untitled

Sleep and Academic Performance in Undergraduates:

A Multi-measure, Multi-predictor Approach

Ana Allen Gomes,

1,2

José Tavares,

and Maria Helena P. de Azevedo

Department of Education, University of Aveiro, Aveiro, Portugal,

IBILI (FCT), Faculty of Medicine, University of Coimbra,

Coimbra, Portugal,

Institute of Medical Psychology, Faculty of Medicine, University of Coimbra, Coimbra, Portugal

The present study examined the associations of sleep patterns with multiple measures of academic achievement of

undergraduate university students and tested whether sleep variables emerged as significant predictors of

subsequent academic performance when other potential predictors, such as class attendance, time devoted to

study, and substance use are considered. A sample of 1654 (55% female) full-time undergraduates 17 to 25 yrs of

age responded to a self-response questionnaire on sleep, academics, lifestyle, and well-being that was administered

at the middle of the semester. In addition to self-reported measures of academic performance, a final grade for

each student was collected at the end of the semester. Univariate analyses found that sleep phase, morningness/

eveningness preference, sleep deprivation, sleep quality, and sleep irregularity were significantly associated with at

least two academic performance measures. Among 15 potential predictors, stepwise multiple regression analysis

identified 5 significant predictors of end-of-semester marks: previous academic achievement, class attendance,

sufficient sleep, night outings, and sleep quality (R

= 0.14 and adjusted R

= 0.14, F(5, 1234) = 40.99, p < .0001).

Associations between academic achievement and the remaining sleep variables as well as the academic, well-being,

and lifestyle variables lost significance in stepwise regression. Together with class attendance, night outings, and

previous academic achievement, self-reported sleep quality and self-reported frequency of sufficient sleep were

among the main predictors of academic performance, adding an independent and significant contribution,

regardless of academic variables and lifestyles of the students. (Author correspondence: [email protected])

Keywords: Academic performance, Adolescents, Chronotype, Multiple regression, Questionnaire, Sleep, Students

INTRODUCTION

Controlled studies that manipulated sleep in healthy

adults through a variety of methods, e.g., post-training

sleep and total, partial, or selective-stage sleep depri-

vation, have found that sleep is associated with a range

of cognitive activities, such as attention (Fafrowicz et al.,

2010; Lim & Dinges, 2010; Van Dongen et al., 2003a;

Wimmer et al., 1992), insight (Wagner et al., 2004 ), diver-

gent thinking (Horne, 1988; Wimmer et al., 1992),

decision-making (Harrison & Horne, 1999, 2000),

speech (Harriso n & Horne, 1997), and most notably

learning and memory (Diekelmann, 2009; Dotto, 1996;

Ficca & Salzarulo, 2004; Fogel et al., 2007; Li et al., 1991;

Maquet, 2001; Roehrs & Roth, 2000; Smith, 1995, 2001;

Stickgold & Walker, 2007; Stickgold et al., 2001; Walker

& Stickgold, 2004, 2006). Both total (Lim & Dinges,

2010) and partial (chronic) (Banks & Dinges, 2007; Van

Dongen et al., 2003) sleep deprivation may impair

daytime neurobehavioral functions in adults. However,

the exact mechanisms underlying the associations are

still unclear. A concise, up-to-date, discussion about the

main theoretical viewpoints on the effects of sleep

deprivation on cognitive functions may be found else-

where (Lim & Dinges, 2010).

A considerable amount of research has focused on the

role of sleep on memory. The role of sleep on memory is

not merely a passive one (interference reduction); rather,

research indicates that sleep actively facilitates memory

(Diekelmann et al., 2009; Ficca, 2010). Sleep appears to

be related to (i) distinct memory types such as working

memory (Kopasz et al., 2010; Lim & Dinges, 2010; Van

Dongen et al., 2003a) and long-term memory; (ii)

several kinds of materials, namely, memory for pro-

cedural/nondeclarative and declarative knowledge

(Dotto, 1996; Fogel et al., 2007; Smith, 1995, 2001); and

(iii) different memorization stages, such as encoding,

Part of this paper was presented as Open Communication at the 20th Congress of the European Sleep Research Society, Lisbon, 14–18

September 2010.

Address correspondence to Ana Allen Gomes, Departamento de Educação, Universidade de Aveiro, Campus Universitário de Santiago,

3810-193 Aveiro, Portugal. Tel.: +351 234 370 353; Fax: +351 234 370 640; E-mail: [email protected]

Submitted January 4, 2011, Returned for revision January 27, 2011, Accepted July 13, 2011

Chronobiology International, 28(9): 786– 801, (2011)

ISSN 0742-0528 print/1525-6073 online

DOI: 10.3109/07420528.2011.606518



consolidation, and reconsolidation (Walker & Stickgold,

2006). A full discussion of the role of sleep on memory,

and its underlying mechanisms, is beyond the scope of

the present study, but may be found in other works

(e.g., Diekelmann et al., 2009; Fica & Salzarulo, 2004;

Maquet, 2001; Stickgold & Walker, 2007; Walker &

Stickgold, 2004, 2006).

Although findings based on standardized cognitive

tasks of controlled studies cannot directly be generalized

to community samples in natural environments, the

above-mentioned cognitive activities seem intuiti vely

important for academic performance; therefore, it is

reasonable to suppose that sleeping behaviors and

patterns might also influence academic achievement in

real-life circumstances. In line with experimental

research, ecological studies have found significant

associations between sleep patterns and academic

achievement meas ures, such as grade point averages

(GPAs; for an overview see Curcio et al., 2006; Dewald

et al., 2010; Gomes et al., 2002; Wolfson & Carskadon,

2003). Although the focus of the present paper is on

undergraduates, it is worth mentioning that the relation-

ships between sleep parameters and school performance

have been more regularly investigated in children and

adolescents of several age and educational levels (Bruni

et al., 1995; Buckhalt et al., 2009; Dewald et al., 2010;

Giannotti & Cortesi, 2002; Hofman & Steenhof, 1997;

Meijer & Wittenboer, 2004; Pagel & Kwiatkowski, 2010;

Pagel et al., 2007; Ravid et al., 2009; Roberts et al., 2001;

Wolfson & Carskadon, 1998). In addition, a growing

number of experimental studies on younger children

and adolescents show, for instance, that sleep facilitates

memory (Kopasz et al., 2010), and that sleep restriction

or extension in school-aged children by only 1 h during

consecutive nights leads to differential impact on neuro-

behavioral measures (Sadeh et al., 2003). However, it

cannot be assumed that the effects of sleep on cognition

or academic performance are the same in all ages or at all

stages of human development (Dewald et al., 2010). For

example, academic outcome measures seem to be

differentially influenced by sleep, depending on student

educational level and age (Pagel et al., 2010), and

recent reviews (Diekelmann et al., 2009; Kopasz et al.,

2010) report that procedural memory consolidation in

children may not benefit from sleep to the same extent

as it does in adults.

Our focus is sleep and academic performance of uni-

versity undergraduate students. University students are

required to perform at demanding levels. In addition,

sleep patterns are likely to change from high school to

university due to alterations in zeitgebers, such as class

schedules and lifestyle preferences (Urner et al., 2009).

Specifically, in samples of university students observed

in their natural environment, poorer academic results

have been consistently associated with shorter sleep dur-

ation (Borisenkov et al., 2010; Jean-Louis et al., 1996;

Kelly et al., 2001; Medeiros et al., 2001; Trockel et al.,

2000), with later sleep-wake schedules (Elliason et al.,

2010; Johns et al., 1976; Medeiros et al., 2001, 1996;

Smith et al., 1989; Trockel et al., 2000), and/or with

related variables, such as delayed sleep phase (Lack,

1986) and eveningness orientation (Bes¸oluk et al., 2011;

Borisenkov et al., 2010; Medeiros et al., 2001; Randler &

French, 2006; Smith et al., 1989). Eveningness preference

has also been found to be associated with other variables

apparently related to academic achievement, such as

procrastination (Digdon & Howell, 2008; Hess et al.,

2001). In a study on personality, conscientiousness was

associated with earlier class schedules, which in turn

were associated with academic performance (Gray &

Watson, 2002). Lower academic grades in college were

also found to be associated with other sleep variables,

such as irregular sleep-wake cycle (Medeiros et al.,

2001), poor sleep quality (Gilbert & Weaver, 2010;

Howel et al., 2004; Johns et al., 1976), complaints of

onset and maintenance insomnia (Pagel & Kwiatkowski,

2010), excessive daytime sleepiness (Rodrigues et al.,

2002), and frequent snoring (Ficker et al., 1999).

Summarizing the vast literature on sleep and chrono-

biology, we may assume that four fundamental sleep

patterns are expected to be associated with academic

achievement: sleep quantity, sleep quality, sleep regu-

larity, and sleep phase schedules. Specifically, sleep

restriction, poor sleep quality, and irregular and late

sleep schedules are expected to be associated with

poorer school performance.

As to sleep amount, given the vast research on sleep

deprivation (both partial and total) and on hypothetical

sleep functions (e.g., restoration theory), especially in

relation to the impact of these sleep functions on cogni-

tive/neuropsychological functioning as already reviewed,

it is reasonable to expect students with greater sleep debt

to demonstrate worse academic performance. In fact,

both rapid eye movement (REM) and non-REM sleep

stages appear to play a role in memory and learning con-

solidation. Therefore, it is also expected that poor sleep

quality, which may manifest itself through difficulties

with sleep onset, and/or light sleep, and/or fragmented

sleep, might also have an impact on academic perform-

ance. As to sleep irregularity, shiftwork and jetlag research

shows that abrupt changes in the sleep-wake schedule

lead to internal dissociation of circadian rhythms, which

may result in a variety of undesirable effects, including

performance decrements (AASM, 2005). As jetlag symp-

toms may arise when three or more time zones are

rapidly traversed, it is reasonable to suppose that univer-

sity students showing comparable irregularities in their

sleep-wake schedule will suffer undesirable conse-

quences, such as higher fatigue, mood deterioration,

reduced performance (Taub & Berger, 1973, 1976), and

excessive daytime somnolence (Manber et al., 1996).

