Content uploaded by Arn Eliasson
Author content
All content in this area was uploaded by Arn Eliasson
Content may be subject to copyright.
ORIGINAL ARTICLE
Early to bed, early to rise! Sleep habits and academic
performance in college students
Arne H. Eliasson &Christopher J. Lettieri &
Arn H. Eliasson
Received: 3 May 2009 / Revised: 13 June 2009 / Accepted: 24 June 2009 /Published online: 15 July 2009
#US Government 2009
Abstract
Purpose Prior studies have placed emphasis on the need for
adequate total sleep time for student performance. We
sought to investigate the relative importance of total sleep
time compared to the timing of sleep and wakefulness for
academic performance.
Methods We performed a questionnaire-based survey of
college students in October 2007. The questionnaire
gathered detailed information on sleep habits including
naps, reasons for missing sleep, academic performance,
study habits, time spent working outside of school, and
stimulant use.
Results Compared to those with the lowest academic
performance, students with the highest performance had
significantly earlier bedtimes (p=0.05) and wake times (p=
0.008). Napping tended to be more common among high
performers (p=0.07). Of importance, there were no signif-
icant differences in total sleep time with or without naps,
weekend sleep habits, study time, gender, race, reasons for
staying up at night, nor in use of caffeinated beverages,
over-the-counter stimulant pills, or use of prescription
stimulants.
Conclusions Timing of sleep and wakefulness correlated
more closely with academic performance than total sleep
time and other relevant factors. These findings have
important implications for programs intended to improve
academic performance by targeting sleep habits of students.
Keywords Academic performance .Total sleep time .
Bed time .Wake time .Circadian rhythm
Introduction
Habitual sleep patterns undergo substantial changes from
childhood to adolescence [1,2] and young adulthood [3].
These changes are characterized by progressive delay in the
sleep phase without a decrease in need for sleep (internal
factors) [4,5]. At the same time, schedules (external
factors) frequently require earlier wake times and lead to
shorter total sleep time (TST) [6,7]. Sleep debt accumu-
lated during the week often leads to prolonged sleep
periods or catch-up sleep on weekends causing severe
day-to-day irregularities of sleep patterns in adolescents and
young adults [8]. Insufficient sleep time, with associated
sleepiness, fatigue, and inattentiveness, has been identified
as a major cause of poor academic performance among high
school and college-aged students [9–11]. In fact, various
school districts have delayed their high school start times to
mitigate the effects of circadian sleep phase delay [12].
Even the US Congress has taken active interest in the issue
and has considered legislation known as “Z’stoA’s”to
encourage later school start times for adolescents [13].
The opinions expressed herein are those of the authors and should not
to be construed as official or as reflecting the policies of either the
Department of the Army or the Department of Defense.
A. H. Eliasson
Scholars Program, Montgomery College,
Rockville, MD 20850, USA
C. J. Lettieri :A. H. Eliasson
Pulmonary, Critical Care and Sleep Medicine,
Department of Medicine, Walter Reed Army Medical Center,
Washington, DC 20307-5001, USA
A. H. Eliasson (*)
Pulmonary Service, Walter Reed Army Medical Center,
12515 Davan Drive,
Silver Spring, MD 20904, USA
e-mail: aheliasson@aol.com
Sleep Breath (2010) 14:71–75
DOI 10.1007/s11325-009-0282-2
While it is clear that insufficient sleep is a major factor
governing mood, alertness, concentration, learning, and
ultimately performance in the academic environment [14],
the precise role of sleep quantity versus the impact of
circadian rhythms on performance remains ill-defined. We
sought to clarify the roles of TST and circadian rhythm on
academic performance among young adults by surveying a
population of college students.
Methods
Study sample and setting
In October 2007, we conducted a questionnaire-based
survey of students at the main campus of Montgomery
College, a community college in a northern suburb of
Washington, DC with enrollment of 60,000 full- and part-
time students. The college is ethnically and culturally
diverse with student body composition of Caucasian
(40%), African-American (28%), Asian (16%), and His-
panic (16%). Age distribution data show 41% of students
are 20 years old or younger, 35% age 21 to 29, and 24%
age 30 or older. The student body is 55% female and 45%
male.
Members of the teaching faculty distributed a two-page
questionnaire by hand to all students present in their classes
over a 2-day period. These faculty members informed their
students that participation was anonymous and voluntary.
Students were given the questionnaire to complete during
class time for expeditious collection. The classes incorpo-
rated a cross-section of curriculum offerings, such as:
biology, history, English, astronomy, and art. The number
of questionnaires distributed and collected was tallied by
the principle investigator. Approval of this research was
granted by the college’s Institutional Review Board.
