sleep and stress

JOURNAL OF RESEARCH ON ADOLESCENCE, 21(2), 342 – 349

Rise and Fall of Sleep Quantity and Quality With Student Experiences Across the First Year of University
Nancy L. Galambos and Andrea L. Howard

Jennifer L. Maggs

University of Alberta

The Pennsylvania State University

Covariations of self-reported sleep quantity (duration) and quality (disturbances) with affective, stressful, academic, and social experiences across the first year of university in 187 Canadian students (M age 5 18.4) were examined with multilevel models. Female students reported sleeping fewer hours on average than did male students. In months when negative affect and general levels of stress were higher, sleep quantity was lower. Poorer sleep quality was seen in students living away from home and reporting more financial stress at baseline. In addition, sleep quality was poorer in months when negative affect and general levels of stress were higher (attenuating the effect of financial stress) and better in months when students spent more days with friends. Three themes are presented to explore the mechanisms by which sleep quantity and quality rise and fall in tandem with experiences of the first year of university.

Poor sleep quantity (e.g., shorter sleep times) and quality (e.g., waking up in the night) are related to lower academic performance among adolescents of all ages (Wolfson & Carskadon, 2003). At no other time is this relation more relevant than when adolescents enter their first year of university and face new and more intense academic demands at the same time as they may experience changes in family and peer relations, finances, and living arrangements. Indeed, up to 70% of university students have regular sleep difficulties (Buboltz, Brown, & Soper,
2001), and first-year students have greater odds of being poor sleepers than second-year students (Suen,
Hon, & Tam, 2008). Partial sleep deprivation (e.g., o5 hours sleep in 24 hours, Pilcher & Huffcutt, 1996) and delayed sleep phase syndrome (i.e., difficulty in getting to sleep and waking up at normal times; F. C.
Brown, Soper, & Buboltz, 2001) are not uncommon among university students. The practice of pulling an ‘‘all nighter’’ is associated with lower grade point averages (GPAs; Thacher, 2008), and less sleep and lower quality sleep are associated with poorer mental health in university students (Buboltz et al., 2006).
To learn more about sleep as it occurs naturally during a demanding transitional period, we observed how ebbs and flows in sleep quantity and quality are accompanied by affective, stressful, academic, and social experiences as students negotiate their way through the first year of university.
In a recent study of students in their first semester of university, day-to-day covariations of sleep
This research was supported by a Social Sciences and Humanities
Research Council of Canada grant to N. Galambos and J. Maggs.
Requests for reprints should be sent to Nancy L. Galambos,
Department of Psychology, University of Alberta, Edmonton, AB,
Canada T6G 2E9. E-mail: galambos@ualberta.ca

quantity and quality with other experiences (affect, stress, academic demands, and social events) were observed over 2 weeks (Galambos, Dalton, & Maggs,
2009). Higher sleep quantity (sleeping longer) was associated with decreased negative affect and spending more time on schoolwork the next day. Higher sleep quality on a given night was related to higher positive affect, lower negative affect, and lower stress the next day. In addition, several daily experiences predicted variation in students’ sleep. For example, drinking alcohol that day and expecting to take a test the next day predicted lower sleep quantity that night, and higher positive affect during the day predicted better sleep quality that night. These results were similar to those of a daily experience study of ninthgrade students (Fuligni & Hardway, 2006). Less sleep at night was associated with higher anxiety, depressed mood, and fatigue the next day, whereas more time studying and higher stress during the day predicted less sleep that night. These rare repeated-measures studies of naturally occurring behaviors show that on a day-to-day basis the extent to which students are sleeping well may carry over into their academic and psychological functioning and vice versa.
Daily experience studies are useful for learning about relations between variables as they co-occur in the short term, but leave open questions about similar relations over the longer term. The first year of college or university in particular is an important transitional time in which adjusting to a new environment may interfere with sleep, impacting academic performance and well-being (Buboltz et al., 2006). To our know-

r 2010 The Authors
Journal of Research on Adolescence r 2010 Society for Research on Adolescence
DOI: 10.1111/j.1532-7795.2010.00679.x