Finally, studies on student samples have consistently

reported poorer school performance to be associated

with a later sleep-wake schedule and/or chronotype pre-

ference towards eveningness. In this case, it is worth men-

tioning that morningness-eveningness is a continuum of

Sleep and Academic Performance in Undergraduates 

normal interindividual differences. That is, along this

continuum, sleep should be normal in quantity (Roenne-

berg et al., 2004) and quality, providing the individual is

able to adapt to her/his preferred schedule (AASM,

2005). Therefore, in contrast to sleep restriction, sleep

irregularity, or poor sleep quality, it seems that later

sleep-wake schedule and eveningness are not per se pro-

blematic in the sense that there is no alteration of sleep

duration or architecture. Thus, the hypothesized associ-

ations between later sleep-wake schedule and academic

performance are not direct, but are most probably

mediated by oth er variables, such as sleep restriction

and sleep irregularity, both being higher in evening

types (Giannotti et al., 2002; Gomes et al., 2008; Taillard

et al., 1999), and/or lower class attendance, which may

occur as a consequence of conflict between late sleep-

wake schedule/eveningness tendencies and externally

imposed morning class schedules.

Despite the numerous publications on sleep and aca-

demic performance among university students, to date

very few studies have examined the relative impact of

sleep variables on academic results in real-life circum-

stances and when other potential predictors are con-

sidered, such as well-being, lifestyle, and academic

variables. Jean-Louis et al. (1996) considered several psy-

chosocial factors, such as personal, medical, social, sleep

habits, academic, mood, and substance abuse, that could

be related to the academic performance of college stu-

dents. Multiple regression analysis identified six signifi-

cant predictors, three of which were sleep variables

(weekend sleep amount, sleep latency, and falling

asleep at school). Trockel et al. (2000) analyzed the

associations of first-year undergraduate academic per-

formance with health-related variables relative to exer-

cise, nutrition, sleep habits, mood states, perceived

stress, time management, social support, religious or

spiritual habits, extra number of hours worked per

week, sex, and age. Multiple regression analyses selected

five significant predictors, two of which were sleep vari-

ables (weekday and weekend wake-up times), that

accounted for the highest proportion of explained var-

iance in GPA. In a study focused on links between

alcohol use, sleep, and academic performance in

college students, Singleton and Wolfson (2009) found

Scholastic Aptitude Test scores were the strongest predic-

tor of GPA, with the other significant predictors being sex,

alcohol consumption, sleep duration, and daytime slee-

piness. By consider ing the multiple variables that might

explain academic performance, these types of studies

provide valuable data and help to better assess the rela-

tive effect of the various aspects of sleep.

In a relatively recent literature review, Curcio et al.

(2006) state that the first step in the research agenda on

sleep and academic performance should be to find

reliable measures of academic performance or, alterna-

tively, adopt the recommendation of Wolfson and

Carskadon’s (2003) of using a multimeasure approach

(e.g., grades, tests, teacher reports). In addition,

Wolfson and Carskadon stress the need to gather longi-

tudinal data (see also Dewald et al., 2010) and emphasize

that future studies should assess a variety of other vari-

ables, besides sleep, that influen ce academic per form-

ance. In spite of these recommendations, to date very

few published studies have adopte d such a multimea-

sure, multipredictor approach. Moreover, research

aimed at studying the associations between sleep and

academic performance has rarely measured neuroticism,

the one exception being the study of Gray and Watson

(2002), apparently overlooking the literature that has em-

phasized neuroticism to be one of the most important

individual predictors of intolerance to shiftwork, as neu-

rotic individuals are more prone to experience undesir-

able consequences following abrupt changes in their

sleep-wake schedule (e.g., Costa et al., 2001; Härma,

1993; Saksvik et al., 2011). In addition, there appears to

be interindividual variability in the susceptibility to

sleep restriction (Banks & Dinges, 2007; Van Dongen

et al., 2003b) as well as to desynchronization of circadian

rhythms (Reinberg et al., 1989), and neuroticism is likely

related to this greater vulnerability, at least with respect to

sleep debt (Blagrove & Akehurst, 2001; Taylor & McFatter,

2003). For these reasons, it seemed important to consider

neuroticism in the present study.

The present study had two major aims: (i) to analyze

associations between sleep patterns and multiple

measures of academic performance of university stu-

dents (self-reported retention, previous GPA, subjective

impact of sleep patterns on academic results, and end-

of-semester marks as indicated by university records);

and (ii) to examine longitudinally whether sleep vari-

ables emerge as significant predictors of subsequent aca-

demic performance when other potential predictors,

such as class attendance, time devoted to study, sub-

stance use, and neuroticism, are considered. Included

in the second aim was determination of whether neuroti-

cism plays a moderating role in the associations between

sleep patterns and subsequent end-of-semester marks.

Indeed, it may be that inadequate sleep, such as sleep

curtailment, has a greater detrimental effect on daytime

functioning in neurotic subjects than in stable subjects.

In this study, we focus on four fundamental sleep vari-

ables: sleep amount, sleep quality, sleep regularity, and

sleep phase schedule. Specifically, we hypothesize that

sleep restriction, poor sleep quality, and irregular and

late sleep schedules are associated with poorer

academic performance.

METHODS

Sample

Participants were 1654 full-time students, 55% female

and 45% male, aged 17 to 25 yrs (M = 19.98, SD = 1.65),

from a public Portuguese University, located in a city at

the littoral, center-north region of Portugal. They were

distributed across the 1st (31.3%), 2nd (39.5%), and 3rd

(29.2%) yrs of university study of 18 undergraduate

 A. A. Gomes et al.

Chronobiology International

degree programs, representative of 50% of the existing

undergraduate degree programs of the university

grouped into five academic fields: engineering (40%),

sciences (30%), pre/primary-school education (12%),

management (10%), and languages (9%). Inclusion cri-

teria were ≤25 yrs of age, full-time student status, non-

working, nonpregnant, and nonparent (no children).

Exclusion criteria were age >25 yrs of age, part-time

student status, working, elite athlete, significant univer-

sity extracurricular activities, having children, and being

pregnant. Based on the students’ residency status on

school days versus weekends/holidays, three groups

were formed and labeled as “moved students” (66%),

those presumably studying outside their home and

living in the university city during the week; “nonresi-

dents” (23.7%), students that presumably travel daily

from their home town to the university to attend

classes; and “residents” (10.3%), students whose family

home is presumably located in the university city.

Instruments and Measures

Sleep-Wake Questionnaire

A sleep-wake questionnaire covering demographic,

sleep, academic, lifestyle, and well-being variables was

developed for a large research project on sleep, well-

being, and academic success of university students that

was to be completed during the school semester. Based

on existing sleep-wake questionnaires, and lacking a

specific Portuguese instrument to access sleep-wake pat-

terns in undergraduates, the questionnaire was princi-

pally constructed by the first and third authors, both of

whom had at least 5 yrs of clinical practice at a sleep

clinic at the University Hospital of Coimbra and research

experience in the adaptation, development, and vali-

dation of psychological and psychiatric assessment

tools. The second author contributed with his experience

in supervising research projects about the diagnosis and

intervention strategies for the promotion of academic

success at the university level. The questionnaire was

also built upon the experience of all authors having

served as university teachers. The first version was

tested with 103 undergraduates using “think aloud” pro-

cedures. After this pilot study, several improvements

were made. The resulting version was then examined

by a group of five teachers. Again, minor improvements

were introduced based on their feedback. After these

steps, the authors agreed on a definitive version of the

questionnaire. The entire questionnaire is shown in

Appendix, and items used in the present work are ident-

ified with an asterisk. The psychometric properties are

further addressed below.

Composite Morningness Questionnaire

The Composite Morningness Questionnaire (CMQ;

Smith et al., 1989), Portuguese version (Silva et al.,

1995), was used to measure chronotype. In our sample,

the internal consistency of the CMQ, assessed through

Cronbach α, was 0.81.

Eysenck Personality Inventory

The Eysenck Personality Inventory (EPI), the 12-item

version, from the Standard Shiftwork Index (SSI)

(Barton et al., 1995) (Portuguese version: Silva et al.,

1995), was used to measure neuroticism and extrover-

sion. Two main reasons led us to prefer this tool

instead of other measures of these constructs. First, the

small number of items is advantageous when researchers

need to collect large amounts of data in an already big

booklet (as in our study). Second, this version was

chosen to integrate the SSI, a battery of tests utilized by

leading researchers to investigate shiftwork through the

adoption of standardized measures (Barton et al.,

1995). A two-factor structure was found in our study

(principal components analysis with varimax rotation)

in accordance with the expected, with the exception

of two item s that did not load in any dimension

and were, therefore, excluded from further analyses.

Five items loaded on the extraversion factor (22.21% of

the explained variance; Cronbach α = 0.68), and an

additional five items loaded on the neuroticism factor

(22.07% of the explained variance; Cronbach α = 0.66).

Variables derived from the self-response question-

naires that are relevant for the analyses of the present

study covered several domains. (Except when otherwise

specified, the described variables are taken from the

sleep-wake que stionnaire; further details about the

respective items may be found in Appendix.)

Demographics variables included sex, age, residential

status, curricular year, academic field (cf. initial ques-

tions in Appendix).