Survey instrument
The principle investigator, three college professors, and two
physicians board-certified in sleep medicine constituted a
panel to develop the research questionnaire. Their mandate
was to limit the survey to two pages printed with 12-font
size. The survey instrument asked students about their sleep
habits (bed time, wake time, nap frequency and length,
week day versus weekend sleep) and how satisfied they
were with their sleep. The survey tool further queried about
daytime alertness versus grogginess and early-morning
versus late-night tendency. School performance was mea-
sured by self-reported grade point average (GPA), and
students were asked to report their usual amount of study
time (hours per day). Students also answered demographic
questions, recorded their employment time outside of
school (hours per week), and use of stimulants (caffeinated
drinks, energy drinks, caffeine pills, and prescription
stimulants).
The survey instrument was then administered to a pilot
group of students for feedback on question readability,
clarity, and question-and-answer format issues. Adjust-
ments in the survey instrument were implemented based
upon feedback from the pilot group leading to a two-page
tool with 29 questions that required approximately 5 min
for completion.
Data analysis
Continuous variables were analyzed using the Student’st
test. All tests were two-tailed, and pvalues of <0.05 were
assumedtorepresentstatistical significance. Data are
presented as mean ± standard deviation or as mean/median
with range as appropriate. All analyses were performed
using the Statistical Package for the Social Sciences 12.0
(SPSS Inc, Chicago, IL).
Results
Of 170 questionnaires, 157 (92%) were returnedby 76 women
(48%) and 81 men (52%). The mean age of the surveyed
students was 22.4±6.8 years (median 20 years, range 17 to
69 years). Racial composition was 42% Caucasian, 18%
Hispanic, 15% African-American, 13% Asian, and 12% self-
designated as other. Mean bedtime on nights before class was
12:16A.M.(range9:00P. M . to 4:00A.M.);meanwaketime
7:37A.M.(range3:30A.M.to12:30P. M.), with nightly sleep
time of 7 h 23 min. Mean total sleep time (including naps)
was7h32min.Onnightswithnoclass,bedtimewasan
hour later (1:12A.M.) and wake time 2.5 h later (9:58A.M.)
with nightly sleep time 8 h 43 min.
Among the whole group of students, a minority
(42%) expressed satisfaction with their sleep, and those
who were satisfied slept for an average of 47 min
longer each 24-h period. For all students, there was a
sizable majority admitting to feeling sleepy during the
day and more specifically feeling groggy during class.
Data for the highest quintile of performers by GPA were
compared with the data gathered on the lowest quintile of
performers (see Table 1). The racial composition of the
highest performing quintile of students and the lowest
quintile of performers were similar to the composition of
the whole group and were not different from each other.
While TST did not differ between groups, there was a
significant difference in the timing of sleep between high
and low academic performers. Specifically, those with
the highest GPA had earlier bed times (12:00A.M.versus
12:38A.M., p=0.05) and earlier wake times (7:13A.M.
72 Sleep Breath (2010) 14:71–75
versus 8:02A.M., p=0.008) than those with the lowest
GPA. There were no statistically significant differences in
TST with and without naps, weekend sleep habits, or
study time between the high performers and the low
performers. High performers were more likely to take naps
regularly than low performers (52% versus 29%, p=0.07).
The percentages of students who acknowledged a morning
tendency and those who were satisfied with their sleep
were not statistically different between high and low
performers (see Table 2). However, low-performing
students were more likely to factor sleep into their class
schedules (p=0.04). No other variables were associated
with academic performance. Specifically, there were no
differences in reasons provided for staying up at night, use
of caffeinated beverages, use of over-the-counter stimulant
pills, or use of prescription stimulants.
Discussion
Chief among the findings of this study is that timing of
sleep and wakefulness appears to be a more important
contributor to academic performance than total amount of
sleep. TST did not correlate with self-reported grade point
average, while earlier bed times and wake times did
correlate with higher grades.
