Daily Emotional States As Reported by Childrens and Adolescents.

Daily Emotional States as Reported by Children and Adolescents
Author(s): Reed Larson and Claudia Lampman-Petraitis

Source: Child Development, Vol. 60, No. 5 (Oct., 1989), pp. 1250-1260
Published by: Wiley on behalf of the Society for Research in Child Development
Stable URL: http://www.jstor.org/stable/1130798 .
Accessed: 07/04/2014 07:13
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Wiley and Society for Research in Child Development are collaborating with JSTOR to digitize, preserve and
extend access to Child Development.
http://www.jstor.org
This content downloaded from 155.54.213.13 on Mon, 7 Apr 2014 07:13:21 AM

All use subject to JSTOR Terms and Conditions
Daily Emotional States as Reported by Children

and Adolescents
Reed Larson
University of Illinois at Urbana-Champaignand Michael Reese Hospital
Claudia Lampman-Petraitis
Loyola University of Chicago
LARSON, REED, and LAMPMAN-PETRAITIS,CLAUDIA.Daily Emotional States as Reported by Children and Adolescents. CHILDDEVELOPMENT,
1989, 60, 1250-1260. Hour-to-houremotional states
reportedby children, ages 9-15, were examined in orderto evaluatethe hypothesis thatthe onset of
adolescence is associated with increased emotional variability.These youths carried electronic
pagers for 1 week and filled out reportson their emotional states in response to signals received at
randomtimes. To evaluate possible age-relatedresponse sets, a subset of children was asked to use
the same scales to rate the emotions shown in drawingsof 6 faces. The expected relationbetween
daily emotional variabilityand age was not found among the boys and was small among the girls.
There was, however, a linear relationbetween age and averagemood states,with the older participants reportingmore dysphoricaverage states, especially more mildly negative states. An absence
of age difference in the ratingsof the faces indicatedthat this relationcould not be attributedto age
differences in response set. Thus, these findings provide little supportfor the hypothesis that the
onset of adolescence is associatedwith increased emotionalitybut indicate significantalterationsin
everyday experience associated with this age period.
Adolescence has long been characterized

as a time of increased emotionality. Aristotle
described youth as "passionate, irascible, and
apt to be carried away by their impulses"
(cited by Fox, 1977). G. Stanley Hall wrote
that it is adolescents' "natural impulse to experience hot and perfervid states" (1904, Vol.
2, pp. 74-75). Similar descriptions of adolescent moodiness or emotionality have been
formulated by social psychologists (Becker,
1964; Lewin, 1938), anthropologists (Benedict, 1938; Mead, 1928), psychoanalysts (Blos,
1961; Freud, 1937), novelists (Spacks, 1981),
and professionals concerned with junior and
senior high school education (Eichhorn,
1980). Laypersons interviewed by Hess and
Goldblatt (1957) and Musgrove (1963) also
used such words as "impulsive," "unstable,"
and "wild" to describe the "average teenager." The underlying hypothesis is that adolescents, perhaps because of the physiological, psychological, and social changes of this
age period, experience emotional states that
are more extreme, change more quickly, and
are less predictable than those experienced at
earlier or later periods.
To evaluate the hypothesis that adolescents are more emotionally variable than
preadolescents, we examined time-sampling
reports obtained from 9-15-year-olds concerning the emotional states they experienced
during their daily lives. The vantage point for
this investigation was phenomenological; we
were concerned with the hour-to-hour range
of emotional states children and adolescents
attribute to themselves-with
their "conscious states" (Harris & Olthof, 1982)acknowledging that a different pattern of
findings might emerge were it possible to
measure components of emotion that a person
does not consciously detect, such as physiological arousal or unconscious affect. Emotional variability was defined for this study as
the standard deviation of a person's daily
mood states as assessed on a continuum from
positive to negative states. The research question is, then, do adolescents experience a
wider range of emotional states than preadolescents? In their daily lives do they attribute
to themselves the greater variation in states
that have long been attributed to them by observers? The study also addresses the related
This research was supported by NIMH grant 1 R01 MH38324, "Stressin Daily Life during
Early Adolescence," awardedto Reed Larsonand carriedout throughMichael Reese Hospital and
Medical Center. Send requests for reprintsto the first authorat the Division of Human Development and Family Ecology, University of Illinois at Urbana-Champaign,1105 West Nevada Street,
Urbana, IL 61801.
[Child Development, 1989, 60, 1250-1260. ? 1989 by the Society for Researchin Child Development, Inc.
All rightsreserved.0009-3920/89/6005-0008$01.00]