Academic antecedents variables included past aca-

demic achievement (as measured by self-reported pre-

vious GPA, rated on a 6-point scale), vocational

preferences (mis)match (1st, 2nd, 3rd, or other), and

academic failure in most courses of the previous curri-

cular year. As regards prior GPA, in Portugal marks are

expressed on a 0- to 20-point scale (at the university

level) or, similarly, on a 0- to 200-point scale (admis-

sion to university classification as the result of a

weighted mean between high-school GPA and admis-

sion tests). In the sleep-wake questionnaire, the partici-

pant was asked to report his/her prior GPA on a 6-point

scale such that ≤10 (coded as 1), 10–11 (coded as 2),

12–13 (coded as 3), 14–15 (coded as 4), 16–17 (coded

as 5), ≥18 (coded as 6) for university classification or

as similar options formulated in terms of the to 0- to

200-point scale (rather than 0 to 20) for admission to

the university (cf. item 35 in Appendix). As to voca-

tional preference match, student admission to majors

in public universities in Portugal occurs once a year

throughout the country. Each student must complete

a form indicating his/her preference for an under-

graduate degree program and university. Candidates/

Sleep and Academic Performance in Undergraduates 

applicants to each undergraduate degree program are

sorted by the Portuguese Ministry of Science and

Higher Education according to their grades. Depend-

ing on the applicant’s position on each list and on

the limit of admissions defined for each degree

program, students may or may not be admitted to

their first preference, and, alternatively, they may be

admitted to their second, third, or other choice. In

the present study, we have assumed that the student’s

choice (assessed in the “Demographics/Academics”

section of the questionnaire) reflects his/her vocational

preference. As to the previous year academic failure in

most courses (7th question on the “Demographics/

Academics” part of the questionnaire), in many Portu-

guese faculties and universities, students may pass or

fail a curricular year (or grade) at the university level,

just as they can at the high-school level. Undergraduate

programs are structured in academic years (two seme-

sters) in such a way that each course is matched to a

given curricular year. To obtain approval (pass) in

any single course, the student must attain a final

mark of ≥10 points on a 20-point scale (<10 points,

the student fails that course). Each individual course

corresponds to a certain number of credits. Universities

and faculties establish a minimum number of credits/

curricular year that students must obtain by passing

courses in order to proceed (pass) to the subsequent

curricular year. When a student fails more than half

the courses in a given academic year, he/she does

not obtain enough credits. Consequently, he/she

remains, technically, in the same curricular year and

must repeat the failed courses. Herein, we term this

situation “previous year failure in most courses.”

Current academic engagement variables include class

attendance (rated on a 5-point scale, cf. item 34 in

Appendix), and study time (h/wk, cf. item 32 in

Appendix).

Lifestyle and substance usage variables include (cf.

items 26–31 and 33 in Appendix) exercise (h/wk),

other extracurricular activities (h/wk), night outings

(frequency), cigarette use (weighted mean = [units/

week day × 5 + units/weekend day × 2]/7 days),

alcohol consumption (weighted mean = [units/week

day × 5 + units/weekend day × 2]/7 days), coffee con-

sumption (weighted mean = [units/week day × 5 +

units/ weekend day × 2]/7 days), and consumption of

other substances (frequency).

Daytime subjective well-being variables include vigor,

mood/anxiety complaints, cognitive functioning, and

daytime somnolence (Manber et al., 1996). The first

three indices resulted from a factor analysis of 14

items asking how the student felt during the day, with

each item rated from 0 to 4 or 4 to 0, as appropriate

(cf. items 19a to 19n in Appendix). The selected

method was a principal component analysis with

varimax rotation for components with eigenvalues

≥1.0. A three-factor solution was found, explaining

55.99% of the total variance. Factor 1 items (active,

energetic, efficient, alert, happy, relaxed) corresponded

to vigor and accounted for 23.01% of the variance.

Factor 2 items (tired, irritable, depressed, nervous) cor-

responded to mood/anxiety complaints and accounted

for 17.63% of the variance. Factor 3 items (productive,

attentive, motivated, difficulty concentrating) corre-

sponded to cognitive functioning and accounted for

15.35% of the variance. The Cronbach α values were

0.77, 0.74, and 0.73, respectively, indicating good

internal consistency of each factor. The daytime som-

nolence index consisted of 5 items adapted from

Manber et al. (1996), plus 1 item about somnolence

during class (items 18a to 18e and 18f in Appendix).

Cronbach α was 0.84.

Neuroticism was measured by our neuroticism factor of

the EPI-12 item version.

Sleep quantity variables include perceived (in)sufficient

sleep using a frequency scale (cf. item 14 in Appendix).

Sleep quality variables include 7 items, each rated from

0 to 4 or 4 to 0, as appropriate, covering sleep-onset,

early and night awakenings, perceived sleep depth,

and quality of sleep (cf. items 4–8 plus 15a and 15b

in Appendix). A sleep quality index was obtained

through summation of these items. Higher scores

equate to poorer sleep quality. Th e internal consistency

was good as indicated by the Cronbach α = 0.73.

Sleep phase/chronotype variables include sleep phase

during weeknights, sleep phase during weekend

nights, and morningness-eveningness as expressed

by the CMQ total score. Each sleep phase variable

was determined as a mid-sleep point expressed in

hours and minutes, according to the specified formulas

[weeknights mid-sleep = bedtime on weeknights +

(time in bed on weeknights/2) and weekend nights

mid-sleep = bedtime on weekends + (time in bed on

weekends/2)], where time in bed was the time interval

between bedtime and rise time (either on weeknights

or on weekends, as appropriate).

Sleep irregularity variables include bedtime irregularity

during the school week, rise-time irregularity during

the school week (cf. items 9 and 13 in Appendix), and

week/weekend sleep-phase irregularity (difference

between week and weekend night sleep phases,

expressed in hours and minutes).

Academic Performance Measures

a. Self-report measures: Three academic achievement

items were included in the sleep-wake questionnaire.

Previous GPA and academic failure of most courses

during the previous curricular year were previously

described. In addition, we also considered it important

to assess the subjective impact of sleep on academic

performance as perceived by each student through

the following question: “Do you feel your sleep

patterns have been negatively influencing your aca-

demic performance at the university?” (cf. item 36 in

Appendix).

 A. A. Gomes et al.

Chronobiology International

b. Objective prospective measure: This measure consisted

of the final mark obtained by each student at the end of

the semester as shown in the university records.

Specifically, for each undergraduate degree program,

we selected one course per year (the most representa-

tive and relevant). For instance, for all participants in

the 1st year of the mathematics degree program (n =

43), we collected the final marks received in the Math-

ematical Analysis II course; for all participants study-

ing in the 3rd year of the biology degree program

(n = 72), the end-of-semester marks obtained in the

Genetics course were collected. The end-of-semester

classification of the student in a given course, on a

0- to 20-point scale, represents a weighted mean

score of a certain number of tasks, i.e., assignments

and written examinations. The assessment tasks are

defined by each professor in agreement with university

rules. To enable combination and comparison among

marks between different courses, within each given

course raw classifications were transformed into stan-

dardized z values.

Procedures

The research project was approved by the department

and university scientific councils, which were the local

sanctioning boards of the university, and conformed to

international ethical standards as described in Portaluppi

et al. (2010) for biological rhythm research.

A teacher from each university year and selected

degree program was approached, and the voluntary

nature and the general format of the research were

explained. With the approval of the teachers and the

consent of the students, the questionnaires were com-

pleted at the end of the class sessions. It was emphasized

that participation was voluntary, and confidentiality was

assured. The aims of the study were briefly explained in

the beginning of the questionnaire and were also orally

explained to the students by the principal researcher,

who was present in all sessions.

To assess the typical sleep-wake patterns of the stu-

dents when they must attend classes, the survey was con-

ducted in the middle of the semesters, at least 1 mo after

the beginning of the classes. Data collection was carefully

planned so events that could potentially influence sleep-

wake patterns, e.g., holidays and student festivities, were

excluded. All questionnaires were collected after 12:00 h

to avert underrepresentation of evening-type students in

the sample. Participation rate, determined by the differ-

ence between the number of questionnaires distributed

(2018) and number of questionnaires returned (1819),

was 90.1%. From the questionnaires collected, a total of

165 were excluded due to missing answers on key ques-

tions, i.e., sleep-wake schedules and sleep durations

(33 participants); atypical cases or circumstances, e.g.,

pregnancy (4 participants); age >25 years (61 partici-

pants); having children (5 participants); non-full-time

student, i.e., students with part- or full-time jobs, elite

athlete students, and those involved in University

Student Union activities (62 participants). A final classifi-

cation for each participant was obtained through univer-

sity records at the end of the semester.

Data Analyses

First, associations between each sleep pattern and aca-

demic performance measures were examined. t tests for

independent samples were used to compare mean

values on sleep patterns between students who passed

versus students who failed the previous curricular year.

Mean values for each variable were compared between

sleep groups for the remaining academic achievement

measures, previous GPA, subjective detrimental effect

of sleep on academic performance, and end-of-semester

marks. First, following Neale and Libert’s (1986) rec-

ommendations, for each sleep variable three to four (as

appropriate) nonextreme groups of similar size were

formed based on quartiles or frequencies. Then, aca-

demic performance mean values were determined for

each sleep group and compared by analysis of variance

(ANOVA). End-of-semester raw classifications in all ana-

lyses were converted to standardized z scores to enable

combination and comparability among marks from

different courses.

Second, considering the entire set of potential predic-

tors of academic performance, the main predictors for

end-of-semester marks were identified through multiple

regression analysis using the stepwise method to select

the most relevant variables:

1. The normality assumption was checked for numeric

variables. For variables not showing normal distri-

bution, data were transformed to approximate a Gaus-

sian curve (log

transformations were used to correct

for skewed distributions).

2. A correlation matrix was examined to select potential

predictors and avoid multicollinearity; whenever

nominal variables were involved, we used other mag-

nitude of association measures instead of correlation

coefficients, e.g., η.

Variables showing nonsignificant associations with z

scores were excluded.

Among variables significantly associated with z

scores, redundant variables were also removed; for

each set of interrelated variables, the rule was to

retain the variable showing the highest correlation

coefficient with z scores. In addition, a collinearity

diagnosis of the model was made, and indepen-

dence was assumed if the following criteria were

met: variance inflation factor [VIF] <2, tolerance

values distant from 0, Durbin Watson statistic ∼2,

condition index <15.

3. The stepwise regression analysis was conducted, en-

tering as potential predictors only those variables

showing significant associations with marks in uni-

variate analyses and with minimal redundancy

Sleep and Academic Performance in Undergraduates 

among each other. The criterion variable was the z

scores. The rule to enter or remove a variable was

p ≤ .05 and p ≥ .10 (SPSS default option). As to

missing values, listwise deletion was used (Afifi

et al., 2004). Eight outliers were detected and excluded

from the database for the regression analysis. An

additional multiple regression analysis was conducted

to test for neuroticism moderator effects. To represent

the relevant interactions, variables were first centered

and then multiplied together.

