Content uploaded by Konrad S. Jankowski
Author content
All content in this area was uploaded by Konrad S. Jankowski on Feb 15, 2015
Content may be subject to copyright.
This article was downloaded by: [Konrad S. Jankowski]
On: 12 February 2015, At: 12:00
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Click for updates
Biological Rhythm Research
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/nbrr20
Is the shift in chronotype associated
with an alteration in well-being?
Konrad S. Jankowskia
a Faculty of Psychology, University of Warsaw, Stawki 5/7, 00-183,
Warsaw, Poland
Accepted author version posted online: 11 Nov 2014.Published
online: 27 Nov 2014.
To cite this article: Konrad S. Jankowski (2015) Is the shift in chronotype associated
with an alteration in well-being?, Biological Rhythm Research, 46:2, 237-248, DOI:
10.1080/09291016.2014.985000
To link to this article: http://dx.doi.org/10.1080/09291016.2014.985000
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Is the shift in chronotype associated with an alteration in well-being?
Konrad S. Jankowski*
Faculty of Psychology, University of Warsaw, Stawki 5/7, 00-183, Warsaw, Poland
(Received 22 September 2014; accepted 31 October 2014)
The study aimed to test whether a shift in chronotype (determined by mid-sleep on
free days) is associated with alterations in psychological well-being and sleep parame-
ters. One hundred and seventeen undergraduates were tested in longitudinal study with
four repeated measures. Measurements were taken during spring in three-week inter-
vals and each measurement consisted of self-reported sleep parameters on work and
free days (i.e. bedtime, sleep latency, wake time, sleep onset, mid-sleep time, social
jetlag), satisfaction with life, and mood (energetic arousal, tense arousal, hedonic
tone). Between-subjects analyses revealed earlier chronotypes, as compared to the later
ones, showing lower tense arousal, higher energetic arousal and life satisfaction, earlier
bedtime, sleep onset and offset on both work and free days, longer sleep duration and
shorter sleep latency on workdays, and less social jetlag. Within-subjects analyses
revealed increasing photoperiod associated with a shift toward earlier chronotype,
decrease in social jetlag, and shortening sleep latency. The seasonal shift toward earlier
chronotype was not associated with alterations in mood or life satisfaction, but it was
associated with a shift toward earlier bedtimes and longer sleep duration on workdays,
decrease in sleep latency, and social jetlag. Results from the within-subjects analyses
were consistent with the results of between-subjects analyses regarding sleep–wake
functioning, but inconsistent regarding psychological outcomes.
Keywords: chronotype; morningness–eveningness; sleep timing; social jetlag; mood;
life satisfaction
Introduction
Among individual characteristics of circadian functioning (Putilov et al. 2010; Ogińska
2011), chronotype, also named morningness–eveningness, gained most attention of
researchers. Chronotype reflects individual differences in the phase of entrainment and
is the final outcome of intrinsic components (e.g. genes) and environmental factors (e.g.
light; Roenneberg et al. 2007). People with earlier chronotype (e.g. morning types) exhi-
bit advanced circadian phase position in a number of physiological and psychological
characteristics, as compared to later chronotypes (e.g. evening types; Adan et al. 2012).
Probably, the easiest for self-observation facet of chronotype is sleep–wake rhythm,
namely sleep timing. Sleep timing on free days (precisely mid-sleep time) has been
identified as a marker of chronotype, assuming that on free days people are more likely
to express their phase of entrainment, due to less pressure of social obligations
(Roenneberg et al. 2003,2007). Thus, in late chronotypes, sleep occurs in later hours,
particularly on free days (Roenneberg et al. 2003). On the other hand, during the
workdays, evening individuals fall asleep much later than morning subjects do, but this
*Email: konrad.jankowski@psych.uw.edu.pl
© 2014 Taylor & Francis
Biological Rhythm Research, 2015
Vol. 46, No. 2, 237–248, http://dx.doi.org/10.1080/09291016.2014.985000
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
difference shrinks (Roenneberg et al. 2003) or disappears (Korczak et al. 2008) for wak-
ing up, as during workdays, the rising time in most people is constrained by social
duties (work, school, etc.). This results in shortened sleep duration in late individuals
during workdays (Zavada et al. 2005; Korczak et al. 2008); something they try to
recover during free days by prolonging sleep (Zavada et al. 2005). However, average
weekly sleep duration seems not related to sleep timing (Roenneberg et al. 2003). Late
chronotype is also related to other phenomenon –social jetlag (Wittmann et al. 2005).
Social jetlag, occurring more in evening chronotypes, conceptually represents misalign-
ment between social and biological time and is expressed as a time shift of sleep phase
between free days (biological time) and workdays (social time).
A number of studies revealed that the most central facets of psychological well-
being –life satisfaction and affect (Ryff 1989)–are associated with chronotype. Lower
life satisfaction has been linked to eveningness (Díaz-Morales et al. 2013) and similar
results have been observed for affect. Among various conceptualizations, a three-dimen-
sional model of mood seems to cover the widest denotation of core affective experience
(Schimmack & Grob 2000). In the framework developed by Matthews et al. (1990),
three dimensions of affect have been distinguished: energetic arousal (EA) (energetic–
tired), tense arousal (TA) (nervous–relaxed), and hedonic tone (HT) (pleasant–unpleas-
ant). Research on the above mood components and chronotype revealed no such time of
day when evening chronotypes exhibited more advantageous mood composition, as
compared to morning chronotypes (Adan & Guardia 1993; Jankowski & Ciarkowska
2008). On the contrary, evening people generally showed lower EA, HT, and
greater tension arousal (Jankowski 2014a)–a depression-like structure of affect
(Gozdzik-Zelazny et al. 2011). Furthermore, adverse mood has been also related to
greater social jetlag (Levandovski et al. 2011).
