p � .05), which was in the direction hypothesized and significant. Hypothesis 2 was supported, in that there was a sig- nificant direct effect of sleep on OCB-O. Thus, although there was not a significant zero-order correlation between sleep and OCB-O, there was a significant relationship when controlling for age and caffeine.

Hypothesis 3 indicated that we expected job satisfaction to mediate the relationship between sleep quantity and OCB-I. To test this hypothesis, we examined the indirect effect of sleep on OCB with a bias-corrected bootstrap 95% confidence interval analysis of the indirect effect (Preacher & Hayes, 2008). A bias-corrected bootstrap 95% confidence interval analysis of the indirect effect of sleep on OCB-I indicated a significant indirect effect that excluded zero, with a lower bound of .0004 and an upper bound of .0035. Thus, although there was not a significant direct effect of sleep on OCB-I, Hypothesis 3 was supported in that there was a significant indirect effect of sleep quantity on OCB-I. See Table 3 for results.

Hypothesis 4 indicated that we expected job satisfaction to mediate the relationship between sleep quantity and OCB-O. A bias-corrected bootstrap 95% confidence interval analysis of the indirect effect of sleep on OCB-O indicated a significant indirect effect that excluded zero, with a lower bound of .0004 and an upper bound of .0038. Thus, hypothesis 4 was supported, in that

Table 1 Study 1 Means, Standard Deviations, and Correlations

Variable Mean SD 1 2 3 4 5

1. Sleep quantity 331.41 86.23 2. Caffeine 0.46 0.5 .08 3. Age 47.32 10.68 �.22� �.08 4. OCB-I 3.4 1.03 �.02 �.14 .09 5. OCB-O 3.59 0.97 .17 �.01 .05 .73��

6. Job satisfaction 3.53 0.9 .15 �.15 .06 .45�� .56��

Note. n � 87. REM � Rapid Eye Movement; NREM � Non-Rapid Eye Movement; OCB-I � Organizational Citizenship Behavior directed toward individuals; OCB-O � Organizational Citizenship Behavior directed to- ward organizations. � p � .05. �� p � .01.

19SLEEP AND ORGANIZATIONAL CITIZENSHIP BEHAVIOR

there was a significant indirect effect of sleep quantity on OCB-O. Moreover, when controlling for job satisfaction, there was no longer a significant effect of sleep on OCB-O. See Table 3 for results.

Study 1: Discussion

In Study 1, we found a direct effect of sleep on OCB-O but not OCB-I. However, sleep had significant indirect effects on both OCB-I and OCB-O through the mediator of job satisfaction. This highlights the relationship between sleep as a nonwork based factor that is relevant to organizational citizenship behavior, indi- cating a spillover effect from home to work. One limitation of this study was the fact that having participants sleep in a laboratory setting may have influenced their sleep patterns. Research indi- cates some small differences in sleep habits when sleeping alone in comparison to sleeping next to another person. In many cases, especially with women, people sleep better next to their partners than they do alone (Dittami et al., 2007). Moreover, it is possible that sleeping in an unfamiliar context while under observation and hooked up to EEG devices may be disruptive to sleep. Indeed, participants in this study had a restricted range in their sleep quantity, such that our test was a conservative test of the hypoth- eses. Thus, our lack of support for Hypothesis 1 may have been a Type II error.

Another limitation of this study is that not everyone who was invited to participate in the study chose to participate. Given that participation was solicited at the sleep clinic, we do not have information regarding which people chose not to participate. If people who chose not to participate had a different relationship between sleep and their work attitudes and behaviors than did those who participated, such a difference was not captured in our data. However, participants from Study 2 were drawn from a subject pool, where this was much less of a concern.

A third limitation is that there are potential chronic or person- level variables that could potentially be relevant to determining sleep and/or OCB that were not captured in the Study 1 sample. However, Study 2 was designed to hold constant such variables, thus addressing this limitation.

A strength of this study was the objective measure of sleep. Moreover, the participants in this sample were from a broad array of organizations and jobs. However, participants in this study were recruited from a group of people seeking treatment for sleep problems, as highlighted by the fact that their sleep means were lower than population estimates. Although a restriction in range in sleep should statistically work against our findings and make it more difficult to detect a relationship between sleep and OCB, the fact that our sample suffered from sleep problems may still call into question the generalizability of our results. Accordingly, we designed Study 2 to see whether our findings would generalize to a sample of working adults who were not seeking treatment for sleep issues. Moreover, in Study 2 we used a daily diary method- ology in order to examine the within-individual effect of sleep on OCB. Such a design holds constant any individual differences or stable work characteristics that might influence both sleep and OCB.

Study 2: Method

Participants

Eighty-five university students (53 female) enrolled in under- graduate management courses at a large university in the western U.S. participated in exchange for course credit. The mean age of the participants was 23.5 years, and 53% of them were Asian, 12% were Caucasian, 2% were Pacific Islander, 2% were Mexican American, 5% were “Other,” and the remainder chose not to answer. To be eligible to participate, students had to be employed

Table 2 Study 1 Regression Results Predicting Job Satisfaction, OCB-I, and OCB-O

Job satisfaction OCB-I OCB-O

Step 1 Step 2 Step 1 Step 2 Step 1 Step 2

Variable � t � t � t � t � t � t

Caffeine �.15 �.38 �.20 �1.89 �.14 �1.29 �.14 �1.35 .00 �.02 �.04 �.38 Age .04 .36 .10 .99 .08 .75 .09 .84 .05 .49 .10 .95 Sleep quantity .29�� 2.73 .06 .51 .24� 2.17 R2 .02 .10� .03 .03 .00 .05�

�R2 .08� .00 .05�

Note. n � 87. Caffeine was coded 0 for those who had no coffee or tea and 1 for those who had coffee or tea. OCB-I � Organizational Citizenship Behavior directed toward individuals; OCB-O � Organizational Citizenship Behavior directed toward organizations. � p � .05. �� p � .01.