A total of 1240 participants were included in the

regression analysis, i.e., 75% of the initial sample (n =

1654). These 1240 were comparable to the total sample

with respect to sex, age, curricular year, academic

domain, and residency status. A total of 406 subjects

were excluded from the analysis due to missing data in

any of the potential predictors or in the criterion variable.

These 406 students were similar to the remaining 1240 for

academic field and residency situation, but were older

(M = 20.47, SD = 1.77 vs. M = 19.82, SD = 1.59; t test,

p < .05) and men were overrepresented (49.8% in the

excluded group vs. 43.2% in the sample group; χ

test,

p < .05).

RESULTS

Sleep and Multi-measures of Academic Achievement

Sleep and Previous Academic Failure (Failed Most Courses in

the Last Year)

Compared to those who passed (88.9%, n = 1457), stu-

dents who failed most of their courses in the preceding

curricular year (11.1%, n = 182) displayed current later

phases of the sleep-wake cycle, both on weeknights

1637

= 7.12, p < .0001) and on weekends (t

1637

= 3.45,

p < .0001), higher eveningness orientation (i.e., lower

morningness scores on the CMQ, t

1637

= 3.92, p < .0001),

and greater rise-time variation during the week (t

1607

3.01, p < .01). This pattern of results was also evident

when examining per sleep group the percentage of

students who had failed most courses (Table 1).

Prior GPA by Sleep Group

From eveningness to morningness chronotype tendency

groups (F = 5.511, p < .001) and across groups of students

showing progressively lower rise-time oscillations during

the week (F = 3.185, p < .05), there was increased im-

provement in past academic achievement grades. Pre-

vious GPAs were also found to be higher in groups with

earlier sleep-wake phases during the school week (F =

14.760, p < .0001), and in groups reporting better sleep

quality (F = 2.710, p < .05), as shown in Table 1.

Perceived Impact of Sleep on Academic Performance

Across groups showing progressively later sleep-wake

phases on weeknights (F = 17.115, p < .0001) and on

weekends (F = 4.247, p < .01), lower morningness scores

(F = 23.284, p < .0001), lower frequency of enough sleep

(F = 91.141, p < .0001), poorer sleep quality (F = 19.699,

p < .0001), and greater bedtime (F = 11.936, p < .0001)

and rise-time (F = 10.853, p < .0001) oscillations during

the week, there was an increase in mean values for the

perception that sleep patterns have had a negative

impact on university academic performance (Table 1).

End-of-Semester Marks by Sleep Group

Students of sleep groups who achieved higher marks

(mean z scores) at the end of the semester (Table 1)

were those who reported earlier sleep phases during the

week (F = 5.335, p < .001) and on weekends (F = 4.649,

p < .01), higher morningness scores (F = 3.486, p < .05),

more stable bedtime schedules during the week (F =

3.240, p < 0.05), better sleep quality (F = 3.722, p < .05),

and higher frequency of enough sleep (F = 3.689, p < .05).

Main Predictors of End-of-Semester Marks: Stepwise

Multiple Regression

After identifying through ANOVA those sleep variables

significantly associated with end-of-semester marks, we

then sought to determine whether or not any sleep

pattern would be selected as a main predictor of academic

performance when considered concurrently with other

potential predictors of academic performance in a step-

wise multiple regression analysis. From a total of 30 vari-

ables initially considered to possibly be associated with

university marks (covering demographic, neuroticism,

lifestyle, well-being, academic, and sleep domains), a pre-

liminary univariate analysis found 19 significantly related

to z scores ( p < .05). Nonsignificant associations were

found for 11 variables (sex, residential status, curricular

year, academic field, passing or failing most courses the

previous year, exercise, other extracurricular activities,

mood/anxiety complaints, neuroticism, and, as already

reported, rise-time irregularity during the school week

and week/weekend sleep-phase irregularity). From

these 19 variables significantly associated with z scores,

4 (vigor, daytime somnolence, sleep phase on week-

nights, QCM total score) were removed from the analysis

to avert multicollinearity with similar variables, thus as-

suring independence among potential predictors. The

intent was to identify, within each group of interrelated

variables, the variable with the strongest association

with z scores. Thus, 15 potential predictors were entered

in the stepwise regression analysis. Descriptive statistics

for the potential predictors and for z scores (criterion vari-

able) are shown in Table 2. From the 15 variab les included

in the model, stepwise regression yielded 5 that were sig-

nificant predictors of marks, explaining 14% of the var-

iance of the z scores (R

= 0.14, adjusted R

= 0.14; F(5,

1234) = 40.99, p < .0001; Table 3). The most important

predictor of z scores was previous academic achievement,

followed by class attendance. The third predictor was the

frequency of enough sleep. Night outings and sleep

quality were the last of the five significant predictors.

Moreover, the selected predictors showed associations

 A. A. Gomes et al.

Chronobiology International

with z scores in the expected directions. Thus, higher

(lower) previous GPA, higher (lower) class attendance,

higher (lower) frequency of enough sleep, lower

(higher) frequency of night outings, and better (worse)

sleep quality during the semester were associated with

an increase (decrease) in end-of-semester marks as

expressed by z scores.

Associations of the remaining sleep, academic, well-

being, and lifestyle variables e.g., (time devoted to

study, alcohol consumption, cognitive functioning) with

academic achievement lost significance in the stepwise

regression. Of the four sleep variables entered in the

analysis, two—sleep quality and enough sleep—re-

mained significantly associated with end-of-semester

grades in the presence of other significant predictors,

thus adding an independent contribution to grades.

The other two sleep variables introduced in the analysis,

sleep phase and regularity of sleep schedules, lost signifi-

cance in a stepwise regression after controlling for the

influence of previous academic results, class attendance,

night outings, sleep amount, and sleep quality.

Finally, to ascertain whether neuroticism interacts

with the two selected sleep variables in predicting

end-of-semester marks, another multiple regression

TABLE 1. Academic achievement measures across sleep groups

Failed most

courses (%)

GPA

(M)

Negative impact of sleep over

performance

(M)

End-of-semester

marks

(M)

Sleep Quality Index Very Good 10.4 3.43 1.55 0.13

Good 9.8 3.28 1.74 −0.05

Poor 13.2 3.30 1.70 −0.04

Very Poor 11.0 3.29 2.10 −0.08

NS <0.05 <0.0001 <0.05

Enough sleep

(frequency)

Never + Rarely 11.2 3.27 2.28 −0.05

1–2 nights/wk 10.5 3.34 2.02 −0.09

3–4 nights/wk 12.4 3.27 1.71 −0.02

Almost all nights/always 10.3 3.42 1.15 0.14

NS NS <0.0001 <0.05

CMQ Morningness tendency 7.6 3.49 1.50 0.14

Intermediate—toward

morning

9.4 3.34 1.62 0.00

Intermediate—toward

evening

13.1 3.27 1.83 −0.06

Eveningness tendency 14.3 3.24 2.07 −0.08

<0.0001 <0.001 <0.0001 <0.05

Sleep phase on week

nights

<03:56 h 5.3 3.51 1.61 0.11

03:56 to 04:25 h 9.2 3.46 1.62 0.09

04:26 to 05:00 h 9.3 3.16 1.74 −0.13

>05:00 h 21.7 3.20 2.09 −0.09

<0.0001 <0.0001 <0.0001 <0.001

Sleep phase on weekend

nights

<05:15 h 7.9 3.44 1.68 0.13

05:15 to 06:00 h 9.1 3.30 1.68 0.05

06:01 to 07:00 h 12.2 3.30 1.79 −0.10

>07:00 h 16.4 3.29 1.92 −0.11

<0.0001 NS <0.01 <0.01

Week-weekend sleep-

phase displacement

<1h 12.7 3.33 1.81 0.09

1h–1h30 11.2 3.30 1.76 0.02

1h31–2h15 10.3 3.33 1.71 −0.01

>2h15 10.3 3.37 1.75 −0.09

NS NS NS NS

Bedtime oscillation along

the school week

≤1h 10.8 3.34 1.58 0.09

1h01–2h 10.6 3.32 1.85 −0.03

>2h 12.3 3.33 1.86 −0.07

NS NS <0.0001 <0.05

Rise-time oscillation

along the school week

<1h 8.5 3.41 1.62 0.07

1h–1h59 9.4 3.32 1.74

−0.01

≥2h 15.8 3.26 1.92 −0.07

<0.01 <0.05 <0.0001 NS

Expressed in a 6-point scale: ≤10 points [coded as 1]; >10 –11 points [coded as 2]; 12–13 points [coded as 3]; 14–15 points [coded as 4]; 16–17

points [coded as 5]; ≥18 points [coded as 6].

Assessed through a 5-point scale, coded from 0 = “strongly disagree” to 4 = “strongly agree.”

Transformed into standardized z scores. M =

mean values. NS = not significant ( p > 0.05).

Sleep and Academic Performance in Undergraduates 

was conducted by entering the five predictors found

plus neuroticism and its interactions with sleep

quality and enough sleep. To represent the interactions

between sleep quality or enough sleep and neuroticism,

the variables were first centered and then multiplied

together. The main effects for the five predictors pre-

viously selected by the stepwise regression analysis

remained statistically significant, but the main effect of

neuroticism was not significant ( p = .572), and nor

was its interaction with enough sleep ( p = .771) or

sleep quality ( p = .783).

DISCUSSION

The present study examined associations between sleep

patterns reported at the middle of the semester, self-

reported performance measures, and, most importantly,

subsequent academic achievement (end-of-semester

marks) in relation to demographics, well-being, and

academic and lifestyle variables in a sample of under-

graduates. Four sleep aspects were considered: quantity,

quality, regularity, and phase/schedule.

Our results with respect to the impact of adequate

sleep duration on end-of-semester marks are in line

with the findings of other naturalistic correlational

investigations (Oginska & Pokorski, 2006), as well as

with those of controlled studies, indicating that cumulat-

ive sleep debt (Van Dongen et al., 2003a), even in appar-

ently small amounts and in healthy young adults, is

associated with a variety of undesirable consequences

(Banks & Dinges, 2007; Pilcher & Huffcutt, 1996;

Spiegel et al., 1999; Van Dongen et al., 2003a, 2003b).