Consequently, there is a growing interest in factors influencing chronotype and in
the question whether these factors also influence sleep parameters and psychological
outcomes interrelated with chronotype. Light seems to be the main environmental factor
influencing the circadian system (Roenneberg et al. 2007), and individuals exposed to
sunlight earlier in the morning tend to be active and sleep at earlier times of day
(Borisenkov et al. 2012; Jankowski et al. 2014). The above observation was derived
from between-subjects comparisons of individuals living in different longitudes within
the same time zone, but this could also be tested in within-subjects research. Namely,
seasonal variability in sleep timing can be observed, as seasonal alteration in
photoperiod is linked to alteration in the timing of sunrise.
In a prospective study of ten Japanese participants (Honma et al. 1992), the earliest
sleep timing was observed in the summer, the latest in the winter, and intermediate in
the spring and autumn. Moreover, the seasonal difference was more pronounced for
sleep offset than sleep onset. In a between-subjects comparison, mean sleep onset
latency exceeding 30 min was more prevalent in winter than summer in southern
Norway, whereas in northern Norway, sleep onset latency exceeding 30 min was less
prevalent in winter than summer (Pallesen et al. 2001). In another study (Nixon et al.
2008), sleep duration in seven-year-old children from New Zealand was shorter in the
summer compared to the winter, autumn, and spring, whereas in Iceland, sleep duration
was shorter in the spring than in the winter in preschool children, but not in older indi-
viduals (Thorleifsdottir et al. 2002); however, there is also indication of lack of seasonal
variation in sleep (Park et al. 2007). Friborg et al. (2012) have shown seasonality of
sleep dependent on latitude. Namely, in Tromsø (69
o
N), sleep was altered in winter as
compared to summer, but not in Accra (5
o
N); in Norwegian university, students sleep
238 K.S. Jankowski
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
timing (only during workdays) was earlier and sleep latency was shorter (both for work-
days and weekend) in summer than it was in winter. Thus, in more central latitudes,
where there is less variation in photoperiod and consequently in time of sunrise, sleep
timing seems stable.
Seasonal variations also occur in moods, dependent on latitude. For example, indi-
viduals exhibited more depressive symptoms during the winter (when compared to the
summer months) (Park et al. 2007), in a city placed more to the north (Rochester
44
o
N), but not in the south (San Diego 32
o
N), where no summer to winter difference in
depression was observed. Comparison of sites more distant in latitude has shown that in
winter, as compared to summer, greater levels of depression and no difference in anxiety
were observed in Norwegian students and no seasonal alteration in affect among
Ghanaians (Friborg et al. 2012). Such lowering of mood in greater latitudes during
winter, in a clinical form known as seasonal affective disorder, has gained much
attention; however, it should be noted that other forms of mood seasonality (e.g.
summer depression) are existent, but to a lesser extent (Murray 2006). Interestingly,
peak of suicides generally occurs in late spring (Christodoulou et al. 2012), but other
seasonal effects in suicidality have been also reported (Aydin et al. 2013).
To sum up, relationships between chronotype or sleep timing and psychological out-
comes or other sleep parameters have resulted mostly from between-subjects compari-
sons; it is often assumed that on the individual level, a shift toward earlier sleep hours
alters these psychological and sleep characteristics. However, such supposition has yet
to be extensively verified. Longitudinal research on sleep timing and mood has shown
their dependence on photoperiod, however, whether or not alteration in sleep timing
itself is related to change in psychological outcomes and other sleep parameters have
not been tested. Namely, observations that in longer photoperiods, individuals sleep in
earlier hours and are less depressive do not imply existence of a link between shift in
sleep timing and shift in mood –it is not known whether mood improves only in those
subjects who shifted toward earlier sleep hours or mood improves in longer photoperi-
ods regardless of shift in sleep timing.
The main aim of the present research is to test whether a shift in chronotype (deter-
mined by mid-sleep time on free days) and social jetlag is associated with alterations in
psychological well-being and sleep parameters. Specifically, it is hypothesized that a
shift toward earlier chronotype and less social jetlag is accompanied by a shift toward
greater well-being, while a shift toward later chronotype and more social jetlag is
accompanied by a shift toward lower well-being, as such changes could be implied from
between-subjects comparisons. Thus, the present research tests state-like (e.g. environ-
mentally or light dependent) alterations in chronotype. These within-subjects relation-
ships are further supplemented with between-subjects associations to find out whether or
not both analyses provide concordant results, and conclusions derived from
between-subjects relationships could be translated into within-subjects recommendations.
Furthermore, this study aims to test whether longitudinal increase in photoperiod is
accompanied by a shift toward earlier sleep times, decrease of social jetlag, alteration of
other sleep parameters, and psychological well-being into more advantageous.