Table 3 Study 1 Regression Results Predicting OCB-I and OCB-O

OCB-I OCB-O

Step 1 Step 2 Step 1 Step 2

Variable � t � t � t � t

Caffeine �.07 �.67 �.05 �.51 .09 1.04 .08 .83 Age .05 .55 .04 .37 .02 .17 .03 .37 Job Satisfaction .42�� 4.30 .44�� 4.33 .57�� 6.32 .55�� .79 Sleep Quantity �.08 �.72 .09 .88 R2 .20�� .20�� .32�� .32��

�R2 .01 .01

Note. n � 87. Caffeine was coded 0 for those who had no coffee or tea and 1 for those who had coffee or tea. OCB-I � Organizational Citizenship Behavior directed toward individuals; OCB-O � Organizational Citizen- ship Behavior directed toward organizations. � p � .05. �� p � .01.

20 BARNES, GHUMMAN, AND SCOTT

in their current position for at least 3 months. Participants held a diverse set of positions across many organizations.

Procedure

This study used an experience sampling method design consist- ing of a series of 11 online surveys that were completed across six days. After completing a consent form and an entry demographic survey on the first day, participants completed two surveys daily over the course of five workdays. For each of the five workdays (not necessarily consecutive workdays), participants completed a preshift survey measuring their sleep quantity before their work shift. Additionally, at the end of each of their five workdays, participants completed a postshift survey measuring their job sat- isfaction and organizational citizenship behavior.

Although 95 participants originally signed up for the study, 10 participants provided one or less days’ worth of surveys and were thus removed from all analyses. Additionally, 12 matched preshift/ postshift surveys were removed from the analyses because they were completed more than a day apart, as indicated by timestamps collected with the submission of each survey. Eighty-Five partic- ipants completed the 330 remaining matched preshift/postshift surveys, resulting in an overall response rate of 78% of those remaining in the study.

Measures

Sleep quantity. At the beginning of participants’ work shifts, self-reported sleep was measured using the Pittsburgh Sleep Diary (Monk et al., 1994). This measure asks participants to report the time at which they went to bed the previous night, how long it took them to fall asleep, how many minutes they spent awake through- out the night after initially falling asleep, and at what time they awoke in the morning. These times were used to obtain a measure of the number of minutes participants spent sleeping that night.

Job satisfaction. At the end of their work shift, participants indicated how satisfied they felt with their job on that specific day on each of the five target days. As in Study 1, we used the five-item Scott & Judge (2006) job satisfaction scale, adapted from Brayfield and Rothe (1951). The average coefficient alpha across the five measurements was .82.

Organizational citizenship behavior. At the end of their workshift, participants reported how often they engaged in orga- nizational citizenship behavior on that specific day on each of their

five target workdays at the end of their work shift. As in Study 1, we used the 16-item scale developed by Lee and Allen (2002) to measure organizational citizenship behavior, with 8 items measur- ing organizational citizenship behavior directed at individuals (OCB-I) and 8 items measuring organizational citizenship behav- ior directed at organizations (OCB-O). The average coefficient alphas for these two measures were .86 and .92, respectively.

Bed time and wake time. Babkoff, Caspy, Mikulincer, and Sing (1991) recommend separating circadian rhythm effects from the effects of sleep quantity. Circadian rhythm effects entail peaks and troughs in various physiological processes (including alertness and affect) that can either mask or confound the effects of sleep quantity. Indeed, contemporary models of sleep explicitly include circadian rhythms in parallel to sleep quantity (Hursh et al., 2004; Jewett & Kronauer, 1999). To control for circadian rhythm effects in sleep/wake patterns, we included as control variables both the time of day participants went to bed (bed time) and the time of day participants woke up that day (wake time). This was taken from the Pittsburgh Sleep Diary (Monk et al., 1994) at the beginning of participants’ work shift. Time of day was indexed as the number of hours elapsed since midnight.

Study 2: Results

Descriptive Statistics and Correlations

Table 4 shows the means, standard deviations, and correlations among the measured variables in Study 2. Participants in Study 2 slept an average of seven hours with a standard deviation of 126 minutes, which is comparable to the population mean of 7.6 hours and population standard deviation of 111 minutes (Bureau of Labor Statistics, 2010).

Tests of Hypotheses

To account for the multilevel nature of our data (because days were nested within employees), we used hierarchical linear mod- eling (HLM; see Raudenbush & Bryk, 2002) to test our hypothe- ses. At level 1 were the repeated, daily observations of the vari- ables in this study. There were no substantive level 2 variables; level 2 was included to account for nonindependence of observa- tions nested within participants.