Table 1 Demographic data, grades, sleep times, and study times
All students Lowest quintile (GPA <2.7) Highest quintile (GPA>3.5) pvalue
a
Age (years) 22.4 ±6.8 22.8± 7.6 23.2± 10.8 0.96
Gender (%male) 52 55 40 0.35
GPA out of 4.0 3.2± 0.5 2.6± 0.3 3.8± 0.2 <0.001
Bedtime before class
(range)
12:16A.M. (9:00 P.M.–4:00A.M.) 12:38 A.M. (10:30P.M.–3:00A.M.) 12:00A.M. (9:00 P.M.–4:00A.M.) 0.05
Wake time before class
(range)
7:37A.M. (3:30 A.M.–12:30 P.M.) 8:02A.M. (6:00A.M.–12:00P.M.) 7:13A.M. (5:00 A.M.–9:30A.M.) 0.008
Total sleep time before class 7 h 23 min 7 h 35 min 7 h 29 min 0.62
Bedtime with no class
(range)
1:12A.M. (9:00 P.M.–10A.M.) 1:29 A.M. (10:00P.M.–5:30A.M.) 12:51A.M. (9:00 P.M.–5:30A.M.) 0.28
Wake time with no class
(range)
9:58A.M.(6A.M.–4:30P.M.) 10:25A.M. (7:00A.M.–2:00 P.M.) 9:27A.M. (7:00 A.M.–2:00P.M.) 0.11
Total sleep time with no class 8 h 43 min 8 h 53 min 8 h 35 min 0.50
Naps (% yes) 42 29 52 0.07
Study time 2 h 48 min 2 h 23 min 2 h 47 min 0.40
a
pvalue derived from ttest comparison between values of lowest and highest quintiles of GPA
Table 2 Morning tendency, daytime symptoms, reasons for staying up at night, and stimulant use
All students Lowest quintile (GPA<2.7) Highest quintile (GPA>3.5) pvalue
a
Morning tendency (% yes) 25 29 40 0.35
Sleep affect class schedule? (% yes) 61 68 42 0.04
Satisfied with sleep (% yes) 42 35 45 0.50
Sleepy during the day (% yes) 62 65 53 0.90
Groggy during class (% yes) 75 81 78 0.97
Stay up for school (% yes) 53 52 55 0.80
Stay up for job (% yes) 30 32 35 0.79
Stay up to socialize (% yes) 54 52 45 0.62
Stay up for other reasons (% yes) 46 42 39 0.80
Job hours per week 19± 15 20± 15 16± 12 0.25
Caffeinated drinks per day 1.6 ± 1.6 1.5±1.6 1.3± 1.2 0.44
Stimulant pill use (% yes) 5 6 6 0.57
Use prescribed stimulant (% yes) 7 3 6 0.57
a
pvalue derived from ttest comparison between values of lowest and highest quintiles of GPA
Sleep Breath (2010) 14:71–75 73
The lack of correlation of TST with grades mirrors
findings of a prior report using similar methods [15].
However, the published literature contains many studies
that have demonstrated the importance of total sleep time
for full enhancement of intellectual functioning as well as
student safety behind the wheel [16]. The concept that sleep
deprivation erodes performance, academic and otherwise, is
not in debate. The importance of the current findings is that
circadian factors or the synchrony effect [17] also plays a
major role in academic performance.
This synchrony effect should be factored into any
interventional program designed for sleep improvement as
a means to enhancing school performance. It has been
previously demonstrated that moving school start times
does have the salutary effect of increasing total sleep time
[18] and improving attendance, but improvements as
measured with grades may not follow as was shown in
the Minneapolis Public Schools Start Time Study [12].
(http://education.umn.edu/carei/Reports). In fact, later
wake-up times may be associated with lower average
grades [19].
Compared to lower-performing students, higher-
performing students in our study had sleep onset almost
40 min earlier on average and awakened almost 50 min earlier
with no significant difference in TST. The importance of sleep
timing is particularly interesting since there does not appear to
be a congruent association of morning tendency with
academic performance in this cohort. These results suggest
that higher-performing students are able to find a way to shift
their sleep phase somewhat earlier than lower-performing
students despite the same degree of morning/evening prefer-
ence. It may be that the findings of the current study would be
more pronounced in sleep phase-delayed students attending
high school where class schedules are less flexible than they
are in college.
It is also interesting that higher-performing students
show a trend of napping more commonly than low
performers, though TST with naps is not statistically
different between groups. None of the other measured
variables serve to suggest a different interpretation. Specif-
ically, use of caffeinated beverages, over-the-counter
stimulants, and prescription stimulants were the same for
both groups. Equally interesting, gender, race, and total
study time did not predict GPA for this group of students.
The implications of the findings of the current study with
regard to the development of an educational program may
include the need for greater flexibility in the timing of
course offerings. Flexibility of class schedules, as opposed
to rigid schedule requirements, may allow students to find a
more natural sleep–wake cycle and improve academic
performance. An institution interested in creating a sleep
improvement program may benefit from the use of the
questionnaire created for this study (available as online
supplement). Additionally, it may be useful to include
questions on majors, year of study, and individually
important information such as body mass index and sleep
quality. Such information may greatly enhance the under-
standing of specific sleep issues facing the college students.
Many of the other findings of the current study parallel
those reported in populations of similar ages and circum-
stances. The sleep habits of our student population lead to a
sleep debt during the school week followed by attempts to
catch up on weekends. Furthermore, our students sleep less
than recommended amounts and on average do not go to
sleep before midnight despite schedules that require waking
before a full night’s sleep is accomplished. The prevalence
of stimulant pills and drug use is also similar to that
previously reported [20].