Methods
Questionnaires
Self-reported actual sleep parameters were measured based on the method proposed by
Roenneberg et al. (2003). It allows for quantification of sleep–wake parameters with a
Biological Rhythm Research 239
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
single administration of a questionnaire that takes about three minutes and shows good
concordance with sleep diaries (Roenneberg et al. 2003; Jankowski 2014b). Such a par-
adigm allows for repeated measurements without a need to use sleep diaries or actigra-
phy, which would make multiple repeated measures hard to accomplish (e.g. overload
of participants with sleep diaries or high costs of over a hundred actigraphs). The ques-
tionnaire consisted of questions referring to the most current living conditions and asked
about self-reported bedtime, time needed to fall asleep and wake-up time (sleep offset),
separately for work and free days, and the number of free days during a week. Based
on the above variables, secondary measures were calculated: sleep onset (bed-
time + sleep latency), mid-sleep time (half way between sleep onset and wake-up time)
and sleep duration; these variables were calculated separately for free and workdays.
Mid-sleep time on free days (MSF) was also calculated via a formula correcting for
sleep need (MSFsc; Roenneberg et al. 2007). MSF and MSFsc are considered to be
indicators of chronotype, as during free days, individuals are assumed to express their
phase of entrainment, as main obstacles, like social obligations (work, school, etc.) are
absent during free days. Consequently, in the present paper whenever the term “chrono-
type”appears, it refers to mid-sleep on free days, unless otherwise specified. Moreover,
mean weekly sleep duration and social jetlag were calculated. Social jetlag is the differ-
ence between MSF and mid-sleep time on workdays (MSW) and shows misalignment
of biological (MSF) and social time (MSW) –the greater the value, the greater the
misalignment (Wittmann et al. 2005).
Satisfaction with life was measured using the Polish translation of the Satisfaction
with Life Scale (SWLS) (Diener et al. 1985), which consists of five items scored with a
seven-point Likert-type response format. The Polish translation of SWLS was conducted
by the author of this manuscript using parallel blind technique (Behling & Law 2000).
SWLS measures global cognitive judgments of satisfaction with one’s life, and its lower
scores indicate lower satisfaction with life. The internal consistency of the SWLS
assessed by Cronbach’sαwas high in the present sample: .86. Test–retest reliability in
the present sample was: .85–.93 (three-week intervals); .87–.88 (six-week intervals); and
.86 (nine-week interval).
Mood was assessed with the UWIST Mood Adjective Check List (UMACL) devel-
oped by Matthews et al. (1990) in the Polish adaptation provided by Goryńska (2005).
The scale has 29 items scored with a four-point Likert-type response format and is
divided into three subscales measuring: EA (with poles: energetic–tired, 10 items); TA
(nervous–relaxed, nine items); and HT (pleasant–unpleasant, 10 items). Higher scores
indicate greater levels of each mood domain. Internal consistency for each subscale, as
indicated by Cronbach’sα, was high in the present sample (.88 for each scale).
Morningness–eveningness preference was assessed for descriptive purposes and for
between-subjects comparisons, using the Polish version of the reduced Morningness–
Eveningness Questionnaire (Jankowski 2013). The scale has 4 items scored with 4 or 5
response options. Internal consistency for the scale, as indicated by Cronbach’sα, was
high both in previous work (.73; Jankowski 2013) and in this study (.81). Test–retest
reliability with one-week interval in the present sample was .90.
Participants and procedure
The participants were 117 psychology students (101 females) aged between 19 and 31
(M= 22.2, SD = 1.9) from two universities situated in Warsaw (51
o
N). Subjects were
tested anonymously between 10:00 in the morning and 18:00 in the afternoon in groups
240 K.S. Jankowski
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
of 15 in classrooms just before beginning the classes, and each subject was tested at the
same time of a day across all measurements. According to the cutoff scores based on a
morningness–eveningness scores distribution in a student population (Jankowski 2013),
the present sample consisted of 21% morning types, 51% neither types, and 28% even-
ing types. There were four measurements taken from the beginning of April (a week
after a shift into summer time) to the beginning of June in three-week intervals. Each
measure consisted of self-reported sleep parameters, satisfaction with life, and mood.
Sixty-five individuals participated in all 4 measurements, 42 in 3 measurements, 5 in 2
measurements, and 5 in 1 measurement. Subjects who accomplished only one measure-
ment (5 females) were not considered in within-subjects analyses.
Statistical analyses were conducted using SPSS 21 software. At first, Pearson corre-
lations were calculated between rMEQ scores, MSF, MSFsc, social jetlag, and the
remaining variables. This analysis aimed to replicate previous findings; thus, all vari-
ables were averaged across all repeated measurements at first, and rcoefficients at
p< .05 one-tailed were considered statistically significant. The above analyses were also
conducted to establish reference rvalues for the succeeding analyses of within-subjects
relationships. Within-subjects relationships were studied using regression models for
repeated measures as proposed by Bland and Altman (1995). This method allows for
revealing bivariate relationships and might be viewed as equivalent of Pearson correla-
tion for within-subjects associations. Here, in the regression models, single predictors
were photoperiod, MSF, MSFsc, or social jetlag, and outcomes were the remaining vari-
ables. Reported are standardized coefficients (beta –here equivalence of Pearson rcoef-
ficient, as requals beta from regression with a single predictor) to check for strength of
pairwise relationships and unstandardized coefficients (B) to show how an increase of
one hour in a given predictor alters dependent variables (in raw scores for psychological
outcomes or in minutes for sleep related variables). Within-subjects relationships were
studied to test the main research question. Coefficients at p< .05 two-tailed were
considered statistically significant.