Before testing our hypotheses, we inspected the percentage of variance lying at the within-individual and between-individual

Table 4 Study 2 Means, Standard Deviations, and Correlations

Mean SD 1 2 3 4 5 6

1. Sleep quantity 420 126 �.53�� .51�� �.03 .00 .04 2. Bed time 24.4 1.3 �.80�� .45�� .07 .13 .049 3. Wake time 7.7 1.4 .45�� .14 .05 .15 .09 4. Job Satisfaction 3.3 0.7 .06 �.05 0 .35�� .41��

5. OCB-I �.06 .06 �.06 .24�� .78��

6. OCB-O �.05 .03 �.07 .32�� .58��

Note. Correlations below the diagonal are within-individual and correlations above the diagonal are between-individual. OCB-I � Organizational Citizenship Behavior directed toward individuals; OCB-O � Organizational Citizenship Behavior directed toward organizations. n � 330 observations nested within 85 individuals. � p � .05. �� p � .01.

21SLEEP AND ORGANIZATIONAL CITIZENSHIP BEHAVIOR

levels for our level 1 variables. Null models in HLM, with no level 1 or level 2 predictors, revealed that a substantial portion of the variance was within-person. Specifically, 81% of the variance in sleep quantity, 77% of the variance in bed time, 43% of the variance in wake time, 30% of the variance in job satisfaction, and 30% of the variance in OCB-I, and 20% of the variance in OCB-O was within-person. These analyses suggest that multilevel model- ing was appropriate for our analyses and that there is sufficient within-person variance to predict. As shown in Table 5, there was a positive relationship between sleep and job satisfaction (� � .26, p � .05). Moreover, as shown in Table 6, there was a positive relationship between job satisfaction and OCB-I (� � .28, p � .01) and between job satisfaction and OCB-O (� � .40, p � .01). These results are consistent with previous research and also consistent with the between-individual results in Study 1. Moreover, they point in the direction of mediation (formal mediation test below).

Hypothesis 1 indicated that sleep quantity would have a positive relationship with OCB-I. As indicated by Table 6, sleep from the previous night (measured preshift) had a positive relationship with OCB-I engaged in that day (measured at the end of the same shift, � � .48, p � .01). Hypothesis 2 indicated that sleep quantity would have a positive relationship with OCB-O. As indicated by Table 6, sleep from the previous night (measured preshift) had a positive relationship with OCB-O engaged in that day (measured at the end of the same shift, � � .34, p � .01).

Hypothesis 3 indicated that job satisfaction would mediate the influence of sleep on OCB-I, and Hypothesis 4 indicated that job satisfaction would mediate the influence of sleep on OCB-O. Sobel (1982) tests indicated a significant indirect effect (z � 2.02, p � .05) of sleep on OCB-I, and a significant indirect effect (z � 2.14, p � .05) of sleep on OCB-O. This provides support for Hypotheses 3 and 4. The effects of sleep on both OCB-I and OCB-O were still significant even when controlling for job satisfaction, indicating the possibility of other additional mediators.

Our results indicate the importance of following the recommen- dation of Babkoff et al. (1991) to separate the effects of circadian rhythms from sleep quantity. The zero-order correlations, which allow for these effects to mask each other, were not significant. However, controlling for circadian rhythms allowed us to detect an effect of sleep on OCB that was both statistically and practically significant.

Study 2: Discussion

In Study 2, we found that sleep had direct effects on OCB-I and OCB-O. Moreover, we found that sleep also had indirect effects on both OCB-I and OCB-O through the mediator of job satisfaction.

These findings are largely consistent with our findings from Study 1 and support our mediated model of sleep, job satisfaction, and OCB. Furthermore, they highlight the dynamic nature of sleep and OCB and important within-individual effects.

The strengths and weaknesses of Study 2 are offset by those of Study 1, and vice versa. Whereas Study 1 relied on an objective measure of sleep, Study 2 used self-reports of sleep (separated in time from OCBs). Whereas Study 1 drew from a sample of people seeking treatment for sleep problems, Study 2 was drawn from a sample of working adults. Whereas Study 1 utilized a design where only one observation was collected per person, Study 2 used an experience-sampling design that allowed us to better control for potential between-person confounds (e.g., individual differences, stable work characteristics, etc.) by testing our hypotheses strictly at the within-person level.

One limitation of Study 2 is that all of our variables were derived from the same source (the focal participant), which raises the potential for common method variance. However, by tempo- rally separating the daily measures of sleep quantity from the daily measures of job satisfaction and OCB, and by individually mean- centering the daily measures, we avoided some potential sources of common-method variance (e.g., measurement context effects and common rater effects such as response tendencies; see Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). That being said, we could not avoid all potential sources of common-method variance, and thus future research able to incorporate independent sources of data (e.g., OCB assessed by one’s supervisor) into experience- sampling designs could determine whether our results hold.

A second limitation is that these participants were all college students. It is important to note that working in a paid job was a requirement for participation in this study. Nevertheless, college students do not represent all working adults. However, it is helpful to note that findings from this student sample were largely con- sistent with findings from Study 1, which was not composed of students. Future research should extend this topic to multiple types of samples to help further broaden the generalizability of our findings.

General Discussion

Consistent with previous research, we found that sleep quantity influenced job satisfaction, and that job satisfaction influenced organizational citizenship behavior. Moreover, we revealed part of the process by which sleep quantity influences OCB by finding evidence that job satisfaction mediated this relationship across two studies. Although significant indirect effects of sleep quantity on both types of OCB through job satisfaction were found in both studies, and significant direct effects of sleep on OCB-O were found in both studies, sleep influenced OCB-I directly only in Study 2. We see several potential reasons for this. First, Study 1 examined these relationships at the between-person level, whereas Study 2 examined these relationships at the within-person level. It therefore may be that the relationship between sleep quantity and OCB-I is stronger at the within-person level. In addition, Study 1 used an objective measure of sleep quantity, whereas Study 2 used a more subjective measure of sleep quantity. It may also be that the relationship between sleep quantity and OCB-I is stronger when more subjective measures are used. Of course, the differences that we observed may also be attributable to sample-related differences

Table 5 Study 2 Within-Individual Effects of Sleep On Job Satisfaction

Variable � t

Intercept 3.31�� 46.83 Bed time .25� 2.36 Wake time �.26� �2.44 Sleep quantity .26� 2.49

Note. n � 330 observations nested within 85 individuals. � p � .05. �� p � .01.