RISE AND FALL OF SLEEP

ledge, no study has examined covariations between sleep and student experience in multiple assessments across the first year of university, yet such a study is important for understanding how sleep is related to behavior and well-being during this period.
In the current study, modeled after the daily experience study of Ga-lambos et al. (2009), we followed 187 university students across five occasions from October to April of their first year. Given that
Galambos and colleagues found relations of affect, stress, academic demands, and social experiences with sleep on a daily level, the current study also asked for reports of sleep, affect, stress, academic effort, and social behavior, but these reports were designed to cover a broader span of time. In the original daily study, between-persons factors such as gender and living situation were not related to average sleep patterns (Galambos and colleagues), but they may be more influential in explaining variations in sleep across the longer period of one academic year. In this study we asked: (1) To what extent do positive and negative affect, stress, academic effort, and social experiences (i.e., spending time with friends and alcohol use) covary with sleep quantity and quality across months? (2) Do gender, living situation, social support, financial stress, and high school GPA predict average levels of sleep quantity and quality across the first year of university?
METHOD
Participants
Participants were 187 full-time first year students at a large Canadian university participating in Making the
Transition II, a web-based study of health behaviors and academic performance. Mean age at study outset was
18.4 years (SD 5 .44, range 5 17.3 – 19.8). Ethnicity was
72% White, 13% Asian, 6% mixed, 3% Indo-Canadian
(originating from India), and 5% another visible minority (e.g., Black, Arabic). Half (52%) lived with parents, 28% in campus residences, 14% in an apartment alone or with roommates, and 5% with relatives. Most students (85%) lived in two-parent homes while growing up, and the majority of students’ mothers (73%) and fathers (74%) had completed college (i.e., 2-year degree) or university (i.e., 4-year degree or higher).
Comparative data suggest that the sample is a good cross section of first-year students in the university.
Procedure
In fall 2005, 198 participants were recruited from compulsory first-year English classes serving most

343

undergraduate students; Engineering students were recruited separately due to different English requirements. Interested students meeting the criteria
(full-time, first year in any college or university, under age 20) attended an initial group session at which they completed consent forms and pen-andpaper baseline questionnaires. Beginning in October
2005, participants completed up to seven monthly, web-based questionnaires for which they were offered CDN$10 each. An additional CDN$5 was offered for completing the final questionnaire in
April, 2006. Each questionnaire was made available to participants for 1 week at the beginning of each month; questionnaires could be completed anytime during that week. Data collection did not take place during scheduled midterm or final exam periods.
Retention was high, with 88% of the 198 participants completing at least four questionnaires. Sleep measures were administered in October, November, December, February, and April; therefore, analyses for the current study relied on data from these months.
Ten of the original participants were excluded from analysis due to missing sleep measures on all occasions, and one was excluded due to multiple missing between-persons predictors at baseline, resulting in the final sample of 187. Six missing data points for between-persons predictors (four cases in which one baseline predictor was missing; one case in which two predictors were missing) were replaced using a regression-based estimation procedure. Recruitment of the sample continued past administration of the
October web-based questionnaire, resulting in an n of 100 for October.
Measures
Sleep quantity and quality. The Pittsburgh Sleep
Quality Index (PSQI; Buysse, Reynolds, Monk,
Berman, & Kupfer, 1989) is a widely used measure of subjective sleep experience in the previous month, consisting of seven subscales (components) and a global score. Scores on two components were selected for the current study. Sleep quantity (or duration) was measured by asking ‘‘During the past month, how many hours of actual sleep did you get at night? (This may be different than the number of hours you spend in bed.)’’ The range in mean sleep duration across waves was small (M 5 6.66, SD 5 1.18, to M 5 6.90, SD 5 1.18). Sleep quality was assessed with the sleep disturbances component. A stem question asked: ‘‘During the past month, how often have you had trouble sleeping because you . . .’’ followed by nine problems (e.g., ‘‘wake up in the middle of the