Results
Between-subjects relationships
Analyses of between-subjects relationships, as indicated by rPearson correlations,
revealed that morningness preference (rMEQ) and sleep timing on free days (MSF,
MSFsc) were highly intercorrelated. Individuals with greater morningness preference or
earlier sleep times on free days had lower TA and greater EA, but morningness and
sleep timing on free days were unrelated to HT (Table 1). Furthermore, individuals suf-
fering greater social jetlag were higher on TA. Subjects with earlier sleep times on free
days were more satisfied with their lives (SWLS), but morningness preference and
social jetlag turned out to be unrelated to life satisfaction.
As for sleep variables, morningness and sleep timing on free days showed consistent
relationships with bedtime, sleep onset, and sleep offset. Namely, individuals higher on
morningness were earlier going to bed, falling asleep, and waking up both on workdays
and free days. Moreover, individuals with earlier MSF and MSFsc needed less time to
fall asleep on workdays. Individuals with earlier MSF had longer sleep on workdays,
and individuals with greater morning preference had shorter sleep on free days. Individ-
uals higher on morningness and sleep timing on free days also had earlier mid-sleep on
workdays (MSW) and experienced less social jetlag.
Biological Rhythm Research 241
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
Social jetlag itself was also related to a number of sleep variables. Individuals suf-
fering greater jetlag were, on workdays, falling asleep later but waking up earlier, thus
their sleep duration was shorter. On free days, individuals higher on social jetlag were
waking up later, going to bed, and falling asleep later, thus, having shorter sleep latency.
Subjects with greater social jetlag had shorter mean sleep duration across a week.
Within-subjects relationships
Descriptive statistics for four measurements are presented in Table 2, but averaged
levels are not compared using statistic test due to missing measurements (see
participants and procedure). Thus, within-subjects regressions, which handle missing
data, are further used.
Within-subjects regressions revealed that increasing photoperiod resulted in a shift
toward earlier sleep timing in the studied sample (Table 3). Namely, every one-hour
increase in photoperiod resulted in an earlier MSF of 5.73 min and an earlier MSFsc of
5.54 min. Further, a one-hour increase in photoperiod produced 6.62 min less social
jetlag, almost one minute shorter sleep latency on both work and free days, and an ear-
lier bedtime, sleep onset, and offset on free days. Interestingly, increment in photoperiod
also resulted in lowering of HT and heightening TA. However, when the fourth (June)
measurement was removed from the analyses, effects of photoperiod on mood were
absent, while its effects on sleep remained unchanged. This removal of the June mea-
sure was done only for a test and all the analyses included the June measurement, and
Table 1. Correlation coefficients between studied variables averaged across all measurements
(between-subjects analyses).
rMEQ MSF MSFsc Social jetlag
HT .10 −.14 −.11 −.15
TA −.24** .19* .17* .18*
EA .18* −.18* −.18* −.15
SWLS .05 −.24** −.24** −.11
Bedtime W −.52*** .74*** .70*** .15
Sleep latency W −.15 .20* .19* .04
Sleep onset W −.54*** .76*** .71*** .16*
Sleep offset W −.45*** .50*** .64*** −.24**
Sleep duration W .07 −.23** −.05 −.38***
Bedtime F −.57*** .91*** .91*** .58***
Sleep latency F −.04 −.01 .01 −.18*
Sleep onset F −.58*** .92*** .92*** .56***
Sleep offset F −.72*** .93*** .83*** .63***
Sleep duration F −.25** .12 −.02 .15
Mean sleep duration −.01 −.15 −.02 −.26**
MSW −.57*** .73*** .78*** −.05
Social jetlag −.38*** .65*** .50***
MSFsc −.61*** .94***
MSF −.70***
Note: rMEQ –morningness preference, MSF –mid-sleep on free days, MSFsc –mid-sleep on free days
corrected for sleep need, MSW –mid-sleep on workdays, HT –hedonic tone, TA –tense arousal, EA –
energetic arousal, SWLS –life satisfaction, F –free days, W –workdays.
*p < .05; **p < .01; ***p < .001.
242 K.S. Jankowski
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
Table 2. Means (standard error) of the studied variables at the four measurements.
