Journal of Behavioral Medicine, Vol. 29, No. 3, June 2006 (C ?2006)
Assessment of Sleep Hygiene Using the Sleep Hygiene Index
David F. Mastin,1,5Jeff Bryson,2,3,4and Robert Corwyn1
Accepted for publication: January 23, 2006
Published online: March 24, 2006
The Sleep Hygiene Index was developed to assess the practice of sleep hygiene behaviors.
The Sleep Hygiene Index was delivered to 632 subjects and a subset of the subjects partici-
pated in a readministration of the instrument. Test–retest reliability analyses suggested that
sleep hygiene behaviors are relatively stable over time for a nonclinical population. Results
confirmed that sleep hygiene is strongly related to sleep quality and modestly related to per-
ceptions of daytime sleepiness. As predicted, support of the sleep hygiene construct was also
providedby strongcorrelationswiththeassociatedfeaturesof a diagnosisof inadequatesleep
hygiene. The Sleep Hygiene Index, a much shorter sleep hygiene instrument than previously
published, demonstrated comparable psychometric properties with additional evidence of va-
lidity and a clear item selection rationale.
KEY WORDS: sleep; hygiene; quality; sleepiness; causal indicators.
In recent years there has been increased atten-
tion and interest in poor sleep habits. A steady de-
cline in the time adults in the United States spend
asleep (National Sleep Foundation, 2002) and an es-
timated yearly cost to society of sleep problems rang-
ing in the tens of billions of US dollars (i.e. Leger,
1994) has created an interest in sleep quality and
sleep habits (also known as sleep hygiene). Sleep hy-
giene may be described as practicing behaviors that
facilitate sleep and avoiding behaviors that interfere
with sleep (Riedel, 2000). Inadequate sleep hygiene
is defined in the International Classification of Sleep
Disorders (American Sleep Disorders Association,
1990) as a “sleep disorder due to the performance
of daily living activities that are inconsistent with the
maintenance of good quality sleep and full daytime
alertness” (p. 73).
1University of Arkansas at Little Rock, Little Rock, Arkansas.
2Jacksonville State University, Jacksonville, Alabama.
3Fielding Graduate University, Santa Barbara, California.
4Behavioral Health Associates of North Alabama, P.C., Gadsden,
5To whom correspondence should be addressed at Depart-
ment of Psychology, University of Arkansas at Little Rock,
2801 South University Avenue, Little Rock, Arkansas; e-mail:
The current authors are aware of two existing in-
struments intended to assess adult sleep hygiene: the
Sleep Hygiene Awareness and Practice Scale (Lacks
and Rotert, 1986) and the Sleep Hygiene Self-Test
(Blake and Gomez, 1998). These instruments have
been found to have only fair internal consistency as
measured by Cronbach’s α (Sleep Hygiene Aware-
ness and Practice Scale = 0.47; Sleep Hygiene Self-
Test = 0.54). The Sleep Hygiene Self-Test has not
been shown to correlate with any external measure
of sleep quality and a reported correlation (Brown
et al., 2002) between the Pittsburgh Sleep Quality In-
dex and the Sleep Hygiene Awareness and Practice
Scale is questionable due to overlapping instrument
items. Further, authors of these instruments failed to
provide a clear rationale for item selection or present
data relating these instruments to subjective experi-
ences of sleepiness.
The objectives of the current paper are to (a) de-
scribe the psychometric assumptions relevant to the
measurement of sleep hygiene; (b) introduce and re-
port on the psychometric properties and construction
of a new instrument, the Sleep Hygiene Index (see
Appendix), an instrument with 13 items derived from
the diagnostic criteria for a diagnosis of inadequate
sleep hygiene as defined in the International Classifi-
0160-7715/06/0600-0223/0 C ?2006 Springer Science+Business Media, Inc.
224 Mastin, Bryson, and Corwyn
Association, 1990); (c) report on the relationship be-
tween sleep hygiene, subjective sleepiness, and sleep
quality in a nonclinical population; and (d) discuss
the utility of the construct of sleep hygiene and im-
plications for clinical practice and research.