344

GALAMBOS, HOWARD, AND MAGGS

night or early morning,’’ ‘‘had bad dreams,’’ ‘‘have pain’’) rated on a 4-point scale ranging from 0 (not during the past month) to 3 (three or more times a week).
Scores for the nine items were summed. Sums were assigned values: 0 (sum of 0), 1 (sum from 1 to 9), 2
(sum from 10 to 18), and 3 (sum from 19 to 27). Mean sleep disturbances across waves ranged from 1.03
(SD 5 .46) to 1.14 (SD 5 .56). This component was selected because it is an indicator of sleep quality that draws on concrete experiences and is sensitive to variability between and within persons. The duration and disturbances components of the PSQI have been shown to distinguish healthy controls from clinical samples (i.e., those with sleep, depressive, or posttraumatic stress disorders; Backhaus, Junghanns,
Broocks, Riemann, & Hohagen, 2002; Buysse et al.,
1989; Calhoun et al., 2007).
Between-persons predictors (baseline measures). Gender was coded as female 5 0 (n 5 114), male 5 1 (n 5 73). Living situation was coded as living with parents 5 0 (n 5 98) versus living away from parents 5 1 (n 5 89). Social support was measured with the mean of seven items (Galambos,
Barker, & Krahn, 2006). Participants were asked ‘‘When you have problems, how much can you rely on each of the following people for help?’’ ‘‘Mother,’’ ‘‘father,’’
‘‘other family members,’’ ‘‘spouse, boyfriend, girlfriend,’’ ‘‘friends,’’ ‘‘people at work,’’ and ‘‘others’’ were options. The extent to which each source provided social support was rated on a scale ranging from 0 (have no such person or not at all) to 3 (very much).
Coefficient a was not computed, as it is not expected that individuals with high support from one source would necessarily have high support from other sources (M 5 1.63, SD 5 .46, range 5 .43 – 2.86).
Financial stress was measured with the mean of eight items from Pearlin and Schooler’s (1978) household economic stress measure, which asks, ‘‘When you think of your current financial situation, how do you feel?’’ followed by feelings such as ‘‘worried’’ and ‘‘tense.’’
Items were rated on a 4-point scale ranging from 1 (not at all) to 4 (very), and were averaged (M 5 1.89,
SD 5 .68, range 5 1.00 – 3.88). Coefficient a was .92.
High school GPA, used for admission into university, was obtained from the Registrar and was a percentage score (M 5 85.30, SD 5 5.97, range 5 70.40 – 97.20).
Within-person predictors (time-varying covariates). The Positive and Negative Affect Schedule
(Watson, Clark, & Tellegen, 1988) was adapted to capture affective experience over the previous 2 weeks.
Participants were asked: ‘‘Over the last 14 days, on how many days did you feel . . .?’’ The number of days

reported for each of 10 positive affect items (e.g., interested, proud) was summed, with a possible range from 0 (no days in which any of 10 positive emotions were experienced) to 140 (experienced all 10 positive emotions every day for 14 days). Positive affect means across waves ranged from 62.66
(SD 5 32.10) to 72.04 (SD 5 29.36). The number of days reported for each of 10 negative affect items
(e.g., distressed, hostile) was summed. Across waves, negative affect ranged from 30.79 (SD 5 21.60) to 36.51
(SD 5 24.08). Coefficient as ranged across waves from
.92 to .95 for positive affect and .88 to .93 for negative affect. A 14-day period was used because recall of emotions over 14 days is accurate within 1 or 2 days
(N. R. Brown, Williams, Barker, & Galambos, 2007).
Stress was measured with the four-item version of the Perceived Stress Scale (Cohen, Kamarck, &
Mermelstein, 1983). Participants were asked, ‘‘Over the last 14 days, how often have you . . .’’ followed by items such as ‘‘felt that you were unable to control the important things in your life’’ and ‘‘felt confident about your ability to handle personal problems’’
(reverse coded). Participants rated these items on a
5-point scale, ranging from 0 (never) to 4 (very often), and ratings were averaged. Across waves, stress scores ranged from a mean of 1.79 (SD 5 .63) to 1.95
(SD 5 .74). Coefficient as ranged across waves from
.67 to .80.
Participants’ academic effort was measured with the mean of two items which asked: ‘‘Over the last 14 days, on how many days did you . . . work as hard as you could on your schoolwork’’ and ‘‘. . . avoid doing any schoolwork at all (other than attending classes).’’
The latter item was reverse scored. Coefficient a was
.62 for the two-item scale. Mean academic effort ranged from 8.23 (SD 5 3.20) to 9.69 (SD 5 3.01).
There were two measures of students’ social experiences. One item assessed the frequency with which students socialized with friends: ‘‘Over the last
14 days, on how many days did you get together with friends (for example, for coffee, a movie, a party, etc.)?’’ Students got together with friends an average of 4.76 (SD 5 3.48) to 5.43 (SD 5 3.72) days in a 2-week period. Alcohol use was assessed by asking ‘‘Over the last 14 days, on how many days have you had alcoholic beverages to drinkFmore than just a few sips?’’ Average days of alcohol use in 14 days ranged from 1.31 (SD 5 1.75) to 1.84 (SD 5 2.49).
RESULTS
There were no significant correlations among between-persons predictors. Within-person intercorrelations (po.05) showed that more negative affect