22 BARNES, GHUMMAN, AND SCOTT

(i.e., those with sleeping problems vs. more normal working adults). Future research is therefore needed to sort out the condi- tions under which sleep is more or less strongly associated with OCB.

Implications

There is a dearth of research in the management and applied psychology literatures examining the topic of sleep. Recent theo- retical extensions, through the use of entrainment theory (Ancona & Chong, 1996), the Sleep Activity Fatigue and Task Effective- ness (SAFTE) depletion model of fatigue (Hursh et al., 2004), and Affective Events Theory (Weiss & Cropanzano, 1996), have be- gun to bring the topic of sleep into management and applied psychology theories (cf. Barnes & Van Dyne, 2009; Barnes & Wagner, 2009; Scott & Judge, 2006; Wagner, Barnes, Lim, & Ferris, 2012).

We continue that trend by including sleep as a predictor of organizational citizenship behavior. Existing research examining sleep in the workplace has tended to focus primarily on cognitive and affective outcomes. We extend this research to examine be- haviors occurring in organizations. Indeed, whereas many manag- ers in organizations may assume that the amounts of sleep expe- rienced by their employees may not be relevant in determining organizationally relevant outcomes, we find that even natural variation in sleep can still have important effects on employee behaviors. This is complementary to previous research outside the domain of management and applied psychology indicating that even small amounts of sleep loss have important effects on indi- viduals (e.g., Dinges et al., 1997).

As noted in the introduction, theory examining organizational citizenship behavior has primarily focused on either individual differences or work-based antecedents (LePine et al., 2002; Organ & Ryan, 1995). This literature has largely ignored nonwork-based antecedents of organizational citizenship behavior, especially those that might be dynamic in nature. Our results indicate that nonwork-based factors can indeed be important predictors of or- ganizational citizenship behavior, with sleep as one such important factor. This indicates that theory examining organizational citizen- ship behavior should be extended to consider not only sleep but also potentially other dynamic nonwork-based factors. Indeed, future research would do well to integrate the larger work-family

spillover literature with the organizational citizenship behavior literature.

Future Research

One limitation across both studies in this article is that organi- zational citizenship behavior was captured through self-report. Although a benefit of using a self-report measure is that partici- pants are aware of more of their own organizational citizenship behavior than are others who may not observe them 100% of the time, participants may nevertheless overestimate their own behav- iors or engage in impression management. Participants were aware that none of their colleagues or supervisors would see the data associated with this study, minimizing any incentives for such impression management. Nevertheless, future research examining this topic should consider measuring organizational citizenship behavior through sources other than self-report.

A related limitation is that although sleep was measured tem- porally antecedent to job satisfaction and organizational citizen- ship behavior, job satisfaction and organizational citizenship be- havior were measured at the same time. This limits the degree to which causality between job satisfaction and organizational citi- zenship behavior can be determined in our study. Future research should temporally separate job satisfaction and organizational cit- izenship behavior. However, given a recent meta-analysis indicat- ing that job satisfaction is temporally antecedent to organizational citizenship behavior rather than the other way around (Riketta, 2008), we expect similar causality in future replications of our research.

We encourage researchers to replicate our findings. Our studies were conducted with two very different samples and had generally consistent results. However, to further explore the generalizability of our findings, future research should replicate these effects in other settings and samples, and with other methods. However, in addition to replication, our studies highlight several other potential avenues for future research. First, researchers should give greater attention to spillover between home and work in determining behaviors such as OCB. Our research indicated that sleep is one important such variable to consider when examining OCB. It is likely that there are other work states and behaviors that are important, such as strain experienced at home or recovery activi- ties engaged in at home (Fritz, Yankelevich, Zarubin, & Barger,

Table 6 Study 2 Within-Individual Effects of Sleep and Job Satisfaction on OCB-I and OCB-O

OCB-I OCB-O

Step 1 Step 2 Step 1 Step 2

Variable � t � t � t � t

Intercept 2.67�� 28.88 2.67�� 28.88 2.46�� 22.35 2.46�� 22.35 Bed time .49�� 3.07 .43�� 2.85 .35� 2.51 .12 1.42 Wake time �.51�� �3.02 �.44�� �2.80 �.38� �2.52 �.15 �1.62 Sleep quantity .48�� 3.03 .41�� 2.73 .34� 2.51 .10 1.2 Job satisfaction .28�� 3.32 .40�� 3.94

Note. n � 330 observations nested within 85 individuals; OCB-I � Organizational Citizenship Behavior directed toward individuals; OCB-O � Organizational Citizenship Behavior directed toward organizations. � p � .05. �� p � .01.

23SLEEP AND ORGANIZATIONAL CITIZENSHIP BEHAVIOR

2010; Sonnentag, Kuttler, & Fritz, 2010). Given that contemporary research is finding many links between home and work life, it is likely that there are many such linking mechanisms, potentially including both cognitive and affective pathways.