For example, Banks and Dinges (2007) found that

across nights of partial sleep deprivation, i.e., 14 nights

with 6 h of sleep/night, the accumulation of neurobeha-

vioral deficits may achieve levels equivalent to those

found after 1 to 3 nights of total sleep deprivation. If

these findings of controlled studies hold true in real-life

circumstances, it is then understandable that those stu-

dents obtaining enough sleep more frequently achieve

greater marks.

The association found in our sample between enough

sleep and end-of-semester marks is also comparable to

the results of similar correlational studies in undergradu-

ates reporting decreased academic performance with

shorter sleep duration (Borisenkov et al., 2010; Jean-

Louis et al., 1996; Kelly et al., 2001; Medeiros et al.,

2001; Trockel et al., 2000). Compared with these natura-

listic studies, one contributing factor of our investigation

is that rather than sleep length, per se, we have

TABLE 2. Descriptive statistics for the variables in the regression analyses—z scores (criterion variable) and potential predictors (n = 1240*)

Variable M SD Min Max

z scores 0.03 0.97 −2.73 3.06

Age 19.82 1.58 17 25

Vocational preferences match (1st, 2nd, 3rd, etc.) 1.45 0.73 1 3

Previous academic achievement (self-reported GPA) 3.41 0.96 1 6

Class attendance—transf. [log

] 0.18 0.24 0 0.70

Study h/wk—transf. [log

] 0.86 0.31 0 1.75

Cognitive functioning 9.56 2.46 0 16

Night outings (past midnight) 2.38 1.25 0 5

Cigarettes—transf. [log

] 0.16 0.35 0 1.43

Coffee—transf. [log

] 0.22 0.21 0 0.78

Alcohol—transf. [log

] 0.12 0.20 0 1.11

Other substances—transf. [log

] 0.06 0.15 0 0.70

Enough sleep (frequency) 2.56 1.12 0 4

Sleep Quality Index (7 items) 8.75 3.92 0 24

Weekend sleep phase 6:05 1h21 1:15 12:00

Bedtime irregularity along the school week 1h58 1h24 0h 9h

*Excluding 8 outliers and participants with missing data for 1 of the 16 variables (listwise deletion).

TABLE 3. Significant predictors of z scores selected through stepwise regression

Model (variables added at each step) RR

Adjusted R

Standard error of the estimate

Change statistics

Fdfp

1. Previous GPA 0.307 0.094 0.093 0.926 0.094 128.572 (1, 1238) <0.001

2. Class attendance* 0.362 0.131 0.130 0.908 0.037 52.849 (1, 1237) <0.001

3. Enough sleep (frequency) 0.369 0.136 0.134 0.905 0.005 6.814 (1, 1236) 0.009

4. Night outings 0.373 0.139 0.137 0.904 0.003 4.825 (1, 1235) 0.028

5. Sleep Quality Index 0.377 0.142 0.139 0.903 0.003 4.475 (1, 1234) 0.035

Constant included. Criterion variable: z scores. Durbin-Watson: 1.852. Condition Index = 14.270. Collinearity statistics for each predictor:

tolerance values > 0.89, VIF < 1.1. *Transf. log

 A. A. Gomes et al.

Chronobiology International

considered a measure of insufficient sleep, i.e., frequency

of enough sleep. This seemed to be more in line with the

notion of sleep deb t in the sense of chronic sleep restric-

tion (Van Dongen et al., 2003a).

Our findings on sleep quality are in agreement with pre-

vious studies showing associations between poor sleep,

lower academic performance of university students

(Gilbert & Weaver, 2010; Howel et al., 2004; Johns et al.,

1976), and other aspects of daytime functioning impair-

ment, e.g., subjective complaints such as increased feel-

ings of depression, tension, and fatigue, not only in sleep

clinical samples (AASM, 2005) but also in community

samples of young adults (Alapin et al., 2000; Oginska &

Pokorski, 2006; Pilcher & Ott, 1998; Pilcher et al., 1997).

Because of the nonexistence of a European Portuguese

version of the Pittsburgh Sleep Quality Index (PSQI;

Buysse et al., 1989), we adopted another measure of

sleep quality; however, it should be stressed that our

sleep-quality items show similarities with some of the

PSQI items. Similar results on the associations of both

sleep length and sleep quality with academic achievement

have also been reported for other age groups, including

children, e.g., a recent meta-analytic review of Dewal

et al. (2010). However, precaution is needed in generaliz-

ing the current findings to other age groups or educational

levels (e.g., Kopasz et al., 2010; Pagel et al., 2007).

As to sleep regularity, several associations with self-

reported academic performance emerged, and one

significant association with end-of-semester marks was

found by univariate analysis, but it lost significance in

multiple regression analysis. Abrupt shifts of sleep-wake

schedules lead to internal dissociation among circadian

rhythms (Reinberg et al., 1989), which may be accom-

panied by complaints of somnolence, attention deficit,

concentration difficulty, and performance decrement

that are commonly found in shiftwork and rapid travel

across multiple time zones (AASM, 2005). In undergradu-

ates, irregularities of 2 to 4 h in the sleep-wake schedule

are associated with increased fatigue, deterioration of

mood, and lower performance (Taub & Berger, 1973),

and irregular sleep-wake schedules of healthy students

are associated with excessive daytime somnolence com-

pared to regular colleagues (Manber et al., 1996). In our

sample, sleeping enough and with good quality were

more important in determining subsequent academic

performance than maintaining a regular sleep-wake sche-

dule, but no definitive conclusion should be drawn. It

seems critical to further scrutinize the potential impact

of irregularities of sleep-wake schedules in academic

samples using longitudinal measures, such as 1-wk sleep

logs or actigraphy, to assess irregularity (Medeiros et al.,

2001), as this topic seems to be under investigated.

Each of the three variables used to measure chrono-

type/sleep phase in the present study showed significant

relationships in univariate analysis with previous year

academic failure in most courses, perceived influence

of sleep in academic performance, and university

marks, and all (but one) were associated with prior

GPA. The literature consistently reports associations

between undergraduate later sleep-wake schedules and

lower academic performance (Bes¸oluk et al., 2011;

Borisenkov et al., 2010; Elliason et al., 2010; Johns et al.,

1976; Lack, 1986; Medeiros et al., 2001, 1996; Randler &

French, 2006; Smith et al., 1989; Trockel et al., 2000).

Nevertheless, the association between sleep phase and

end-of-semester grades lost its significance in stepwise

multiple regression after controlling for the effects of

class attendance, previous academic achievement,

night outings, sleep quantity, and sleep quality. Thus,

the association between late sleep-wake schedules/

phases and academic performance appears to be

mediated by other variables, such as sleep restriction,

which is more pronounced in evening-type students

(Giannotti et al., 2002; Gomes et al., 2008; Taillard

et al., 1999), and class attendance, which was already

found to be lower in evening-tendency students

(Gomes, 2006). Indeed, there are normal interindividual

differences linked to the peak time (acrophases) of circa-

dian rhythms, which manifests along a continuum from

morning- to evening-type individuals, with a majority

of intermediate persons between the two extremes

(Kerkhof, 1985). These would be normal variations, but

uniform work and school schedules do not consider

these differences; thus, inconsistencies may arise

between the individual preferred schedules and exter-

nally imposed time schedules, which may lead to

several undesirable consequences, such as when an

evening-type student is confronted with morning class

schedules (Bes¸oluk et al., 2011; Giannotti et al., 2002).

As to multiple regression analysis, in our study, based

on an initial pool of 30 demographic, academic, well-

being, neuroticism, and sleep variables, a stepwise

regression entering 15 potential predictors of academic

achievement for university students led to the selection

of 5 variables that remained significantly associated with

marks. Two were sleep variables, sleep quantity and

sleep quality. These two sleep variables added a small,

but independent and significant, contribution to univer-

sity grades beyond previous academic achievement and

class attendance. This low contribution may lead to a

questioning of their pertinence, and it may be argued

that significant associations emerged only by chance or

because of large sample size. Although this might be

true, both objections would also apply to any one of the

other variables considered. Furthermore, these two

sleep variables were better predictors of school achieve-

ment than other potential predictors, such as time

devoted to study, subjective cognitive functioning, or vo-

cational preferences. Using a stepwise regression method,

which tends to select a relatively small number of predic-

tors (Afifi et al., 2004), and considering the 15 potential

predictors of academic results, two of the four sleep vari-

ables emerged in the group of the five selected significant

predictors of student performance, whereas other appar-

ently logical predictors of academic achievement lost sig-

nificance and were not selected for the main group. In

Sleep and Academic Performance in Undergraduates 

addition, no evidence was found that neuroticism moder-

ates or influences the impact of sleep variables on end-

of-semester marks. In conclusion, the results of our

regression analysis study found that in students with

similar previous academic achievement and class attend-

ance, both sleep quantity and sleep quality may influence

academic results in such a way that sleep reduction or

poor-quality sleep may be associated with decreased

academic performance.

The current results are in line with controlled studies

cited at the beginning of our paper, showing consistent

associations between sleep and a set of cognitive activi-

ties, memory in particular. Our results are also consistent

with the review pap er on sleep and academic perform-

ance by Curcio et al. (2006), who concluded that, as

both REM and NREM sleep stages appear to be necessary

for learning and memory, sleep loss and sleep fragmen-

tation constitute a risk for efficient consolidation of

declarative and procedural knowledge and skills. In

light of this conclusion, it is understandable that sleep

restriction and poor-quality sleep may impact academic

performance of undergraduates.