RISE AND FALL OF SLEEP

days were associated with more stressful (r 5.66) and alcohol use days (r 5.13), and with fewer positive affect (r 5 À.22) and academic effort days (r 5 À.19).
More positive affect days were related to fewer stressful days (r 5 À .45) and more academic effort
(r 5.22), socializing (r 5.32), and alcohol use days
(r 5. 11). Additionally, a higher number of stressful days was associated with fewer days of academic effort (r 5 À.18) and socializing (r 5 À.15); more academic effort days were related to fewer days of alcohol use (r 5 À.18) and socializing (r 5 À .15; all po.05). Finally, socializing and drinking were associated on the same days (r 5.30, po.05).
Multilevel models (HLM 6.06; Raudenbush &
Bryk, 2002) examined between-persons (stable) and within-person (time-varying) predictors of sleep quantity (Table 1) and quality (Table 2). First, unconditional means models (not shown) containing no between-persons predictors or time-varying covariates determined the proportions of between-persons and within-person variance in sleep quantity and quality. Second, effects of gender, living situation, social support, financial stress, and high school GPA
(assessed at baseline) on the intercept (the sleep

345

measure averaged across months) were examined
(Model 1). Third, affective (positive, negative), stressful, academic, and social (socialized, used alcohol) experience variables were included as withinperson predictors in Models 2 through 5, respectively, to examine how sleep covaried with these experiences across months, controlling for betweenpersons differences. Concerns about having sufficient statistical power and including more covariates than time points led us to test separate models rather than an integrative model including all covariates
(Singer & Willett, 2003). Between-persons and withinperson predictors were grand-mean centered, except for dichotomous variables. Full information maximum likelihood estimation was used to generate parameter estimates and to preserve cases containing within-person missing values. Intercepts in all models were estimated as randomly varying across persons, and the slopes of all time-varying covariates were specified as nonrandomly varying.
Likelihood-ratio tests assessed the significance of the differences in fit between the unconditional means model and Model 1 and between Model 1 and
Models 2 through 5.

TABLE 1
Results of Multilevel Models Predicting Covariation of Sleep Quantity and Affective, Stressful, Academic, and Social Experiences Across
Months, Controlling for Between-Persons Effects on the Intercept

Model 1
Fixed Effect

Coeff

Intercept (average)
6.60Ã
Between-persons effects on intercept
Gendera
0.33Ã b Living situation
0.06
Social support
0.12
Financial stress
À 0.20Ã
High school GPA
0.00
Recent experiences
(time-varying covariates)
Positive affect
Negative affect
Stress
Academic effort
Socialized
Used alcohol w2 (df)
10.58 (5)
Comparison model
UnMs

Model 2 Affect
SE

Model 3 Stress

Coeff

Coeff

SE

SE

Model 4
Academic
Coeff

Model 5 Social
SE

Coeff

SE

.12

6.63Ã

.12

6.61Ã

.11

6.61Ã

.12

6.62Ã

.12

.14
.14
.12
.11
.01

0.28Ã
0.05
0.06
À 0.12
À 0.00

.14
.15
.12
.12
.01

0.30Ã
0.06
0.04
À 0.13
0.00

.14
.14
.13
.12
.01

0.31Ã
0.05
0.13
À 0.22
0.01

.14
.14
.13
.11
.01

0.30Ã
0.05
0.15
À 0.21
0.00

.14
.15
.12
.11
.01

0.00
À 0.01Ã

.00
.00

À 0.23Ã

.08
À 0.02

.01

83.47Ã (2)
Model 1

127.06Ã (1)
Model 1

5.53Ã (1)
Model 1

Note. Coeff 5 unstandardized coefficient; SE 5 standard error; UnMs 5 unconditional means model; N 5 187. a Male 5 1. b Away from parents 5 1.
Ãpo.05.