Sleep hygiene is assessed by measuring be-
haviors and environmental variables thought to
cause or lead to relatively poor sleep quality rather
than measuring outcomes. For example, a typical
sleep hygiene item would query the subject as to
caffeine intake before bedtime. This use of items
thought to have a causal impact on the variable being
measured (causal items) necessitates a different set
of assumptions and design processes than is typical
in classical test theory (Bollen and Lennox, 1991;
Diamantopoulos and Winklhofer, 2001; MacCallum
and Browne, 1993; Streiner, 2003). For example,
one does not necessarily expect causal indicators
to be highly related with one another. Therefore,
Cronbach α should not be the only measure to report
in assessing the reliability of instruments that consist
of causal items. Following the guidelines of classical
test theory, authors of previous sleep hygiene instru-
ments may have discarded indicators with evidence
that items reduced overall or subscale coefficient α,
exhibited poor item-total correlations, or failed to fit
a factor analysis. Alternately, items with question-
able selection rationale may have been retained due
to their contributions to a favorable coefficient α.
Instead, especially in the case of the use of causal
items, authors are urged to look toward additional
indices of reliability relevant to the instrument being
designed (Bollen and Lennox, 1991). Test–retest
reliability is reported for the Sleep Hygiene Index
as an indicator of reliability over time as well as
Cronbach’s α as a measure of internal reliability.
It should also be understood that an instru-
ment that consists primarily of causal items is defin-
ing the variable of interest by these items (Bollen
and Lennox, 1991; Diamantopoulos and Winklhofer,
2001). It is understandable then that possessing a
clear rationale for item selection is critical. Although
the diagnosis of inadequate sleep hygiene has been
met with some criticism (Reynolds et al., 1991), the
current authors believe the International Classifica-
tion of Sleep Disorders diagnostic criteria for this dis-
order provides a suitable reference for agreement as
hygiene as a construct may be defined.
The model proposed in Fig. 1 presents the causal
items derived from the diagnostic criteria for inade-
quate sleep hygiene as x1through x13. The variable
of interest, sleep hygiene, is identified by η1. It should
benoted that thearrows point fromthecausal indica-
tors to the latent construct. Effect indicators derived
from the associated features of inadequate sleep hy-
giene (American Sleep Disorders Association, 1990)
are identified by y1through y5. Here, arrows repre-
senting the relation between latent construct and in-
dicators for these effect indicators point from the la-
tent construct to the indicators. Bollen and Lennox
(1991) have suggested that some measurement of ef-
fect is necessary for adequate latent variable iden-
tification. Two linked constructs, sleep quality and
subjective sleepiness, are identified by η2and η3, re-
spectively, measured here with the Pittsburgh Sleep
Quality Index (Buysse et al., 1989) and the Epworth
Sleepiness Scale (Johns, 1991).
Subjects were recruited from a midsized uni-
versity in the Midwest United States. Data were
collected from 632 volunteering psychology univer-
sity students over two academic semesters: alpha and
beta (alpha set: 103 males and 205 females, mean age
21.6; beta set: 125 males and 199 females, mean age
22.7). Less than 5% of the data was spoiled due to
skipped items or illegible handwriting. Students were
offered extra points upon completing the research
during class. No student rejected the opportunity. A
subset of group beta 141 (55 males and 86 females,
mean age 23.9) retook the Sleep Hygiene Index after
a 4–5-week interval to measure test–retest reliability.
Subset beta was comprised of attending students who
had participated in the first research opportunity
earlier in the semester. The research was reviewed
and approved by a university institutional review
board and participants gave informed consent to
All subjects completed the assessment in a
classroom setting in one sitting of less than 1 h.
All participants completed the Sleep Hygiene In-
dex and Epworth Sleepiness Scale (ESS). Group
alpha also completed the Pittsburgh Sleep Quality
Index. Participants were debriefed in an effort
to answer any questions regarding the study and
Sleep Hygiene Assessment 225
1: Sleep hygiene
y1-y5: Inadequate Sleep
Hygiene Associated Features
r2(599) = .06
r2(269) = .23
3: Subjective Sleepiness
2: Sleep Quality
I have more trouble paying attention
and thinking than I used to.
I feel sleepy during the day.
I worry about my sleep.
I feel I am more moody now than
I used to be.