It is interesting to note that the best predictors of end-

of-semester marks were previous academic achievement

and class attendance, in close agreement with findings

reported in the literature. In fact, several studies found

that student academic achievement prior to postsecond-

ary education, e.g., high-school GPA (Robins et al., 2004)

and admission tests (Kuncel et al., 2010; Sackett et al.,

2009; Trapmann et al., 2007), is related, to a lesser or

greater extent, with subsequent outcomes in college. As

to class attendance, a recent meta-analytic review con-

cluded that it explains a large degree of unique variance

in college grades, superior to other known predictors of

academic performance (Credé et al., 2010). However, to

date, surprisingly few studies, if any, interested in the

impact of sleep/biological rhythms on academic per-

formance have considered this variable.

The present study has several further important

strengths: (i) it encompasses a large sample of under-

graduates from a variety of academic fields (engineering,

exact and natural sciences, languages, education, and

management) that may be found in universities all over

the world; thus, our results have the potential to be

extrapolated to other universities and countries; (ii) the

sample represented in a balanced way both sexes and

three academic years; (iii) the study was conducted in a

single university; thus, all students were subjected to

identical class timings, examination schedules, and

school-year calendar; (iv) in line with expert recommen-

dations (Curcio et al., 2006; Wolfson & Carskadon, 2003),

several measures of academic achievement were col-

lected (multimeasure approach) and were not limited

to self-reported data; (v) the study included a longitudi-

nal analysis of the relationship between sleep patterns

and end-of-semester academic marks; therefore,

although the nonexperimental nature of the present

research does not allow for the extraction of causal

inferences, we can be sure about the tempo ral sequence

of the relationships found; (vi) a multivariable approach

was adopted; that is, besides sleep patterns, a variety of

other variables that might explain academic performance

of university students was considered, which helped to

better assess the relative impact of sleep on academic

performance in real-life circumstances.

Certain limitations of the present work should be

mentioned. First, we were unable to control for the

time-of-day of the assessment tasks and written examin-

ations. In addition, our study, like oth ers in the field of

sleep and biological rhythms, did not cover all of the

possible relevant variables that could be related to

academic performance. Some important variables not

considered by us, but currently known to predict

academic performance, were conscientiousness person-

ality dimension, which is a strong predictor of postse-

condary academic achievement (Kuncel et al., 2010;

O’Connor & Paunonen, 2007; Trapmann et al., 2007);

achievement motivation, the strongest predictor of GPA

in the meta-analysis of Robins et al. (2004); and academic

self-efficacy (Ferla et al., 2009; Robins et al., 2004).

Furthermore, the present study did not control for socio-

economic status (SES) even though Pagel et al. (2007)

reported that the number and type of sleep variables

affecting the GPA of adolescents changed after statisti-

cally controlling for age and household income. On the

other hand, SES is likely to be somewhat restricted in

range in university samples. Furthermore, in our study,

the most important predictor of end-of-semester marks

was previous academic achievement, which is also sup-

posed to have been influenced by SES (Robbins et al.,

2004; Sackett et al., 2009). Therefore, as prior academic

achievement was included in our final regression

model, we believe that, indirectly, the effect of SES is

not completely absent from our results regarding end-

of-semester marks. We also did not control for mental

or other medical disorders that might influence both

sleep and academic outcomes. However, our study con-

sidered neuroticism, which has been reported to be

strongly correlated with several mental disorders,

especially mood disorders, eating disorders, somatoform

disorders, anxiety disorders, and schizophrenia (Lahey,

2009). Furthermore, in our study, the associations

between lower academic performance, poor sleep

quality, and sleep debt were not moderated by neuroti-

cism. In particular, we did not control for sleep disorders,

such as sleep apnea, delayed sleep phase syndrome, or

narcolepsy (AASM, 2005), commonly detected for the

first time in this age group. We did not control either

for medication use (other than medication to promote

sleep), nor for its possible effects. Even though we

measured daytime sleepiness, the present study did not

fully explore the role of this variable over academic per-

formance; however, a very recent meta-analytic review

highlights the need to treat sleep duration, sleep

quality, and sleepiness as separate variables in future

research (Dewald et al., 2010). Although this review was

 A. A. Gomes et al.

Chronobiology International

based on child and adolescent sleep literature, we believe

future studies on undergraduates should follow this rec-

ommendation. Despite its limitations, to date our study

seems to be one of the most comprehensive within the

biological rhythm and sleep research fields, combining

an interesting set of different variables, two of which,

class attendance and previous academic achievement,

were identified in meta-analyses as among the most con-

sistent predictors of academic achievement (e.g., respect-

ively, Credé et al., 2010; Robins et al., 2004). Considering

feasibility, the use of a self-response questionnaire was a

perfectly adequate tool to collect data on sleep patterns

from a large sample at the same point-in-time, but the

limitations of this kind of instrument are well known.

Thus, ideally, self-report measures of sleep should be

complemented by more objective measures, such as

actigraphy or polysomnography. In the present research,

due to university restrictions for accessing student

records, only one mark per participant was collected at

the end of semester, wh ich may not be necessarily

representative of the student’s usual performance. This

limitation could explain the overall small percentage

of variance found in stepwise regression analyses.

Another potential reason for the relatively small variance

is likely a consequence of the longitudinal design

adopted. As we examined the prospective relationships

among sleep, academic performance, and life-style vari-

ables (measured in the middle of the school semester)

with subsequent academic performance (end-of-seme-

ster marks), changes in the variables studied between

the two points-in-time (middle of semester and end of

semester) might have occurred, or other variab les that

were not controlled by the researchers might have in-

truded and impacted end-of-semester achievement. In

summary, the significance of the relative contribution

of each sleep variable should be weighed, bearing in

mind the inherently noisy nature of academic perform-

ance in real-life circumstances.

In conclusion, adequate sleep with respect to quantity,

quality, and timing is likely associated with better marks

for university students. Our results have several impli-

cations. In particular, sleep education should be incor-

porated into existing prevention programs to promote

student health and academic success.

ACKNOWLEDGMENTS

Student participation and lecturers cooperation are

gratefully acknowledged. The assistance of colleagues in

data collection was precious. The present work is cur-

rently financially supported by the IBILI (Institute of Bio-

medical Research in Light and Image; Fundação para a

Ciência e a Tecnologia [FCT]), Faculty of Medicine, Uni-

versity of Coimbra, Portugal. The University of Aveiro,

Department of Educational Scienc es, now Department

of Education, provides logistic support (note: work par-

tially based on a larger research project formerly sup-

ported by F.C. Gulbenkian [LEIES Project] and by FCT-

Portugal [SPASHE project; UI-CCPSF research Unit]).

Declaration of Interest: The authors report no conflicts

of interest. The authors alone are responsible for the

content and writing of the paper.

APPENDIX

Sleep-Wake Questionnaire for University Students [SWQUS] - «during-the-semester» version

*Sex:□ Female. □ Male. *Age: ____ yr-old

* Do you have children? □ Yes. □ No.

* Undergraduate degree: _____________. *Curricular year: _____.

* Your current major was your… □ first □ second □ third or other … choice

* In the last year, have you completed enough course credits to progress to a new curricular year?

□ Yes. □ No, I have failed most courses.

* What is your student status? □ full-ti me student; □ part-time student due to part-/full-time job; □ elite athlete

student; □ student union delegate/representative

* Please indicate the town where you live…

… during the school week: __________ … and on weekends / holidays: ___________

I. SLEEP-WAKE CYCLE DURING THE SEMESTER

Last Month

Please consider

the last month, keeping in mind what usually happens in a typical class week.

* 1. In a typical class week, at what time do you usually…

… go to bed? (on average) ___ h ___ min

…get up? (on average) ___ h ___ min

* 2. On weekends, during the semester, at what time do you usuall y

… go to bed? (on average) ___ h ___ min

… get up? (on average) ___ h ___ min

Sleep and Academic Performance in Undergraduates 

* 3. During the school week, does your bedtime change from night to night?

□ Not at all □ Yes: it varies between ___ h ___ and ___ h ___

* 4. After going to bed, you usually fall asleep within …

□ 1-14 min □ 15-30 min □ 31-45 min □ 46-60 min □ more than 60 min

* 5. How often do you have trouble falling asleep?

□ never □ rarely □ sometimes □ 3-4 nights a week □ almost every night/always

* 6. How many times do you usually wake up during a night’s sleep?

□ none □ once □ 2-3 times per night □ 4-5 times per night □ 6 times or more

* 7. How often do you wake up spontaneously much earlier than needed (i.e., much earlier than your planned waking

time)?

□ never □ rarely □ sometimes □ 3-4 nights a week □ almost every night/always

* 8. Are nocturnal or early morning awakenings a problem for you?

□ not at all □ a bit □ somewhat □ often □ very often

* 9. During the school week, does your wake-up time change from day to day?

□ Not at all □ Yes, it varies between ___ h ___ and ___ h ___

10. After waking up, you usually get up within …

□ 1-14 min □ 15-30 min □ 31-45 min □ 46-60 min □ more than 60 min

11. During the semester, how many hours per night do you usually sleep on weekends?

□ 4h or less □ 4-5h □ 5-6h □ 6-7h □ 7-8h □ 8-9h □ 9-10h □ 10-11h □ 11h or more

12. In a typical class week, how many hours per night do you usually sleep?

□

4h or less □ 4-5h □ 5-6h □ 6-7h □ 7-8h □ 8-9h □ 9-10h □ 10-11h □ 11h or more

13. During a typical week of classes, does your sleep duration change from night to night?

□ Not at all □ Yes, it varies between ___ h ___ and ___ h ___

* 14. In a typical week during the semester, how often do you get the sleep hours you need?

□ never □ rarely □ 1-2 nights a week □ 3-4 nights a week □ almost every night/always

* 15. Regardless of its duration, how would you describe your…

[15.a] … sleep quality? □ very poor □ poor □ fair □ good □ very good

[15.b] .. sleep depth? □ very light □ light □ fairly deep □ deep □ very deep

16. Do you use medication to promote sleep?

□ never □ rarely □ sometimes □ often □ almost every night/always

17. Do you take naps?

□ never □ rarely □ sometimes □ several times a week □ almost always/always

* 18. Usually, during the day:

a) […]§

b) […]§

c) […]§

d) […]§

e) […]§

(§ items on daytime somnolence, each rated in a five-point Likert scale, from Manber et al., 1996)

f) How often do you feel excessively somnolent/sleepy during classes?