À 0.02
.02
0.03
.02
25.44Ã (2)
Model 1

346

GALAMBOS, HOWARD, AND MAGGS

TABLE 2
Results of Multilevel Models Predicting Covariation of Sleep Quality and Affective, Stressful, Academic, and Social Experiences Across
Months, Controlling for Between-Persons Effects on the Intercept
Model 1
Fixed Effect

Coeff

Model 2 Affect
SE

.04
Intercept (average)
1.04Ã
Between-persons effects on intercept
Gendera
À 0.05
.05
Living situationb
0.13Ã
.05
Social support
0.05
.04
Financial stress
0.10Ã
.04
High school GPA
À 0.00
.00
Recent experiences
(time-varying covariates)
Positive affect
Negative affect
Stress
Academic effort
Socialized
Used alcohol w2 (df)
21.94Ã (5)
Comparison model
UnMs

Model 3 Stress

Coeff

Coeff

SE

SE

Model 4 Academic
Coeff

SE

Model 5 Social
Coeff

SE

1.05Ã

.04

1.04Ã

.04

1.04Ã

.04

1.04Ã

.04

À 0.07
0.12Ã
0.05
0.07
À 0.00

.05
.05
.04
.04
.00

À 0.03
0.13Ã
0.08Ã
0.07
À 0.00

.05
.05
.04
.04
.00

À 0.06
0.13Ã
0.05
0.10Ã
À 0.00

.05
.05
.04
.04
.00

À 0.06
0.13Ã
0.06
0.10Ã
À 0.00

.05
.05
.04
.04
.00

À 0.00
0.00Ã

.00
.00

0.08Ã

.03
À 0.01

.01

50.78Ã (2)
Model 1

73.24Ã (1)
Model 1

9.53Ã (1)
Model 1

À 0.01Ã
.00
0.01
.01
13.34Ã (2)
Model 1

Note. Coeff 5 unstandardized coefficient; SE 5 standard error; UnMs 5 unconditional means model; N 5 187. A higher score on sleep quality indicates poorer quality (i.e., higher sleep disturbance). a Male 5 1. b Away from parents 5 1.
Ãpo.05.

Sleep Quantity
The unconditional means model determined that 45% of the variation in sleep quantity was within the person and 55% was between persons. Models 1 through
5 showed that gender was a reliable predictor of sleep quantity, with men reporting that they slept longer hours than did women. Greater financial stress was a significant between-persons predictor of fewer average sleep hours only in Model 1. Turning to the withinperson models, negative affect and stress were significant negative predictors of variation in sleep quantity across months. That is, students reported sleeping fewer hours in months they experienced more negative affect and in months they experienced higher stress. Model 1 did not provide a better fit to the data than did the unconditional means model, but Models 2 through 5 fit the data better than did Model 1.
Sleep Quality
The unconditional means model showed that 67% of the variation in sleep quality was within the person whereas 33% was between persons. Living away from home was a robust predictor of more frequent sleep disturbances. In Model 3 only, higher social

support predicted more sleep disturbance, an unexpected finding. Financial stress was a significant predictor of lower sleep quality (i.e., higher sleep disturbances). However, this association was not observed with negative affect (Model 2) and perceived stress (Model 3) in the models, both of which were related to more sleep disturbance. Sleep quality was also lower in months that students socialized on fewer days (Model 5). Model 1 fit the data better than did the unconditional means model, and Models 2 through 5 all fit better compared with Model 1.
DISCUSSION
Adaptation to increased academic demands, new social horizons, and the semiautonomous living provided by the transition to university bring significant opportunities and challenges as the late adolescent progresses toward adulthood (Schulenberg
& Maggs, 2002). Obtaining a sufficient quantity of high quality sleep is necessary for optimal academic performance, physical health, and psychological well-being, yet too few students new to university get adequate sleep (Buboltz et al., 2006). Fluctuations in sleep quantity and quality in the first year of