Second, researchers should consider other outcomes of sleep that are relevant to organizations. Given that sleep influences affect and job satisfaction at work, there are likely other organi- zational variables that are influenced by sleep. Previous research indicates that job satisfaction is related to turnover (George & Jones, 1996). Future research may reveal that there is a dynamic component to turnover intentions and behaviors associated with turnover. It is reasonable to expect that, though the mediator of job satisfaction, sleep may be related to the dynamic component of turnover intentions and turnover behaviors. A similar relationship may be found with organizational commitment and identification, such that low sleep leads to low job satisfaction, which in turn leads to low organizational commitment and identification.

Third, future research should consider additional mediators of the relationship between sleep and organizational citizenship be- havior.

One such possibility is mood; given established effect of sleep on affect (Bower, Bylsma, Morris, & Rottenberg, 2010; Son- nentag, Binnewies, & Mojza, 2008; Franzen, Siegle, & Buysse, 2008) and affect on job satisfaction (Ilies & Judge, 2002; Ilies, Scott, & Judge, 2006), it may be that mood mediates the effect of sleep on job satisfaction. Such future examinations of more detail between steps in our model will aid in a deeper understanding of the topic.

Whereas our article has focused on overnight sleep, future research should examine sleep at other times of the day (such as naps). Although previous research indicates that the overwhelming majority of human sleep is overnight (Lavie, 1986, 1997; Soehner, Kennedy, & Monk, 2007), naps, which were not captured in our study, may be shown to have important effects on both workplace attitudes and behaviors. Indeed, it may be that allowing employees to take naps during work hours leads to higher levels of organi- zational citizenship behavior. Research examining the tradeoffs between lost labor and improved workplace behavior may reveal that naps are financially defensible.

Future research should examine moderators of the effects noted in this article. It may be that there are individual differences that influence the relationship between sleep and organizational citi- zenship behavior. Sleep physiology research indicates that sleep needs change as people age (Kramer et al., 1999), and that per- sonality traits such as extraversion influence the effects of sleep deprivation (Killgore, Richards, Killgore, Kamimori, & Balkin, 2007). Thus, age and extraversion may moderate the relationship between sleep and organizational citizenship behavior.

Future research should examine sleep over longer periods of time. Previous research examining accumulated sleep loss and sleep debt indicates that it is reasonable to expect that sleep loss over more than just the preceding night would be relevant in determining work attitudes and work behavior (Horne, Anderson, & Platten, 2008; Rupp, Wesensten, Bliese, & Balkin, 2009). Re- searchers may find that chronic sleep deprivation is even more disruptive than acute sleep deprivation, and that two nights of short sleep in a row lead to especially strong decrements in job satis- faction and organizational citizenship behavior. Sleep debt is often reflected in sleep latency, which is how long it takes to fall asleep

(Johns, 1991). Similarly, sleep quality may have important effects on the relationships indicated by our model, including interrup- tions to sleep, and time spent awake while trying to sleep (Barnes, 2012).

Finally, future research should consider organizationally rele- vant antecedents of sleep. Recent research highlights the fact that work schedules can cut into sleep (Barnes, Wagner, & Ghumman, in press). Other research may find that job demands, job control, and work-related stress may serve as additionally important ante- cedents of sleep. Indeed, such research may reveal feedback loops between work and sleep that can explain the buildup of chronic sleep debt. As the literature linking work and sleep grows, re- searchers should focus efforts on potential interventions intended to mitigate some of the negative effects of low sleep quantity (cf. Barnes, 2011). Such knowledge could be applied to improve the effectiveness of organizations and the well-being of employees.

References

Ancona, D., & Chong, C. L. (1996). Entrainment: Pace, cycle, and rhythm in organizational behavior. Research in Organizational Behavior, 18, 251–284.

Babkoff, H., Caspy, T., Mikulincer, M., & Sing, H. C. (1991). Monotonic and rhythmic influences: A challenge for sleep deprivation research. Psychological Bulletin, 109, 411– 428. doi:10.1037/0033-2909.109.3 .411

Baranski, J. V., Cian, C., Esquivie, D., Pigeau, R. A., & Raphel, C. (1998). Modafinil during 64 hr of sleep deprivation: Dose-related effects on fatigue, alertness, and cognitive performance. Military Psychology, 10, 173–193. doi:10.1207/s15327876mp1003_3

Barnes, C. M. (2011). “I’ll sleep when I’m dead”: Managing those too busy to sleep. Organizational Dynamics, 40 (1), 18 –26. doi:10.1016/j.orgdyn .2010.10.001

Barnes, C. M. (2012). Working in our sleep: Sleep and self-regulation in organizations. Organizational Psychology Review, 2, 234 –257, doi: 10.1177/2041386612450181

Barnes, C. M., & Hollenbeck, J. R. (2009). Sleep deprivation and decision- making teams: Burning the midnight oil or playing with fire? Academy of Management Review, 34, 56 – 66. doi:10.5465/AMR.2009.35713280

Barnes, C. M., Schaubroeck, J. M., Huth, M., & Ghumman, S. (2011). Lack of sleep and unethical behavior. Organizational Behavior and Human Decision Processes, 115, 169 –180. doi:10.1016/j.obhdp.2011 .01.009

Barnes, C. M., & Van Dyne, L. (2009). ‘I’m tired’: Differential effects of physical and emotional fatigue on workload management strategies. Human Relations, 62, 57–90. doi:10.1177/0018726708099518

Barnes, C. M., & Wagner, D. T. (2009). Changing to daylight saving time cuts into sleep and increases workplace injuries. Journal of Applied Psychology, 94, 1305–1317. doi:10.1037/a0015320

Barnes, C. M., Wagner, D. T., & Ghumman, S. (in press). Borrowing from sleep to pay work and family: Expanding time-based conflict to the broader non-work domain. Personnel Psychology.