1234
Photoperiod 12:54 (00:00) 14:31 (00:00) 15:33 (00:00) 16:29 (00:00)
MSF 05:27 (00:07) 05:28 (00:07) 05:11 (00:06) 05:14 (00:07)
MSFsc 04:20 (00:07) 04:25 (00:07) 04:13 (00:06) 04:12 (00:07)
MSW 03:38 (00:05) 03:40 (00:05) 03:36 (00:05) 03:45 (00:04)
Social jetlag 01:49 (00:05) 01:47 (00:05) 01:34 (00:05) 01:28 (00:05)
HT 30.14 (.58) 29.40 (.67) 30.21 (.64) 29.02 (.69)
TA 16.55 (.56) 17.33 (.61) 17.15 (.57) 17.98 (.59)
EA 27.50 (.75) 28.35 (.64) 28.72 (.68) 27.28 (.72)
SWLS 22.83 (.60) 22.54 (.62) 23.13 (.67) 22.95 (.72)
Bedtime W 23:40 (00:06) 23:46 (00:06) 23:39 (00:06) 23:51 (00:05)
Sleep latency W 19.40 (1.53) 16.72 (1.05) 15.43 (.93) 14.94 (1.00)
Sleep onset W 00:00 (00:06) 00:02 (00:06) 23:53 (00:06) 00:05 (00:05)
Sleep offset W 07:17 (00:07) 07:19 (00:06) 07:20 (00:07) 07:26 (00:06)
Sleep duration W 07:18 (00:08) 07:17 (00:07) 07:28 (00:07) 07:20 (00:07)
Bedtime F 00:43 (00:08) 00:49 (00:07) 00:31 (00:06) 00:36 (00:07)
Sleep latency F 17.31 (1.39) 15.52 (.87) 15:33 (.91) 13:90 (.93)
Sleep onset F 01:01 (00:07) 01:04 (00:07) 00:46 (00:06) 00:50 (00:07)
Sleep offset F 09:54 (00:08) 09:51 (00:08) 09:35 (00:08) 09:39 (00:09)
Sleep duration F 8:53 (00:07) 08:46 (00:07) 08:48 (00:06) 08:48 (00:07)
Mean sleep duration 07:46 (00:07) 07:46 (00:06) 07:53 (00:06) 07:48 (00:06)
Note: MSF –mid-sleep on free days, MSFsc –mid-sleep on free days corrected for sleep need, MSW –mid-
sleep on workdays, HT –hedonic tone, TA –tense arousal, EA –energetic arousal, SWLS –life satisfaction,
F–free days, W –workdays.
Table 3. Results of within-subjects regression analyses. Values in front of brackets show
standardized coefficients. Values in brackets show alteration of a dependent variable (rows) in raw
scores (for psychological outcomes) and minutes (for remaining variables) if a predictor (columns)
increases for 1 h.
Photoperiod MSF MSFsc Social jetlag
HT −.12 (−.36)* .09 (.75) .06 (.44) .06 (.48)
TA .18 (.42)** −.00 (−.04) .02 (.18) −.02 (−.16)
EA −.06 (−.17) .04 (.31) .02 (.15) .07 (.54)
SWLS .04 (.06) .01 (.05) .06 (.22) −.05 (−.17)
Bedtime W .08 (1.26) .20 (9.44)*** .02 (.89) −.19 (−8.06)**
Sleep latency W −.21 (−.90)*** .15 (1.94)* .09 (.97) .07 (.85)
Sleep onset W .03 (.53) .23 (11.01)*** .04 (1.83) −.17 (−7.64)**
Sleep offset W .06 (1.25) .02 (1.11) .31 (16.27)*** −.45 (−24.71)***
Sleep duration W .03 (.72) .13 (−9.89)* .22 (14.44)** −.25 (−17.08)***
Bedtime F −.20 (−5.23)*** .76 (57.80)*** .88 (57.83)*** .58 (39.50)***
Sleep latency F −.20 (−.81)*** .12 (1.44)* .12 (1.25)* .07 (.79)
Sleep onset F −.23 (−6.00)*** .77 (59.57)**** .89 (59.30)*** .58 (40.39)***
Sleep offset F −.21 (−5.46)*** .78 (60.43)*** .37 (24.52)*** .67 (47.27)***
Sleep duration F .02 (.54) .01 (0.86) −.41 (−34.79)*** .08 (6.87)
Mean sleep
duration
.05 (1.13) −.10 (−6.98) .02 (1.39) −.15 (−9.42)*
MSW .07 (.89) .15 (6.06)** .27 (9.05)*** −.45 (−16.17)***
Social jetlag −.30 (−6.62)*** .81 (53.94)*** .57 (32.87)***
MSFsc −.24 (−5.54)*** .81 (56.15)***
MSF −.28 (−5.73)***
Note: MSF –mid-sleep on free days, MSFsc –mid-sleep on free days corrected for sleep need, MSW –mid-
sleep on workdays, HT –hedonic tone, TA –tense arousal, EA –energetic arousal, SWLS –life satisfaction,
F–free days, W –workdays.
*p < .05; **p < .01; ***p < .001.
Biological Rhythm Research 243
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
all the results and conclusions are based on all available data (including the fourth
measurement). The reason for testing what would happen if the June measurement was
discarded is addressed in the discussion. It should be noted that the strength of relation-
ship between photoperiod and MSF was of −.28, what means that about 8% of variabil-
ity in MSF was explained by changes in photoperiod, and a number of other factors
could affect shifts in MSF.
A shift in sleep timing on free days or in social jetlag was not associated with alter-
ation in either mood or life satisfaction. Alterations in mood and life satisfaction were
also unrelated to change in any other sleep variables, for example, sleep duration (these
results are omitted in the tables as their analyses were not the main aim of this study).
Regarding sleep variables, a shift toward earlier sleep timing on free days, as indicated
by earlier MSF, was linked with a shift toward earlier bedtime, sleep onset, and shorter
sleep latency on both work and free days. Sleep offset on workdays was unrelated to
MSF. Further, an hour shift toward earlier MSF resulted in 9.89 min longer sleep on
workdays, 6.06 min earlier mid-sleep on workdays, and a reduction of social jetlag by
53.94 min.