RISE AND FALL OF SLEEP

university may reflect a process of adaptation as students go about learning how to cope with new demands. Observing how personal circumstances upon entering university predict variations in sleep across the first year of university provides a window into understanding the adaptational process that may be useful for identifying mediational targets for programs designed to facilitate this transition and improve retention. Furthermore, learning how sleep quantity and quality rise and fall from month to month in tandem with affective, stressful, academic, and social experiences improves understanding of the possible short-term costs and benefits of sleep behaviors. Using a within-person design following students multiple times across their first year of university, this study provides new insights into predictors of sleep quantity and quality. Three major themes are suggested.
First, university students who were more independent from parental care experienced lower quality sleep. That is, those living away from home and those experiencing greater burdens due to financial stress had more sleep disturbances. Possible mechanisms underlying these associations include external impediments to sleep such as living in shared and potentially noisy accommodations or needing to work longer hours in paid employment, as well as internal states, such as homesickness (Beck, Taylor, &
Robbins, 2003) or anxiety about paying bills (Furr,
Westefeld, McConnell, & Jenkins, 2001). Greater anxieties about finances or leaving home could also explain why female students slept less than male students, a gender difference consistent with epidemiological and university student research showing a higher prevalence of insomnia and other sleep difficulties among women than men (Buboltz et al., 2001;
Buysse et al., 2008; Tsai & Li, 2004). One practical implication for universities is to introduce living options such as substance-free and noise controlled residences to reduce impediments to sleep. An implication for students and parents is to put sleep into the equation when they discuss the costs and benefits of living away from home and of incurring the need for employment simultaneous with postsecondary education. During economic recessions, such tradeoffs are likely to be widespread, underscoring the need for reliable student financial aid.
Second, times with more negative affect and stress coincided with times of less positive sleep. Withinperson associations (i.e., intercorrelations) among the time-varying predictors, similarly, showed that when negative affect and stressful experiences were higher, academic effort was lower. Moreover, alcohol use was greater in months with more negative affect.

347

This suggests that students experience difficulties in multiple areas of their lives simultaneously. Campus health, residence life, and academic staff should be cognizant of the potential pile-up of difficulties during students’ first year. Although we cannot disentangle cause and effect, students may need assistance to prevent or overcome downward spirals when stressors, negative affect, drinking to cope, and inadequate sleep co-occur. Initial results from a social support intervention designed to ease the transition to university have been promising; students who participated reported less depression, more social support, and higher adjustment to university
(Pratt et al., 2000).
Third, positive experience (e.g., positive affect, academic effort) generally did not predict sleep quantity or quality, but there was covariation of sleep quality with socializing. Several explanations can be advanced. On the one hand, postsecondary students rank social and academic goals as most important, and the majority of their time is allocated to these pursuits (Maggs, 1997). The lack of associations of sleep with social and academic experiences may suggest that students generally manage these demands successfully, at least as evidenced by their sleep quantity and quality over the longer term. Alternatively, like other health domains such as nutrition and exercise, sleep may be a behavior for which enough is enough. That is, once sufficient rest is obtained, positive experiences may not co-occur with higher quality sleep. Finally, the effect size linking positive experiences with sleep may simply be more modest. Some results were consistent with the daily diary study of Galambos et al. (2009), specifically, that sleep, negative affect, and stress go hand-in-hand. In the daily diary study, negative affect and stress increased on days when sleep was of lower quantity or quality the night before. That sleep quantity and quality were associated in the same direction at the monthly level speaks to the robustness of the relations among sleep, negative affect, and stress. These daily and monthly associations may be capturing the process by which sleep relates to mental health difficulties over the longer term. If days of poor sleep run into months of poor sleep, the cumulative effects are likely to be dire. These results underscore the seriousness of inadequate sleep among university students. The positive association between socializing and better sleep quality echoes the daily diary finding that socializing preceded a longer night’s sleepFbut is counter to the finding that more sleep at night was associated with reduced socializing the next day. The