Blau, P. M. (1964). Exchange and power in social life. New York, NY: Wiley.

Bonnet, M. H., & Arand, D. L. (1995). We are chronically sleep deprived. Sleep, 18, 908 –911.

Borbely, A. A. (2009). Refining sleep homeostasis in the two-process model. Journal of Sleep Research, 18, 1–2. doi:10.1111/j.1365-2869 .2009.00750.x

Borbely, A. A., & Achermann, P. (1999). Sleep homeostasis and models of sleep regulation. Journal of Biological Rhythms, 14, 557–568.

Borman, W. C., & Motowidlo, S. J. (1993). Expanding the criterion

24 BARNES, GHUMMAN, AND SCOTT

domain to include elements of contextual performance. In N. Schmitt & W. C. Borman (Eds.), Personnel selection in organizations (pp. 71–98). San Francisco, CA: Jossey-Bass.

Bower, B., Bylsma, L. M., Morris, B. H., & Rottenberg, J. (2010). Poor reported sleep quality predicts low positive affect in daily life among healthy and mood-disordered persons. Journal of Sleep Research, 19, 323–332. doi:10.1111/j.1365-2869.2009.00816.x

Brayfield, A. H., & Rothe, H. F. (1951). An index of job satisfaction. Journal of Applied Psychology, 35, 307–311. doi:10.1037/h0055617

Brief, A. P., Butcher, A. H., & Roberson, L. (1995). Cookies, disposition, and job-attitudes: The effects of positive mood-inducing events and negative affectivity on job satisfaction in a field experiment. Organiza- tional Behavior and Human Decision Processes, 62, 55– 62. doi: 10.1006/obhd.1995.1030

Bureau of Labor Statistics. (2010). American time use survey. News release USDL-0855.

Caldwell, J. A., Caldwell, J. L., Brown, D. L., & Smith, J. K. (2004). The effects of 37 hours of continuous wakefulness on the physiological arousal, cognitive performance, self-reported mood, and simulator flight performance of F-117A pilots. Military Psychology, 16, 163–181. doi: 10.1207/s15327876mp1603_2

Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review. Journal of Management, 31, 874 –900. doi: 10.1177/0149206305279602

Dahl, R. E., & Lewin, D. S. (2002). Pathways to adolescent health: Sleep regulation and behavior. Journal of Adolescent Health, 31, 175–184. doi:10.1016/S1054-139X(02)00506-2

Dinges, D. F., Pack, F., Williams, K., Gillen, K. A., Powell, J. W., Ott, G. E., . . . Pack, A. I. (1997). Cumulative sleepiness, mood disturbance, and psychomotor vigilance performance decrements during a week of sleep restricted to 4 –5 hours per night. Sleep, 20, 267–277.

Dittami, J., Keckeis, M., Machatschke, I., Katina, S., Zeitlhofer, J., & Kloesch, G. (2007). Sex differences in the reactions to sleeping in pairs versus sleeping alone in humans. Sleep and Biological Rhythms, 5, 271–276. doi:10.1111/j.1479-8425.2007.00320.x

Emerson, R. M. (1976). Social exchange theory. Annual Review of Soci- ology, 2, 335–362. doi:10.1146/annurev.so.02.080176.002003

Fassina, N. E., Jones, D. A., & Uggerslev, K. L. (2008). Meta-analytic tests of the relationships between organizational justice and citizenship be- havior: Testing agent-system, agent-dominance, and shared-variance. Journal of Organizational Behavior, 29, 805– 828. doi:10.1002/job.494

Franzen, P. L., Siegle, G. J., & Buysse, D. J. (2008). Relationships between affect, vigilance, and sleepiness following sleep deprivation. Journal of Sleep Research, 17, 34 – 41. doi:10.1111/j.1365-2869.2008.00635.x

Friedman, R. C., Bigger, J. T., & Kornfeld, D. S. (1971). Interns and sleep loss. New England Journal of Medicine, 285, 201–210. doi:10.1056/ NEJM197107222850405

Frijda, N. H. (1988). The laws of emotion. American Psychologist, 43, 349 –358. doi:10.1037/0003-066X.43.5.349

Fritz, C., Yankelevich, M., Zarubin, A., & Barger, P. (2010). Happy, healthy, and productive: The role of detachment from work during nonwork time. Journal of Applied Psychology, 95, 977–983. doi: 10.1037/a0019462

George, J. M., & Jones, G. R. (1996). The experience of work and turnover intentions: Interactive effects of value attainment, job satisfaction, and positive mood. Journal of Applied Psychology, 81, 318 –325. doi: 10.1037/0021-9010.81.3.318

Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25, 161–178. doi:10.2307/2092623

Harrison, Y., & Horne, J. A. (1999). One night of sleep loss impairs innovative thinking and flexible decision making. Organizational Be- havior and Human Decision Processes, 78, 128 –145. doi:10.1006/obhd .1999.2827

Harrison, Y., & Horne, J. A. (2000). The impact of sleep deprivation on decision making: A review. Journal of Experimental Psychology- Applied, 6, 236 –249. doi:10.1037/1076-898X.6.3.236

Hobson, J. A. (2005). Sleep is of the brain, by the brain and for the brain. Nature, 437, 1254 –1256. doi:10.1038/nature04283

Horne, J., Anderson, C., & Platten, C. (2008). Sleep extension versus nap or coffee, with the context of ‘sleep debt.’. Journal of Sleep Research, 17, 432– 436.