□ never □ rarely □

sometimes □ often □ very often/always

*19.

Not at all A bit Somewhat Much Very much

a) Energetic □□ □ □ □

b) Tired □□ □ □ □

c) Irritable □□ □ □ □

d) Alert □□ □ □ □

e) Depressed □□ □ □ □

f) Nervous □□ □ □ □

g) Happy □□ □ □ □

h) Productive □□ □ □ □

i) Relaxed □□ □ □ □

j) Efficient □□ □ □ □

k) Attentive □□ □ □ □

l) Motivated □□ □ □ □

m) Active □□ □ □ □

n) Having difficulties concentrating □□ □ □ □

 A. A. Gomes et al.

Chronobiology International

Other Sleep Aspects

20. How many hours of sleep per night do you need to feel well?

□ 4h or less □ 4-5h □ 5-6h □ 6-7h □ 7-8h □ 8-9h □ 9-10h □ 10-11h □ 11h or more

21. Did your sleep habits change at the university in comparison to high school?

□ not at all □ a bit □ somewhat □ much □ very much

22. In your opinion, do you have any sleep problems?

□ No □ Yes – Please describe: _________________________________

23. This academic year, did you ever stay awake all night to complete academic tasks?

□ No □ Yes – Please specify how many sleepless nights: _____

24. This academic year, did you ever stay awake all night due to other reasons?

□ No □ Yes – Please specify how many sleepless nights: _____

25. With respect to the place where you sleep most of the time when you are at the university:

a) Do you share your sleeping room with someone else?

□ No □ Yes – specify (e.g., colleague; brother): ______

b) Is your sleep disturbed…

… by noise? □ not at all □ a bit □ somewhat □ much □ very much

… by your roommate? □ not at all □ a bit □ somewhat □ much □ very much □ not applicable

II. WAKING LIFE

* 26. How many cigarettes do you smoke

per day (on average)? Week days: _____ Weekends: _____

* 27. How many glasses of alcoholic beverages do you drink

per day (on average)? Week days: ____ Weekends: _____

* 28. How many cups of coffee do you have

per day (on average)? Week days: _____ Weekends: _____

* 29. How often do you use other substances?

□ never □ rarely □ sometimes □ often/many times □ very often/always

* 30. How many hours a week do you exercise (on average)? ____

* 31. How many hours a week (on average) do you spend engaging in other extracurricular activities ? ____

* 32. During the class semester, how many hours a week (on average) do you spend studying? ____

* 33. How often do you go out at night (e.g., party, club, disco) until after midnight?

□ almost never □ once a month □ 2-3 times per month □ 1-2 times a week □ 3-4 times a week □ almost

always/every night

* 34. How many lectures do you attend (on average)?

□ every or almost every lecture □ more than half □ half □ less than half □ almost none or none

* 35.

For 2

or 3

year students: Indicate the answer that best describes your classifications at the university

(0-20 scale), on average:

□ 10 or less □ 10-11 □ 12-13 □ 14-15 □ 16-17 □ 18 or more

For 1

year students- Indicate your admission classification to the university (0-200):

□ less than 100 □ 100-114 □ 115-134 □ 135-154 □ 155-174 □ 175 or higher

* 36. Do you feel your sleep patterns have been negatively influencing your academic performance at the university?

□ strongly disagree □ disagree □ do not know/neither agree nor disagree □ agree □ strongly agree

REFERENCES

AASM (American Academy of Sleep Medicine). (2005). ICSD—Inter-

national Classification of Sleep Disorders, 2nd ed.: Diagnostic and

coding manual. Westchester, IL: AASM, 297 pp.

Afifi A, Clark VA, May S. (2004). Computer-aided multivariate analysis.

4th ed. Boca Raton, FL: Chapman & Hall/CRC, 489 pp.

Alapin I, Fichten CS, Libman E, Creti L, Bailes S, Wright J. (2000). How

is good and poor sleep in older adults and college students related

to daytime sleepiness, fatigue, and ability to concentrate? J. Psycho-

som. Res. 49:381–390.

Banks S, Dinges DF. (2007). Behavioral and physiological conse-

quences of sleep restriction. J. Clin. Sleep Med. 3:519–528.

Barton J, Spelten E, Totterdell P, Smith L, Folkard S, Costa G. (1995).

The Standard Shiftwork Index: a battery of questionnaires for asses-

sing shiftwork-related problems. Work Stress 9:4–30.

Bes¸oluk S, Önder I, Deveci I. (2011). Morningness-eveningness prefer-

ences and academic achievement of university students. Chrono-

biol. Int. 28:118–125.

Blagrove M, Akehurst L. (2001). Personality and the modulation of

effects of sleep loss on mood and cognition. Pers. Individ. Dif.

30:819–828.

Borisenkov MF, Perminova EV, Kosova AL. (2010). Chronotype, sleep

length, and school achievement of 11- to 23-year-old students in

northern European Russia. Chronobiol. Int. 27:1259–1270.

Bruni O, Antignani M, Innocenzi M, Ottaviano P, Ottaviano S. (1995).

Influence of sleep and temperament on school achievement. Sleep

Res. 24:91.

Buckhalt JA, Wolfson AR, El-Sheikh M. (2009). Children’s sleep and

school psychology practice. Sch. Psychol. Q. 24:60–69.

Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. (1989).

The Pittsburgh Sleep Quality Index: a new instrument for psychia-

tric practice and research. Psychiatry Res. 28:193–213.

Costa G, Sartori S, Facco P, Apostoli P. (2001). Health conditions of bus

drivers in a 6 year follow up study. J. Hum. Ergol. 30:405–410.

Credé M, Roch SG, Kieszczynka UM. (2010). Class attendance in

college: a meta-analytic review of the relationship of class

Sleep and Academic Performance in Undergraduates 

attendance with grades and student characteristics. Rev. Educ. Res.

80:272–295.

Curcio G, Ferrara M, De Gennaro L. (2006). Sleep loss, learning

capacity and academic performance. Sleep Med. Rev. 10:323–337.

Dewald JF, Meijer AM, Oort FJ, Kerkhof GA, Bögels SM. (2010). The

influence of sleep quality, sleep duration and sleepiness on

school performance in children and adolescents: a meta-analytic

review. Sleep Med. Rev. 14:179–189.

Diekelmann S, Wilhelm I, Born J. (2009). The whats and whens of

sleep-dependent memory consolidation. Sleep Med. Rev.

13:309–321.

Digdon NL, Howell AJ. (2008). College students who have an evening-

ness preference report lower self-control and greater procrastina-

tion. Chronobiol. Int. 25:1029–1046.

Dotto L. (1996). Sleep stages, memory and learning. Can. Med. Assoc. J.

154:1193–1196.

Eliasson AH, Lettieri CJ, Eliasson AH. (2010). Early to bed, early to rise!

Sleep habits and academic performance in college students. Sleep

Breath. 14:71–75.

Fafrowicz M, Oginska H, Mojsa-Kaja J, Marek T, Golonka K, Tucholska

T. (2010). Chronic sleep deficit and performance of a sustained at-

tention task—an electrooculography study. Chronobiol. Int.

934–944.

Ferla J, Valcke M, Cai Y. (2009). Academic self-efficacy and academic

self-concept: reconsidering structural relationships. Learn.

Individ. Dif. 19:499–505.

Ficca G. (2010). Caveats and pitfalls for the use of psychological models

in the study of sleep and memory [text and handouts]. Sleep and

Cognition Teaching Course, 20th Congress of the European Sleep

Research Society, Lisbon, 14–16 September 2010.

Ficca G, Salzarulo P. (2004). What in sleep is for memory. Sleep Med.

5:225–230.

Ficker JH, Wiest GH, Lehnert G, Meyer M, Hahn EG. (1999). Are

snoring medical students at risk of failing their exams? Sleep

22:205–209.

Fogel SM, Smith CT, Cote KA. (2007). Dissociable learning-dependent

changes in REM and non-REM sleep in declarative and procedural

memory systems. Behav. Brain Res. 180:48–61.

Giannotti F, Cortesi F. (2002). Sleep patterns and daytime functioning

in adolescence: an epidemiological survey of an Italian high

school student sample. Carskadon MA Adolescent sleep patterns.

Biological, social, and psychological influences. United Kingdom:

Cambridge University Press, pp. 132–147.

Giannotti F, Cortesi F, Sebastiani T, Ottaviano S. (2002). Circadian pre-

ference, sleep and daytime behaviour in adolescence. J. Sleep Res.

11:191–199.

Gilbert SP, Weaver CC. (2010). Sleep quality and academic perform-

ance in university students: a wake-up call for college psycholo-

gists. J. Coll. Stud. Psychother. 24:295

–306.

Gomes ACA. (2006). Sono, sucesso académico e bem estar em estu-

dantes universitários. [Sleep, well being and academic success in

undergraduates]. PhD dissertation. University of Aveiro, Aveiro,

Portugal.

Gomes AA, Tavares J, Azevedo MHP. (2002). Sleep-wake patterns and

academic performance in university students. Paper presented at

the European Conference on Educational Research, University of

Lisbon, Accessed 19 March, 2009, from 11–14 September 2002.

Education-Line [electronic database]. http://www.leeds.ac.uk/

educol/documents/00002200.htm.

Gomes AA, Tavares J, Azevedo MHP. (2008). Morningness-

eveningness in undergraduates: consequences over sleep-wake

patterns. Sleep 31(abstract suppl.):A53.

Gray EK, Watson D. (2002). General and specific traits of personality

and their relation to sleep and academic performance. J. Pers.

70:177–206.

Härmä M. (1993). Individual differences in tolerance to shiftwork: a

review. Ergonomics 36:101–109.

Harrison Y, Horne JA. (1997). Sleep deprivation affects speech. Sleep

20:871–877.

Harrison Y, Horne JA. (1999). One night of sleep loss impairs innova-

tive thinking and flexible decision making. Organ. Behav. Hum.

Decis. Process 78:128–145.

Harrison Y, Horne JA. (2000). The impact of sleep deprivation on

decision making: a review. J. Exp. Psychol. Appl. 6:236–249.