A shift in MSFsc, although highly correlated with the timing of MSF, was, in some
aspects, differently related to sleep variables and similarly in others, as compared to
MSF. Namely, an earlier MSFsc was unrelated to bedtime, sleep onset and latency on
workdays, while it was linked to an earlier sleep offset on workdays. Further, unlike
MSF, earlier MSFsc was associated with shorter sleep duration on workdays and longer
sleep duration on free days. Relationships between MSFsc and other variables were sim-
ilar to those for MSF.
Decrease in social jetlag was related to later bedtime, sleep onset and sleep offset on
workdays and to earlier bedtime, sleep onset and sleep offset on free days. Thus, reduc-
tion of social jetlag was associated with a shift toward later mid-sleep on workdays and
earlier mid-sleep on free days, however, still, sleep timing on workdays remained earlier
than that on free days (reduction in social jetlag meant more similar sleep timing across
all week). Further, an hour reduction in social jetlag was linked to a 17.08 min increase
in sleep duration on workdays, which resulted in a 9.42 min increase in weekly mean
sleep duration.
Discussion
The present research aimed to test whether conclusions that earlier chronotypes, as com-
pared to the later ones, present more advantageous sleep–wake rhythm and well-being,
could be confirmed by observations that a shift in chronotype is related to a change in
well-being and alterations in sleep–wake functioning.
Results of between-subjects relationships showed indeed, that earlier chronotypes, as
compared to later ones, exhibit lower TA, higher EA and higher life satisfaction. These
results are concordant with previous findings regarding mood (Adan & Guardia 1993;
Jankowski & Ciarkowska 2008; Jankowski 2014a), life satisfaction (Díaz-Morales et al.
2013), and other indicators of well-being (Haraszti et al. 2014), and provide deepened
insight, showing that individual differences in chronotype indicated by MSF and MSFsc
are related to better mood and higher life satisfaction. This provides evidence that indi-
vidual differences in chronotype might be important for individual levels of TA, EA and
life satisfaction –the observation also resembling the previous one regarding depressive-
ness (Levandovski et al. 2011). Further, individuals with greater social jetlag showed
greater TA; this is coherent with a study showing greater social jetlag related to
244 K.S. Jankowski
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
depressiveness (Levandovski et al. 2011). However, relationships of social jetlag with
other mood components and life satisfaction were statistically non-significant –presum-
ably because of insufficient sample size, as the study was designed primarily for within-
subjects analyses. Furthermore, the between-subjects relationships replicated previous
findings showing later chronotype related to greater social jetlag, later bedtimes, and
sleep onsets and offsets both on free days and workdays (Roenneberg et al. 2003;
Zavada et al. 2005; Korczak et al. 2008). Moreover, in the present study, later chrono-
type, as indicated by MSF, was also related to longer sleep latency and shorter sleep
duration on workdays –the former relationship presumably can be ascribed to too early
timing of sleep on workdays as compared to the internal time.
Within-subjects analyses showed that, during the spring, increasing photoperiod was
related to a shift toward earlier chronotype (MSF and MSFsc). As sleep timing on
workdays was not related to photoperiod, it could be concluded that on workdays indi-
viduals mainly follow social time, while on free days individuals are allowed to be cou-
pled to sun time –a relationship previously suggested in between-subjects analyses
(Borisenkov et al. 2012; Jankowski et al. 2014). Moreover, the present results showed
sleep latency, both on free and workdays, and social jetlag decreasing with increasing
photoperiod, suggesting that longer daylight has a positive effect on the speed of falling
asleep and harmony between internal and social time.
The present results of within-subjects analyses showed that a shift toward earlier
chronotype, as indicated by MSF, was related to advantageous alteration in sleep–wake
functioning, whereas a shift toward later chronotype was related to disadvantageous
alterations in sleep. Namely, when individuals shifted toward earlier chronotype, not
only did their sleep timing on workdays become earlier, but also, their sleep latency on
workdays and free days decreased; in addition, sleep duration on workdays increased
and social jetlag diminished. Interestingly, some discrepancy between the results based
on MSF and MSFsc could be observed, which suggests that the correction formula used
to calculate MSFsc advances MSF beyond the physiological meaningfulness. Namely,
the results obtained using MSFsc showing that people shifting toward earlier chronotype
also start to sleep less on workdays and sleep more on free days, are counterintuitive
(these are like for evening chronotypes) and against conclusions derived from between-
subjects analyses (Zavada et al. 2005; Korczak et al. 2008). The correction formula for
sleep debt applied to the MSFsc indicator (Roenneberg et al. 2007) assumes that on free
days people, in a linear manner, catch up the exact number of hours of sleep lost during
workdays, by recovery oversleeping. This, however, seems not be the mechanism, as
people recover from sleep debt by prolonging their sleep only to some extent (much less
than the number of hours of sleep loss) due to increased sleep efficiency of recovery
sleep (Herscovitch & Broughton 1981; Van Dongen et al. 2003; Alhola & Polo-Kantola
2007).