Horne, J. A. (1988). Why we sleep: The functions of sleep in humans and other mammals. Oxford, England: Oxford University press.

Horne, J. A. (1993). Human sleep, sleep loss and behavior - Implications for the prefrontal cortex and psychiatric-disorder. British Journal of Psychiatry, 162, 413– 419. doi:10.1192/bjp.162.3.413

Hursh, S. R., Redmond, D. P., Johnson, M. L., Thorne, D. R., Belenky, G., Balkin, T. J., . . . Eddy, D. R. (2004). Fatigue models for applied research in warfighting. Aviation Space and Environmental Medicine, 75, A44 – A53.

Ilies, R., Fulmer, I. S., Spitzmuller, M., & Johnson, M. D. (2009). Person- ality and citizenship behavior: The mediating role of job satisfaction. Journal of Applied Psychology, 94, 945–959. doi:10.1037/a0013329

Ilies, R., & Judge, T. A. (2002). Understanding the dynamic relationships among personality, mood, and job satisfaction: A field experience sam- pling study. Organizational Behavior and Human Decision Processes, 89, 1119 –1139. doi:10.1016/S0749-5978(02)00018-3

Ilies, R., Scott, B. A., & Judge, T. A. (2006). The interactive effects of personal traits and experienced states on interaindividual patterns of citizenship behavior. Academy of Management Journal, 49, 561–575. doi:10.5465/AMJ.2006.21794672

Jewett, M. E., & Kronauer, R. E. (1999). Interactive mathematical models of subjective alertness and cognitive throughput in humans. Journal of Biological Rhythms, 14, 588 –597. doi:10.1177/074873099129000920

Johns, M. W. (1991). A new method for measuring daytime sleepiness. Sleep, 14, 540 –545.

Killgore, W. D. S., Richards, J. M., Killgore, D. B., Kamimori, G. H., & Balkin, T. J. (2007). The trait of introversion-extraversion predicts vulnerability to sleep deprivation. Journal of Sleep Research, 16, 354 – 363. doi:10.1111/j.1365-2869.2007.00611.x

Kirn, W. (2004). Sleep is for sissies. Time, 164, 56. Kramer, C. J., Kerkhof, G. A., & Hofman, W. F. (1999). Age differences

in sleep-wake behavior under natural conditions. Personality and Indi- vidual Differences, 27, 853– 860. doi:10.1016/S0191-8869(99)00034-3

Kronholm, E., Partonen, T., Laatikainen, T., Peltonen, M., Härmä, M., Hublin, C., . . . Sutela, H. (2008). Trends in self-reported sleep duration and insomnia-related symptoms in Finland from 1972 to 2005: A com- parative review and re-analysis of Finnish population samples. Journal of Sleep Research, 17, 54 – 62. doi:10.1111/j.1365-2869.2008.00627.x

Kronholm, E., Sallinen, M., Suutuma, T., Sulkava, R., Era, P., & Partonen, T. (2009). Self report sleep duration and cognitive functioning in the general population. Journal of Sleep Research, 18, 436 – 446. doi: 10.1111/j.1365-2869.2009.00765.x

Lavie, P. (1986). Ultrashort sleep-waking schedule 3—Gates and forbidden zones for sleep. Electroencephalography and Clinical Neurophysiology, 63, 414 – 425. doi:10.1016/0013-4694(86)90123-9

Lavie, P. (1997). Melatonin: Role in gating nocturnal rise in sleep propen- sity. Journal of Biological Rhythms, 12, 657– 665. doi:10.1177/ 074873049701200622

Lazarus, R. S. (1994). Appraisal: The long and short of it. In P. Ekman & R. J. Davidson (Eds.), The nature of emotion: 208 –215. New York, NY: Oxford University Press.

Lee, K., & Allen, N. J. (2002). Organizational citizenship behavior and workplace deviance: The role of affect and cognitions. Journal of Applied Psychology, 87, 131–142. doi:10.1037/0021-9010.87.1.131

LePine, J. A., Erez, A., & Johnson, D. E. (2002). The nature and dimen- sionality of organizational citizenship behavior: A critical review and

25SLEEP AND ORGANIZATIONAL CITIZENSHIP BEHAVIOR

meta-analysis. Journal of Applied Psychology, 87, 52– 65. doi:10.1037/ 0021-9010.87.1.52

Lieberman, H. R., Tharion, W. J., Shukitt-Hale, B., Speckman, K. L., & Tully, R. (2002). Effects of caffeine, sleep loss, and stress on cognitive performance and mood during US Navy SEAL training. Psychophar- macology, 164, 250 –261. doi:10.1007/s00213-002-1217-9

Lim, J., & Dinges, D. F. (2010). A meta-analysis of the impact of short- term sleep deprivation on cognitive variables. Psychological Bulletin, 136, 375–389. doi:10.1037/a0018883

Luckhaupt, S. E., Tak, S., & Calvert, G. M. (2010). The prevalence of short sleep duration by industry and occupation in the national health inter- view survey. Sleep, 33, 149 –159.