I feel it takes more effort to get
things done than it used to.
r2(600) = .21
r2(598) = .14
r2(597) = .20
r2(600) = .19
r2(600) = .19
Fig. 1. Diagnostic criteria for inadequate sleep hygiene are x1through x11. Variable of interest, sleep hygiene,
is identified by η1. Associated features of inadequate sleep hygiene are identified by y1through y5. Two linked
constructs, sleep quality and subjective sleepiness, are identified by η2and η3, respectively. All correlations in this
figure were significant at the 0.05 level or less.
were provided an educative handout of sleep hy-
giene guidelines. Participants were offered access
to the results of the research upon completion of
Sleep Hygiene Index
The Sleep Hygiene Index, first presented here,
is a 13-item self-administered index intended to as-
sess the presence of behaviors thought to comprise
sleep hygiene. Participants were asked to indicate
how frequently they engage in specific behaviors (al-
ways, frequently, sometimes, rarely, never). Items
constructing the Sleep Hygiene Index were derived
from the diagnostic criteria for inadequate sleep
hygiene in the International Classification of Sleep
Disorders (American Sleep Disorders Association,
1990). Item scores were summed providing a global
assessment of sleep hygiene. Higher scores are in-
dicative of more maladaptive sleep hygiene status.
Epworth Sleepiness Scale
The Epworth Sleepiness Scale is a self-report
8-item questionnaire producing scores from 0 to
24. Scores greater than 10 suggest significant day-
time sleepiness (Johns, 1991). The Epworth Sleepi-
ness Scale has good psychometric properties (Johns,
1991), correlates with objective measures of sleepi-
ness (Chervin et al., 1997), and has been shown to
differentiate between individuals with and without
sleep disorders (Chervin et al., 1997) and those who
are and are not sleep deprived (Johnson, 1997).
Pittsburgh Sleep Quality Index
The Pittsburgh Sleep Quality Index is a self-
rated 19-item instrument intended to assess sleep
quality and sleep disturbance over a 1-month period
in clinical and nonclinical populations (Buysse et al.,
1989). Scores range from 0 to 21 with the higher
scores indicating poorer sleep quality. The Pittsburgh
Sleep Quality Index has been demonstrated to have
226Mastin, Bryson, and Corwyn
good internal reliability, stability over time, evidence
of validity (Buysse et al., 1989), and is well regarded
in the sleep research community.
There were 603 complete data sets for the Sleep
Hygiene Index resulting in M = 34.66 (SD = 6.6)
and a range of 17–55.
Although Cronbach’s α for the Sleep Hygiene
Index (α = 0.66) was found to be superior to previ-
ously published sleep hygiene instruments, we only
report it to give a general idea of internal con-
sistency. More importantly, the Sleep Hygiene In-
dex was found to have good test–retest reliability
(r(139) = 0.71, p<0.01).
The Sleep Hygiene Index was positively cor-
related (p<0.01) with all associated features of
inadequate sleep hygiene (American Sleep Dis-
orders Association, 1990) (y1 through y5; Pearson
r values ranged from 0.371 to 0.458). The Sleep
Hygiene Index correlated positively with the Ep-
worth Sleepiness Scale (r(599) = 0.244, p<0.01)
and the Pittsburgh Sleep Quality Index total score
(r(269) = 0.481, p<0.01). The Sleep Hygiene Index
also positively correlated (p<0.05 or less) with
all the Pittsburgh Sleep Quality Index component
scores (this is noteworthy in that some subcompo-
nents are more clearly independent from the sleep
hygiene construct in item design than others).
Our results support the sleep hygiene model
proposed in Fig. 1. Reliability analyses suggested that
sleep hygiene behaviors as measured by the Sleep
Hygiene Index are relatively stable over time for a
nonclinical population. Results confirmed the finding
by Brown et al. (2002) that sleep hygiene is strongly
related to sleep quality. For the first time, we are able
to report that sleep hygiene is modestly related to
perceptions of daytime sleepiness as would be ex-
pected in that poor sleep hygiene is thought to be
related to poor sleep quality. As predicted, support
of the sleep hygiene construct, as measured by the
Sleep Hygiene Index, was also provided by strong
correlations with the associated features of a diagno-
sis of inadequate sleep hygiene. The Sleep Hygiene
Index, a much shorter sleep hygiene instrument than
previously published, demonstrated comparable psy-
chometric properties with additional evidence of va-
lidity and a clear item selection rationale. This study
was limited in that a nonprobability sample was used.