Confronting the results of within-subjects relationships with between-subjects associ-
ations it can be concluded that the two provide similar results regarding sleep. However,
a shift in chronotype was not accompanied by an alteration in mood or life satisfaction,
despite the observation that these psychological outcomes were more favorable in more
morning oriented subjects than in the evening ones. This suggests that disadvantageous
mood and lower life satisfaction are intrinsic components of late chronotype, with a
common trait-like underpinnings, for example, genes (Mansour et al. 2006) or tempera-
ment (Jankowski 2014a). On the other hand, state-like components of chronotype, for
instance, the environmentally or light-dependent ones, seem to be unrelated to
well-being –conclusion corresponding to the results showing that residents of Western
Biological Rhythm Research 245
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
localization within a time zone do not exhibit more depressive symptoms than those
living on the east (differences in sun time), despite they show later sleep timing
(Jankowski et al. 2014). Nevertheless, it should be noted that externally evoked altera-
tions in sleep timing may evoke changes in well-being in some circumstances. For
instance, changing school start times into later ones was shown to decrease depressive
symptoms in high school students (Wahistrom 2002). Thus, when analyzing effects of
alterations in sleep times, it seems important to consider whether change in sleep times
refer to free or workdays and the origin of the change (e.g. spontaneous, induced by
sunlight vs. imposed by the authorities).
Interestingly, increasing photoperiod was related to a shift toward higher TA and
lower HT –an association that disappeared when, only for the test, the last (June) mea-
surement was excluded from the analyses. This implicates two interpretations. First, lati-
tudes distant from the equator have a high variability in day length, with highly
increasing photoperiods during spring, believed to cause irritability and consequently
peak of suicides in late spring (Christodoulou et al. 2012). Second, in the student sam-
ple studied here, the last measurement was taken near the end of semester when antici-
pation of an examination session could have affected their mood. Thus, the study period
and sample specificity are the main limitations of the present research, and the presented
results require further replications in different study designs. It should be tested whether
or not a shift in chronotype within other seasons or between seasons is also unrelated to
alteration in well-being. It could be also asked whether the presented results occur in
other populations (e.g. workers, students from lower levels of the education system) or
when alterations in chronotype are evoked by experimentally controlled manipulations
instead of spontaneous drifts partly associated with the seasonal shifts in photoperiod.
Moreover, in the studied sample, males were underrepresented to the extent that gender
differences could not be addressed. Considering that gender is known to affect individ-
ual preferences in circadian functioning (Putilov et al. 2008), the presented results might
represent effects most likely to occur in females and further studies with more male
participants should be carried out.
Nevertheless, the present study is the first showing that recommendations regarding
psychological outcomes derived from comparisons of individuals differing in chronotype
might not be transferable into individual level recommendations. Namely, it exhibited
that a shift in chronotype was not accompanied by alterations in mood or life satisfac-
tion. Such a result is of great concern as it gives direction to further studies to consider
intrinsic underpinnings of chronotype in relationship to well-being, rather than environ-
mentally dependent components of circadian phase position.
Funding
The study was supported by a grant [grant number 2011/03/D/HS6/05760] from the National
Science Centre (Poland).
References
Adan A, Archer SN, Hidalgo MP, Di Milia L, Natale V, Randler C. 2012. Circadian typology: a
comprehensive review. Chronobiol Int. 29:1153–1175.
Adan A, Guardia J. 1993. Circadian variations of self-reported activation: a multidimensional
approach. Chronobiologia. 20:233–244.
Alhola P, Polo-Kantola P. 2007. Sleep deprivation: impact on cognitive performance.
Neuropsychiatr Dis Treat. 3:553–567.
246 K.S. Jankowski
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
Aydin A, Gulec M, Boysan M, Selvi Y, Selvi F, Kadak MT, Besiroglu L. 2013. Seasonality of
self-destructive behaviour: seasonal variations in demographic and suicidal characteristics in
Van, Turkey. Int J Psychiatry Clin Pract. 17:110–119.
Behling O, Law KS. 2000. Translating questionnaires and other research instruments: problems
and solutions. Thousand Oaks (CA): Sage.
Bland JM, Altman DG. 1995. Statistics notes: calculating correlation coefficients with repeated
observations: part 1–correlation within subjects. Br Med J. 310:446.
Borisenkov MF, Fradkova LI, Kolomeichuk SN. 2012. Factors affecting chronotype of adolescents
in the north according to multiple regression analysis. Chronobiol Int. 29:1418–1419.
Christodoulou C, Efstathiou V, Bouras G, Korkoliakou P, Lykouras L. 2012. Seasonal variation of
suicide. A brief review. Encephalos. 49:73–79.
Díaz-Morales J, Jankowski KS, Vollmer C, Randler C. 2013. Morningness and life satisfaction:
further evidence from Spain. Chronobiol Int. 30:1283–1285.
Diener E, Emmons RA, Larsen RJ, Griffin S. 1985. The satisfaction with life scale. J Pers Asses.
49:71–75.
Friborg O, Bjorvatn B, Amponsah B, Pallesen S. 2012. Associations between seasonal variations
in day length (photoperiod), sleep timing, sleep quality and mood: a comparison between
Ghana (5°) and Norway (69°). J Sleep Res. 21:176–184.
Goryńska E. 2005. The mood adjective checklist (UMACL) by Matthews, Chamberlain & Jones.
Warsaw: Polskie Towarzystwo Psychologiczne. Polish.
Gozdzik-Zelazny A, Borecki L, Pokorski M. 2011. Depressive symptoms in schizophrenic
patients. Eur J Med Res. 16:549–552.
Haraszti RÁ, Purebl G, Salavecz G, Poole L, Dockray S, Steptoe A. 2014. Morningness–evening-
ness interferes with perceived health, physical activity, diet and stress levels in working
women: a cross-sectional study. Chronobiol Int. 31:829–837.