348

GALAMBOS, HOWARD, AND MAGGS

relation between positive sleep and activities with friends, whether on a daily or monthly level, may reflect the relaxing effects of socializing. It may be good for students’ sleep if they regularly spend time with friends. In the daily study, students who slept longer not only reduced their socializing the next day but they increased time spent on schoolwork, suggesting that they chose academic work over friends on days they had the energy to do so. In the daily study, alcohol use seemed to interfere with sleep quality on a daily level, but no such association emerged here. Similarly, academic effort was unrelated to sleep in this study, but school demands covaried with sleep in complex ways in the daily diary study. Such differences between the monthly and daily results highlight the capacity for daily diary studies to capture temporal relations that may be missed in studies covering longer spans of time, but also suggest that some day-to-day relations may not emerge as reliable longer-term associations. Additionally, the time frame for the measurement of sleep
(past month) did not overlap completely with measures of student experience (past 14 days), which could have attenuated the magnitude of month-tomonth covariations.
Limitations include reliance on subjective rather than objective sleep indicators and collection of data at only one university. Possible directions for research include using actigraph recordings to gather objective sleep data, conducting studies at multiple sites and with non-students to enhance generalizability, measuring sleep and other experiences over the same time period, examining covariations of sleep with a wider range of student experiences (e.g., romantic relationships), following students into their later university years to determine how sleep affects academic performance, and evaluating the effectiveness of psychoeducational interventions (e.g., sleep hygiene instructions; Buboltz et al., 2006) for improving students’ sleep.
Sleep difficulties among university students can have profound academic, mental health, and behavioral consequences. This study bolsters a growing literature on student sleep that has implications for students and universities. For example, the link between sleep, stress, and negative affect may inform the practices and recommendations of on-campus counseling and health care. It may also encourage administrators to consider students’ sleep needs when scheduling classes and examinations
(Buboltz et al., 2006). The seasoned professor has learned to caution students not to sacrifice sleep for the sake of a few extra hours of study time before an exam, but universities must first begin to acknowl-

edge and address student sleep as a significant health concern for these cautions to carry weight in the classroom.
REFERENCES
Backhaus, J., Junghanns, K., Broocks, A., Riemann, D., &
Hohagen, F. (2002). Test-retest reliability and validity of the Pittsburgh Sleep Quality Index in primary insomnia.
Journal of Psychosomatic Research, 53, 737 – 740.
Beck, R., Taylor, C., & Robbins, M. (2003). Missing home:
Sociotropy and autonomy and their relationship to psychological distress and homesickness in college freshmen. Anxiety, Stress, and Coping: An International
Journal, 16, 155 – 166.
Brown, F. C., Soper, B., & Buboltz, W. C., Jr. (2001). Prevalence of delayed sleep phase syndrome in university students. College Student Journal, 35, 472 – 476.
Brown, N. R., Williams, R. L., Barker, E. T., & Galambos, N.
L. (2007). Estimating frequencies of emotions and actions: A web-based diary study. Applied Cognitive
Psychology, 21, 259 – 276.
Buboltz, W. C., Brown, F., & Soper, B. (2001). Sleep habits and patterns of college students: A preliminary study.
Journal of American College Health, 50, 131 – 135.
Buboltz, W. C., Loveland, J., Jenkins, S. M., Brown, F.,
Soper, B., & Hodges, J. (2006). College student sleep:
Relationship to health and academic performance. In
M. V. Landow (Ed.), College students: Mental health and coping strategies (pp. 1 – 39). New York, NY: Nova Science
Publishers Inc.
Buysse, D. J., Angst, J., Gamma, A., Ajdacic, V., Eich, D., &
Rossler, W. (2008). Prevalence, course, and comorbidity
¨
of insomnia and depression in young adults. Sleep, 31,
473 – 480.
Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R., &
Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index:
A new instrument for psychiatric practice and research.
Psychiatry Research, 28, 193 – 213.
Calhoun, P. S., Wiley, M., Dennis, M. F., Means, M. K.,
Edinger, J. D., & Beckham, J. C. (2007). Objective evidence of sleep disturbance in women with posttraumatic stress disorder. Journal of Traumatic Stress, 20,
1009 – 1018.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social
Behavior, 24, 385 – 396.
Fuligni, A. J., & Hardway, C. (2006). Daily variation in adolescents’ sleep, activities, and psychological well-being. Journal of Research on Adolescence, 16, 353 –
378.
Furr, S. R., Westefeld, J. S., McConnell, G. N., & Jenkins,
J. M. (2001). Suicide and depression among college students: A decade later. Professional Psychology: Research and Practice, 32, 97 – 100.
Galambos, N. L., Barker, E. T., & Krahn, H. J. (2006). Depression, self-esteem, and anger in emerging adulthood: Sevenyear trajectories. Developmental Psychology, 42, 350 – 365.