Limitations of this study include the use of self-reported
data for sleep times and school grades rather than objective
data measured by actigraphy and grades reported by the
college registrar. However, sleep survey results have been
shown to be as reliable as objective measures, at least in
studies involving adolescent-aged subjects [21]. Another
limitation is the use of a single question rather than a
validated questionnaire to determine subject tendency for
morningness and eveningness [22]. However, in order to
distribute a survey tool that would not require an inordinate
amount of class time, it was necessary to limit its length.
The importance of adequate amounts of sleep for peak
performance is not being questioned by the findings of this
study. Our data underscore the important contribution of
timing of sleep and wakefulness relative to the academic
schedule. To summarize, while it has been previously
shown that rest is an important contributor to performance,
circadian rhythm or an earlier habitual sleep period may be
more influential than TST with regard to academic
performance.
Acknowledgments The authors wish to thank Professor Sharon
Ward and Dr. Aram Hessami for their expertise in helping develop the
questionnaire. We also thank Dr. Shweta Sen for her administrative
guidance. We thank all three professors for their invaluable contribu-
tion with the distribution and collection of the questionnaire survey.
References
1. Karacan I, Anch M, Thornby JI, Okawa M, Williams RL (1975)
Longitudinal sleep patterns during pubertal growth: four-year
follow up. Pediatr Res 9:842–846
2. Taylor DJ, Jenni OG, Acebo C, Carskadon MA (2005) Sleep
tendency during extended wakefulness: insights into adolescent
sleep regulation and behavior. J Sleep Res 14:239–244
3. Ohayon MM, Roberts RE, Zulley J, Smirne S, Priest RG (2000)
Prevalence and patterns of problematic sleep among older
adolescents. J Am Acad Child Adolesc Psychiatry 39:1549–1556
4. Carskadon MA (1990) Patterns of sleep and sleepiness in
adolescents. Pediatrician 17:5–12
74 Sleep Breath (2010) 14:71–75
5. Carskadon MA, Vieira C, Acebo C (1993) Association between
puberty and delayed phase preference. Sleep 16:258–262
6. Mercer PW, Merritt SL, Cowell JM (1998) Differences in reported
sleep need among adolescents. J Adolesc Health 23:259–263
7. Fukuda K, Ishihara K (2001) Age-related changes of sleeping
pattern during adolescence. Psychiatry Clin Neurosci 55:231–232
8. Reid A, Maldonado CC, Baker FC (2002) Sleep behavior of
South African adolescents. Sleep 25:423–427
9. Wolfson AR, Carskadon MA (1998) Sleep schedules and daytime
functioning in adolescents. Child Dev 69:875–887
10. Beebe DW, Rose D, Amin R (2008) Effect of chronic sleep
restriction on adolescents’learning and brain activity in a
simulated classroom: a pilot study. Sleep 31:A77
11. Estrada A, Killgore WD, Rouse T, Balkin TF, Wildzunas RM
(2008) Total sleep time measured by actigraphy predicts academic
performance during military training. Sleep 31:A134
12. http://education.umn.edu/carei/Reports (last accessed 23 February
2009)
13. http://www.lofgren.house.gov/index.php?option=com_content
&task=view&id=337 (last accessed 13 June 2009)
14. Dinges DF, Pack F, Williams K, Gillen KA, Powell JW, Ott GE,
Aptowicz C, Pack AI (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
15. Eliasson A, Eliasson A, King J, Gould B, Eliasson A (2002)
Association of sleep and academic performance. Sleep Breath
6:45–48
16. Danner F, Phillips B (2008) Adolescent sleep, school start times,
and teen motor vehicle crashes. J Clin Sleep Med 4:533–535
17. Goldstein D, Hahn CS, Hasher L, Wiprzycka UJ, Zelazo PD
(2007) Time of day, intellectual performance, and behavioral
problems in morning versus evening type adolescents: is there a
synchrony effect? Pers Indiv Differ 42:431–440
18. Wolfson AR, Spaulding NL, Dandrow C, Baroni EM (2007)
Middle school start times: the importance of a good night’s sleep
for young adolescents. Behav Sleep Med 5:194–209
19. Trockel MT, Barnes MD, Egget DL (2000) Health-related
variables and academic performance among first-year college
students: implications for sleep and other behaviors. J Am Coll
Health 49:125–131
20. Teter CJ, McCabe SE, Cranford JA, Boyd CJ, Guthrie SK (2005)
Prevalence and motives for illicit use of prescription stimulants in
an undergraduate student sample. J Am Coll Health 53:253–262
21. Wolfson AR, Carskadon MA, Acebo C, Seifer R, Fallone G,
Labyak SE, Martin JL (2003) Evidence for the validity of a sleep
habits survey for adolescents. Sleep 26:213–216
22. Cavallera GM, Giudici S (2008) Morningness and eveningness
personality: a survey in literature from 1995 up till 2006. Pers
Indiv Differ 44:3–21
Sleep Breath (2010) 14:71–75 75