Herscovitch J, Broughton R. 1981. Sensitivity of the Stanford sleepiness scale to the effects of
cumulative partial sleep deprivation and recovery oversleeping. Sleep. 4:83–91.
Honma KI, Honma S, Kohsaka M, Fukuda N. 1992. Seasonal variation in the human circadian
rhythm –dissociation between sleep and temperature rhythm. Am J Physiol. 262:885–891.
Jankowski KS. 2013. Polish version of the reduced morningness–eveningness questionnaire. Biol
Rhythm Res. 44:427–433.
Jankowski KS. 2014a. The role of temperament in the relationship between morningness–evening-
ness and mood. Chronobiol Int. 31:114–122.
Jankowski KS. 2014b. Composite scale of morningness: psychometric properties, validity with
Munich ChronoType Questionnaire and age/sex differences in Poland. Eur Psychiatry.
doi:10.1016/j.eurpsy.2014.01.004.
Jankowski KS, Ciarkowska W. 2008. Diurnal variation in energetic arousal, tense arousal, and
hedonic tone in extreme morning and evening types. Chronobiol Int. 25:577–595.
Jankowski KS, Vollmer C, Linke M, Randler C. 2014. Differences in sun time within the same
time zone affect sleep-wake and social rhythms, but not morningness preference: findings
from a Polish-German comparison study. Time Soc. 23:258–276.
Korczak AL, Martynhak BJ, Pedrazzoli M, Brito AF, Louzada FM. 2008. Influence of chronotype
and social zeitgebers on sleep/wake patterns. Braz J Med Biol Res. 41:914–919.
Levandovski R, Dantas G, Fernandes LC, Caumo W, Torres I, Roenneberg T, Hidalgo MP,
Allebrandt KV. 2011. Depression scores associate with chronotype and social jetlag in a rural
population. Chronobiol Int. 28:771–778.
Mansour HA, Wood J, Logue T, Chowdari KV, Dayal M, Kupfer DJ, Monk TH, Devlin B,
Nimgaonkar VL. 2006. Association study of eight circadian genes with bipolar I disorder,
schizoaffective disorder and schizophrenia. Genes Brain Behav. 5:150–157.
Matthews G, Jones DM, Chamberlain AG. 1990. Refining the measurement of mood: the UWIST
Mood Adjective Checklist. Br J Psychol. 81:17–42.
Murray GW. 2006. Seasonality, personality, and the circadian regulation of mood. New York
(NY): Nova Science.
Nixon GM, Thompson JMD, Han DY, Becroft DM, Clark PM, Robinson E, Waldie KE, Wild CJ,
Black PN, Mitchell EA. 2008. Short sleep duration in middle childhood: risk factors and
consequences. Sleep. 31:71–78.
Ogińska H. 2011. Can you feel the rhythm? A short questionnaire to describe two dimensions of
chronotype. Pers Individ Dif. 50:1039–1043.
Biological Rhythm Research 247
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015
Pallesen S, Nordhus IH, Nielsen GH, Havik OE, Kvale G, Johnsen BH, Skjotskift S. 2001.
Prevalence of insomnia in the adult Norwegian population. Sleep. 24:771–779.
Park DH, Kripke DF, Cole RJ. 2007. More prominent reactivity in mood than activity and sleep
induced by differential light exposure due to seasonal and local differences. Chronobiol Int.
24:905–920.
Putilov AA, Donskaya OG, Verevkin EG, Putilov DA. 2010. Associations of waking EEG
structure with chronotype and trototype of 130 sleep deprived individuals. Biol Rhythm Res.
41:113–136.
Putilov AA, Verevkin EG, Ivanova E, Donskaya OG, Putilov DA. 2008. Gender differences in
morning and evening lateness. Biol Rhythm Res. 39:335–348.
Roenneberg T, Kuehnle T, Juda M, Kantermann T, Allebrandt K, Gordijn M, Merrow M. 2007.
Epidemiology of the human circadian clock. Sleep Med Rev. 11:429–438.
Roenneberg T, Wirz-Justice A, Merrow M. 2003. Life between clocks: daily temporal patterns of
human chronotypes. J Biol Rhythms. 18:80–90.
Ryff CD. 1989. Happiness is everything, or is it? Explorations on the meaning of psychological
well-being. J Pers Soc Psychol. 57:1069–1081.
Schimmack U, Grob A. 2000. Dimensional models of core affect: a quantitative comparison by
means of structural equation modeling. Eur J Personality. 14:325–345.
Thorleifsdottir B, Björnsson JK, Benediktsdottir B, Gislason T, Kristbjarnarson H. 2002. Sleep
and sleep habits from childhood to young adulthood over a 10-year period. J Psychosom Res.
53:529–537.
Van Dongen HPA, Maislin G, Mullington JM, Dinges DF. 2003. 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.
Wahistrom K. 2002. Changing times: findings from the first longitudinal study of later high school
start times. NASSP Bull. 86:3–21.
Wittmann M, Dinich J, Merrow M, Roenneberg T. 2005. Social jetlag: misalignment of biological
and social time. Chronobiol Int. 23:497–509.
Zavada A, Gordijn MCM, Beersma DGM, Daan S, Roenneberg T. 2005. Comparison of the
munich chronotype questionnaire with the Horne–Östberg’s morningness–eveningness score.
Chronobiol Int. 22:267–278.
248 K.S. Jankowski
Downloaded by [Konrad S. Jankowski] at 12:00 12 February 2015