Hess B, Sherman MF, Goodman M. (2000). Eveningness predicts aca-

demic procrastination: the mediating role of neuroticism. J. Social

Behav. Pers. 15:61–74.

Hofman WF, Steenhof L. (1997). Sleep characteristics of Dutch adoles-

cents are related to school performance. Sleep Wake Res. 8:51–55.

Horne JA. (1988). Sleep loss and “divergent” thinking ability. Sleep

11:528–536.

Howell AJ, Jahrig JC, Powell RA. (2004). Sleep quality, sleep propensity

and academic performance. Percept. Mot. Skills 99:525–535.

Jean-Louis G, von Gizycki H, Zizi F, Friedman K, Spielman AJ, Taylor

D, Fullilove R, Taub H. (1996). Psychosocial determinants of sleepi-

ness and performance: consideration of gender differences. Sleep

Res. 25:101.

Johns MW, Dudley HA, Masterton JP. (1976). The sleep habits, person-

ality and academic performance of medical students. Med. Educ.

10:158–162.

Kelly WE, Kelly KE, Clanton RC. (2001). The relationship between sleep

length and grade-point average among college students. Coll. Stud.

J. 35:84–86.

Kerkhof GA. (1985). Inter-individual differences in the human circa-

dian system: a review. Biol. Psychol.

20:83–112.

Kopasz M, Loessl B, Hornyak M, Riemann D, Nissen C, Piosczyk H,

Voderholzer U. (2010). Sleep and memory in healthy children

and adolescents—a critical review. Sleep Med. Rev. 14:167–177.

Kuncel NR, Ones DS, Sackett PR. (2010). Individual differences as

predictors of work, educational, and broad life outcomes. Pers.

Individ. Dif. 49:331–336.

Lahey BB. (2009). Public health significance of neuroticism. Am.

Psychol. 64:241–256.

Li D, Wu Z, Shao D, Liu S. (1991). The relationship of sleep to learning

and memory. Int. J. Ment. Health 20:41–47.

Lim J, Dinges DF. (2010). A meta-analysis of the impact of short-

term sleep deprivation on cognitive variables. Psychol. Bull.

136:375–389.

Manber R, Bootzin RR, Acebo C, Carskadon M. (1996). The effects of

regularizating sleep-wake schedules on daytime sleepiness. Sleep

19:432–441.

Maquet P. (2001). The role of sleep in learning and memory. Science

294:1048–1052.

Medeiros ALD, Mendes DBF, Lima PF, Araujo JF. (2001). The relation-

ships between sleep-wake cycle and academic performance in

medical students. Biol. Rhythm Res. 32:263–270.

Meijer AM, Wittenboer GLH. (2004). The joint contribution of sleep,

intelligence and motivation to school performance. Pers. Indiv.

Dif. 37:95–106.

Neale JM, Liebert RM. (1986). Science and behavior: an introduction to

methods of research. 3rd ed. Englewood Cliffs, NJ: Prentice-Hall,

324 pp.

O’Connor MC, Paunonen SV. (2007). Big Five personality predictors

of post-secondary academic performance. Pers. Indiv. Dif.

43:971–990.

Oginska H, Pokorski J. (2006). Fatigue and mood correlates of sleep

length in three age-social groups: school children, students, and

employees. Chronobiol. Int. 23:1317–1328.

Pagel JF, Kwiatkowski CF. (2010). Sleep complaints affecting school

performance at different educational levels. Front. Neurol. 1:125.

Pagel JF, Forister N, Kwiatkowski C. (2007). Adolescent sleep disturb-

ance and school performance: the confounding variable of socio-

economics. J. Clin. Sleep Med. 3:19–23.

Pilcher JJ, Huffcutt AI. (1996). Effects of sleep deprivation on perform-

ance: a meta-analysis. Sleep 19:318–326.

Pilcher JJ, Ott ES. (1998). The relationships between sleep and

measures of health and well-being in college students: a repeated

measures approach. Behav. Med

. 23:170–178.

 A. A. Gomes et al.

Chronobiology International

Pilcher JJ, Ginter DR, Sadowsky B. (1997). Sleep quality versus sleep

quantity: relationships between sleep and measures of health,

well-being and sleepiness in college students. J. Psychosom. Res.

42:583–596.

Poropat AE. (2011). The Eysenckian personality factors and their corre-

lations with academic performance. Br. J. Educ. Psychol. 81:41–58.

Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methods

for biological rhythm research on animals and human beings.

Chronobiol. Int. 27:1911–1929.

Randler C, French D. (2006). Correlation between morningness-even-

ingness and final school leaving exams. Biol. Rhythm Res.

37:233–239.

Ravid S, Afek I, Suraiya S, Shahar E, Pillar G. (2009). Sleep disturbances

are associated with reduced school achievements in first-grade

pupils. Dev. Neuropsychol. 34:574–587.

Reinberg A, Motohashi Y, Bourdeleau P, Touitou Y, Nouguier J,

Nouguier J, Lévi F, Nicolaï A. (1989). Internal desynchronization

of circadian rhythms and tolerance of shift work. Chronobiologia

16:21–34.

Robbins SB, Lauver K, Le H, Davis D, Langley R, Carlstrom A. (2004).

Do psychosocial and study skill factors predict college outcomes?

A meta-analysis. Psychol. Bull. 130:261–288.

Roberts RE, Roberts CR, Chen IG. (2001). Functioning of adolescents

with symptoms of disturbed sleep. J. Youth Adolesc. 30:1–18.

Rodrigues RN, Viegas CA, Abreu E, Silva AA, Tavares P. (2002). Daytime

sleepiness and academic performance in medical students. Arq.

Neuropsiquiatr. 60:6–11.

Roehrs T, Roth T. (2000). Sleep-wake state and memory function. Sleep

23:564–568.

Roenneberg T, Kuehnle T, Pramstaller PP, Ricken J, Havel M, Guth A,

Merrow M. (2004). A marker for the end of adolescence. Curr. Biol.

14: R1038–R1039.

Sackett PR, Kuncel NR, Arneson JJ, Cooper SR, Water SD. (2009). Does

socioeconomic status explain the relationship between admissions

tests and post-secondary academic performance? Psychol. Bull.

135:1–22.

Sadeh A, Gruber R, Raviv A. (2003). The effects of sleep restriction and

extension on school-age children: what a difference an hour makes.

Child Dev. 74:444–455.

Saksvik IB, Bjorvatn B, Hetland H, Sandal GM, Pallesen S. (2011).

Individual differences in tolerance to shift work—a systematic

review. Sleep Med. Rev. 15:221–235.

Silva CF, Azevedo MHP, Dias MRVC. (1995). Estudo padronizado do

trabalho por turnos – versão portuguesa do SSI. Psychologica

13:27–36.

Singleton RA, Wolfson AR. (2009). Alcohol consumption, sleep, and

academic performance among college students. J. Stud. Alcohol

Drug. 70:355–363.

Smith C. (1995). Sleep states and memory processes. Special Issue: The

function of sleep. Behav. Brain Res. 69:137–145.

Smith C. (2001). Sleep states and memory processes in humans:

procedural versus declarative memory systems. Sleep Med. Rev.

5:491–506.

Smith CS, Reilly C, Midkiff K. (1989). Evaluation of three circadian

rhythm questionnaires with suggestions for an improved measure

of morningness. J. Appl. Psychol. 74:728–738.

Spiegel K, Leproult R, Van Cauter EV. (1999). Impact of sleep debt on

metabolic and endocrine function. Lancet 354:1435–1439.

Stickgold R, Walker MP. (2007). Sleep-dependent memory consolida-

tion and reconsolidation. Sleep Med. 8:331–343.

Stickgold R, Hobson JA, Fosse R, Fosse M. (2001). Sleep, learning and

dreams: off-line memory reprocessing. Science 294:1052–1057.

Taillard J, Philip P, Bioulac B. (1999). Morningness/eveningness and

the need for sleep. J. Sleep Res. 8:291–295.

Taub J, Berger R. (1973). Performance and mood following variations

in the length and timing of sleep. Psychobiology 10:559–570.

Taub J, Berger R. (1976). The effects of changing phase and duration of

sleep. J. Exp. Psychol. Hum. Percept. Perform. 2:30–41.

Taylor DJ, McFatter RM. (2003). Cognitive performance after sleep

deprivation: does personality make a difference? Pers. Indiv. Dif.

33:1179–1193.

Trapmann S, Hell B, Hirn J-OW, Schuler H. (2007). Meta-analysis of the

relationship between the Big Five and academic success at univer-

sity. J. Psychol. 215:132–151.

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.

Urner M, Tornic J, Bloch KE. (2009). Sleep patterns in high school and

university students: a longitudinal study. Chronobiol. Int.

26:1222–1234.

Van Dongen HPA, Maislin G, Mullington JM, Dinges DF. (2003a). The

cumulative cost of additional wakefulness: dose-response effects

on neurobehavioral functions and sleep physiology from chronic

sleep restriction and total sleep deprivation. Sleep 26:117–126.

Van Dongen HPA, Rogers NL, Dinges DF. (2003b). Sleep debt: theoreti-

cal and empirical issues [review]. Sleep Biol. Rhythms 1:5

–13.

Walker MP, Stickgold R. (2004). Sleep-dependent learning and

memory consolidation. Neuron 44:121–133.

Walker MP, Stickgold R. (2006). Sleep, memory, and plasticity. Annu.

Rev. Psychol. 57:139–166.

Wimmer F, Hoffmann RF, Bonato RA, Moffitt AR. (1992). The effects of

sleep deprivation on divergent thinking and attention processes. J.

Sleep Res. 1:223–230.

Wolfson AR, Carskadon MA. (1998). Sleep schedules and daytime

functioning in adolescents. Child. Dev. 69:875–887.

Wolfson AR, Carskadon, M. (2003). Understanding adolescents’ sleep

patterns and school performance: a critical appraisal. Sleep Med.

Rev. 7:491–506.

Sleep and Academic Performance in Undergraduates 

property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a

listserv without the copyright holder's express written permission. However, users may print, download, or

email articles for individual use.