The extent to which the Sleep Hygiene Index results
could be generalized across age groups for example,
Possessing a valid and reliable instrument may
be necessary, but may not be sufficient in under-
standing and modifying sleep hygiene behaviors. An
incongruity between sleep hygiene knowledge and
practices in clinical (Lacks and Rotert, 1986) and
non-clinical populations (Brown et al., 2002) suggests
that an understanding of sleep hygiene in context
may be important. We suggest that clinically it is im-
portant to understand that sleep hygiene does not ex-
ist in isolation and may be better understood when
considering the psychosocial context of the patient
(e.g., precipitation and/or maintenance of maladap-
tive sleep hygiene behaviors may not be addressed
by education alone). Further, it is likely to be unnec-
essary to create a core, or alternately, a comprehen-
sive list of sleep hygiene behaviors in the pursuit of
a quantitative assessment of sleep hygiene. With re-
gard to research it is important to (a) understand the
psychometric implications of the use of causal vari-
ables in the assessment of sleep hygiene (Bollen and
Lennox, 1991), (b) create more complete models of
sleep hygiene in an effort to understand and explain
the precipitation and maintenance of sleep hygiene
related behaviors, and (c) examine sleep hygiene be-
haviors independently in an effort to understand the
relative contribution of each to constructs of interest
(as suggested by Stepanski and Wyatt, 2003).
Sleep Hygiene Index Items
1. I take daytime naps lasting two or more
2. I go to bed at different times from day to day.
Sleep Hygiene Assessment227
3. I get out of bed at different times from day to
4. I exercise to the point of sweating within 1 h
of going to bed.
5. I stay in bed longer than I should two or three
times a week.
6. I use alcohol, tobacco, or caffeine within 4 h
of going to bed or after going to bed.
7. I do something that may wake me up before
bedtime (for example: play video games, use
the internet, or clean).
8. I go to bed feeling stressed, angry, upset, or
9. I use my bed for things other than sleeping or
sex (for example: watch television, read, eat,
10. I sleep on an uncomfortable bed (for exam-
ple: poor mattress or pillow, too much or not
11. I sleep in an uncomfortable bedroom (for ex-
or too noisy).
12. I do important work before bedtime (for ex-
ample: pay bills, schedule, or study).
13. I think, plan, or worry when I am in bed.
American Sleep Disorders Association (1990). International Clas-
sification of Sleep Disorders: Diagnostic and Coding Man-
ual, American Sleep Disorders Association, Rochester, MN,
Blake, D. D., and Gomez, M. H. (1998). A scale for assessing sleep
hygiene: Preliminary data. Psychol. Rep. 83: 1175–1178.
Bollen, K. B., and Lennox, R. (1991). Conventional wisdom on
measurement: A structural equation perspective. Psychol.
Bull. 110: 305–314.
Brown, F. C., Buboltz, W. C., and Soper, B. (2002). Relationship
of sleep hygiene awareness, sleep hygiene practices, and sleep
quality in university students. Behav. Med. 28: 33–39.
Buysse, D. J., Reynolds, C. F., Monk, T. H., and Timothy, H.
(1989). The Pittsburgh Sleep Quality Index: A New Instru-
ment for Psychiatric Practice and Research. Psychiatry Res.
Chervin, R. D., Aldrich, M. S., and Pickett, R. (1997). Comparison
of the results of the Epworth Sleepiness Scale and the multi-
ple sleep latency test. J. Psychosom. Res. 42: 145–155.
Diamantopoulos, A., and Winklhofer, H. M. (2001). Index con-
velopment. J. Mark. Res. 38: 269–277.
Johns, M. W. (1991). A new method for measuring daytime
sleepiness: The Epworth Sleepiness Scale. Sleep 14: 540–
Johnson, S. (1997). How sleepy are practicing physicians? A field
trial of the Epworth Sleepiness Scale. Sleep Res. 26: 671.
Lacks, P., and Rotert, M. (1986). Knowledge and practice of sleep
hygiene techniques in insomniacs and good sleepers. Behav.
Res. Ther. 24: 365–368.
Leger, D. (1994). The cost of sleep-related accidents: A report for
the National Commission on Sleep Disorders Research. Sleep
MacCallum, R. C., and Browne, M. W. (1993). The use of causal
indicators in covariance structure models: Some practical is-
sues. Psychol. Bull. 114: 533–541.
National Sleep Foundation (2002). Sleep in America Poll, National
Sleep Foundation, Washington, DC.
Reynolds, C. F., Kupfer, D. J., and Buysse, D. J. (1991). Subtyp-
ing DSM-III-R primary insomnia: A literature review by the
DSM-IV work group on sleep disorders. Am. J. Psychiatry
Oaks, CA, pp. 125–146.
Stepanski, E. J., and Wyatt, J. K. (2003). Use of sleep hy-
giene in the treatment of insomnia. Sleep Med. Rev. 7: 215–
Streiner, D. L. (2003). Being inconsistent about consistency: When
coefficient α does and doesn’t matter. J. Pers. Assess. 80: 217–