RISE AND FALL OF SLEEP
Galambos, N. L., Dalton, A. L., & Maggs, J. L. (2009). Losing sleep over it: Daily variation in sleep quantity and quality in Canadian students’ first semester of university. Journal of Research on Adolescence, 19, 741 – 761.
Maggs, J. L. (1997). Alcohol use and binge drinking as goaldirected action during the transition to post-secondary education. In J. E. Schulenberg, J. L. Maggs, &
K. Hurrelmann (Eds.), Health risks and developmental transitions during adolescence (pp. 345 – 371). New York,
NY: Cambridge University Press.
Pearlin, L. I., & Schooler, C. (1978). The structure of coping.
Journal of Health and Social Behavior, 19, 2 – 21.
Pilcher, J. J., & Huffcutt, A. I. (1996). Effects of sleep deprivation on performance: A meta-analysis. Sleep, 19, 318 – 326.
Pratt, M. W., Hunsberger, B., Pancer, S. M., Alisat, S.,
Bowers, C., Mackey, K., et al. (2000). Facilitating the transition to university: Evaluation of a social support discussion intervention program. Journal of College Student Development, 41, 427 – 441.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods. Thousand
Oaks, CA: Sage.
Schulenberg, J., & Maggs, J. L. (2002). A developmental perspective on alcohol use and heavy drinking during

349

adolescence and the transition to young adulthood.
Journal of Studies on Alcohol, Suppl 14, 54 – 70.
Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and event occurrence. New York,
NY: Oxford University Press.
Suen, L. K. P., Hon, K. L. E., & Tam, W. W. S. (2008). Association between sleep behavior and sleep-related factors among university students in Hong Kong.
Chronobiology International, 25, 760 – 775.
Thacher, P. V. (2008). University students and ‘‘the all nighter’’: Correlates and patterns of students’ engagement in a single night of total sleep deprivation. Behavioral Sleep Medicine, 6, 16 – 31.
Tsai, L.-L., & Li, S.-P. (2004). Sleep patterns in college students: Gender and grade differences. Journal of Psychosomatic Research, 56, 231 – 327.
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.
Wolfson, A. R., & Carskadon, M. A. (2003). Understanding adolescents’ sleep patterns and school performance: A critical appraisal. Sleep Medicine Reviews, 7,
491 – 506.

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy.
Users should refer to the original published version of the material.

References: Beck, R., Taylor, C., & Robbins, M. (2003). Missing home: Sociotropy and autonomy and their relationship to Brown, F. C., Soper, B., & Buboltz, W. C., Jr. (2001). Prevalence of delayed sleep phase syndrome in university students L. (2007). Estimating frequencies of emotions and actions: A web-based diary study Buboltz, W. C., Brown, F., & Soper, B. (2001). Sleep habits and patterns of college students: A preliminary study. Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress Fuligni, A. J., & Hardway, C. (2006). Daily variation in adolescents’ sleep, activities, and psychological Galambos, N. L., Barker, E. T., & Krahn, H. J. (2006). Depression, self-esteem, and anger in emerging adulthood: Sevenyear trajectories. Developmental Psychology, 42, 350 – 365. Maggs, J. L. (1997). Alcohol use and binge drinking as goaldirected action during the transition to post-secondary education Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Pilcher, J. J., & Huffcutt, A. I. (1996). Effects of sleep deprivation on performance: A meta-analysis. Sleep, 19, 318 – 326. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods Schulenberg, J., & Maggs, J. L. (2002). A developmental perspective on alcohol use and heavy drinking during Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and event occurrence Suen, L. K. P., Hon, K. L. E., & Tam, W. W. S. (2008). Association between sleep behavior and sleep-related factors among university students in Hong Kong. Thacher, P. V. (2008). University students and ‘‘the all nighter’’: Correlates and patterns of students’ engagement in a single night of total sleep deprivation Tsai, L.-L., & Li, S.-P. (2004). Sleep patterns in college students: Gender and grade differences. Journal of Psychosomatic Research, 56, 231 – 327. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales and Social Psychology, 54, 1063 – 1070. Wolfson, A. R., & Carskadon, M. A. (2003). Understanding adolescents’ sleep patterns and school performance: A critical appraisal. Sleep Medicine Reviews, 7, 491 – 506.