Monk, T. H., Reynolds, C. F., Kupfer, D. J., Buysse, D. J., Coble, P. A., Hayes, A. J., . . ., & Ritenour, A. M. (1994). The Pittsburgh Sleep Diary. Journal of Sleep Research, 3, 111–120. doi:10.1111/j.1365-2869.1994 .tb00114.x

Motowidlo, S. J., & Van Scotter, J. R. (1994). Evidence that task perfor- mance should be distinguished from contextual performance. Journal of Applied Psychology, 79, 475– 480. doi:10.1037/0021-9010.79.4.475

National Sleep Foundation. (2009). Sleep in America poll. Washington, DC: Author.

Organ, D. W., Podsakoff, P. M., & MacKenzie, S. B. (2006). Organiza- tional citizenship behavior: Its nature, antecedents, and consequences. Sage: Thousand Oaks, CA.

Organ, D. W., & Ryan, K. (1995). A meta-analytic review of attitudinal and dispositional predictors of organizational citizenship behavior. Per- sonnel Psychology, 48, 775– 802. doi:10.1111/j.1744-6570.1995 .tb01781.x

Pilcher, J. J., Ginter, D. R., & Sadowsky, B. (1997). Sleep quality versus sleep quantity: Relationships between sleep and measures of health, well-being, and sleepiness in college students. Journal of Psychosomatic Research, 42, 583–596. doi:10.1016/S0022-3999(97)00004-4

Pilcher, J. J., & Ott, E. (1998). The relationships between sleep and measures of health and well- being in college students: A repeated measures approach. Behavioral Medicine, 23, 170 –178. doi:10.1080/ 08964289809596373

Podsakoff, P. M., Ahearne, M., & MacKenzie, S. B. (1997). Organizational citizenship behavior and the quantity and quality of work group perfor- mance. Journal of Applied Psychology, 82, 262–270. doi:10.1037/0021- 9010.82.2.262

Podsakoff, P. M., MacKenzie, S. M., Lee, J., & Podsakoff, N. P. (2003). Common method variance in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychol- ogy, 88, 879 –903. doi:10.1037/0021-9010.88.5.879

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple me- diator models. Behavior Research Methods, 40, 879 – 891. doi:10.3758/ BRM.40.3.879

Pressman, M. R. (2002). Primer of polysomnogram interpretation. Boston, MA: Butterworth Heinemann.

Raudenbush, S. W., & Bryk, A. S. (2002). Applications and data analysis methods, 2nd edition. Sage: Thousand Oaks.

Riketta, M. (2008). The causal relation between job attitudes and perfor- mance: A meta-analysis of panel studies. Journal of Applied Psychology, 93, 472– 481. doi:10.1037/0021-9010.93.2.472

Rupp, T. L., Wesensten, N. J., Bliese, P. D., & Balkin, T. J. (2009). Banking sleep: Realization of benefits during subsequent sleep restric- tion and recovery. Sleep, 32, 311–321.

Saper, C. B., Scammell, T. E., & Lu, J. (2005). Hypothalamic regulation of sleep and circadian rhythms. Nature, 437, 1257–1263. doi:10.1038/ nature04284

Scott, B. A., & Judge, T. A. (2006). Insomnia, emotions, and job satisfac- tion: A multilevel study. Journal of Management, 32, 622– 645. doi: 10.1177/0149206306289762

Siegel, J. M. (2005). Clues to the functions of mammalian sleep. Nature, 437, 1264 –1271. doi:10.1038/nature04285

Smith, C. A., Organ, D. W., & Near, J. P. (1983). Organizational citizen- ship behavior: Its nature and antecedents. Journal of Applied Psychol- ogy, 68, 653– 663. doi:10.1037/0021-9010.68.4.653

Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp. 290 –312). San Francisco, CA: Jossey-Bass.

Soehner, A. M., Kennedy, K. S., & Monk, T. H. (2007). Personality correlates with sleep–wake variables. Chronobiology International, 24, 889 –903. doi:10.1080/07420520701648317

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. doi:10.1037/0021-9010 .93.3.674

Sonnentag, S., Kuttler, I., & Fritz, C. (2010). Job stressors, emotional exhaustion, and need for recovery: A multi-source study on the benefits of psychological detachment. Journal of Vocational Behavior, 76, 355– 365. doi:10.1016/j.jvb.2009.06.005

Sonnentag, S., & Zijlstra, F. R. H. (2006). Job characteristics and off-job activities as predictors of need for recovery, well-being, and fatigue. Journal of Applied Psychology, 91, 330 –350. doi:10.1037/0021-9010 .91.2.330

Totterdell, P., Reynolds, S., Parkinson, B., & Briner, R. B. (1994). Asso- ciations of sleep with everyday mood, minor symptoms and social interactions. Sleep, 17, 466 – 475.

Wagner, D. T., Barnes, C. M., Lim, V., & Ferris, D. L. (2012). Lost sleep and cyberloafing: Evidence from the laboratory and a Daylight Saving Time quasi-experiment. Journal of Applied Psychology, 97, 1068 –1076. doi:10.1037/a0027557

Weiss, H. M., & Cropanzano, R. (1996). Affective events theory: A theoretical discussion of the structure, causes and consequences of affective experiences at work. Research in Organizational Behavior, 18, 1–74.

Received January 5, 2012 Revision received August 10, 2012

Accepted August 28, 2012 �

26 BARNES, GHUMMAN, AND SCOTT