Larson and Lampman-Petraitis

question of whether they experience
emotional extremes.
more
BackgroundResearch
One must separate the hypothesis of adolescent emotionality from the broader hypothesis that adolescence is a time of turmoil or
"storm and stress," a view that no longer enjoys general acceptance (Adelsen, 1979; Petersen, 1988; Strober, 1986). The turmoil hypothesis asserts that emotionality is part of
a normal and even desirable adolescent condition of personality disruption and change
(Blos, 1961; Freud, 1937). Research findings,
however, make it clear that such disruption is
neither normative nor desirable. The great
majority of adolescents report feeling globally
happy with their lives (Offer, Ostrov, & Howard, 1981), rates of psychiatric disorder during
adolescence are not greater than at later periods of the life span (Rutter, Graham, Chadwick, & Yule, 1976), and the typical teenager
does not experience a breach in relations with
parents (Hill & Holmbeck, 1986; Montemayor, 1986). Furthermore, when turmoil
does occur, it is associated with worse, not
better long-term adjustment (Dusek & Flaherty, 1981; Kohlberg, LaCrosse, & Ricks,
1974; Offer & Offer, 1975).
The absence of a general condition of turmoil among a majority of teenagers, however,
does not disprove the more specific hypothesis that adolescents are emotional creatures.
The possibility remains that wide emotional
extreme positive as well
swings-including
as negative emotional states (Sharp, 1980)are a normative part of the adolescent experience.
Research provides some indication that
adolescents experience stronger emotions in
their daily lives than do adults. Direct data on
daily emotional experience were obtained by
Larson, Csikszentmihalyi, and Graef (1980),
who used electronic pagers to signal adolescents and adults to report on their subjective
experiences at random times during the day.
They found that the mood states reported by a
sample of high school students had higher
standard deviations than those reported by a
sample of adults. The mean mood levels reported by the two groups were similar, indicating that their emotional states had similar
baselines; the adolescents, however, reported
more occasions both of negative and positive
extremes.
Additional evidence comes from onetime questionnaire studies showing that
youth and young adults report greater emotional extremes than older adults. Sixteen- to
1251
19-year-olds studied by Diener, Sandvik, and

Larsen (1985) obtained higher scores on "affective intensity" than older members of the
same families. Bradburn (1969) and Campbell
(1981) found more frequent reporting of both
positive and negative affect among young
adult survey respondents than among successive groups of older respondents. A set of projective studies reviewed by Malatesta (1981)
also suggests a reduction in affective extremes
from youth to old age, although the author
suggested methodological reasons for avoiding this conclusion and in a questionnaire
study such an age trend was not evidenced
(Malatesta & Kalnok, 1984).
These studies provide initial confirmation for the hypothesis that adolescence is a
period of emotional variability. Their findings, however, are subject to two major limitations. First, while distinguishing the daily
pattern of adolescent emotional states from
that of adults, adolescents' states were not
studied in relation to those of children. It is
possible that emotionality in adolescence is
merely a continuation of childhood lability.
Indeed, infants and children also have a reputation for being emotionally variable (Malmquist, 1975; Weiner & Graham, 1984).
A second problem lies in the possibility
that the results reflect age changes in how
people use self-report mood scales. Research
on response styles indicates that, with age,
children and adolescents use less extreme
scale points in rating ambiguous stimuli.
Light, Zax, and Gardiner (1965), for example,
found less use of response extremes in ratings
of neutral stimuli from the fourth to eighth to
twelfth grades (see also Hamilton, 1965).
Thus, it is conceivable that differences found
between adolescents and adults may reflect
nothing more than a change in response style.
In other words, adults may experience the
same emotional states as adolescents, butowing to their greater base of experiencereport these states as less extreme.
Both of these problems can be resolved
by comparing the emotional variability of adolescents with children. If emotionality is a
distinguishing trait of adolescents-not just a
continuation of a childhood condition-then
adolescents should report wider daily emotional variations than children. But if age
differences in response style influence selfreport data, then children, rather than adolescents, should report greater emotional extremes.
In this study, therefore, we compared the
range of emotional states experienced by chil-

1252
Child Development
dren and adolescents. The onset of adolescence in the United States, where the study
was done, is typically associated with pubescence, the start of junior high school, and
other transitions that occur between the sixth
and ninth grades (Lipsitz, 1977; Steinberg,
1985). In order to obtain an adequate representation of preadolescents, we included fifth
graders; hence our sample includes students
from the fifth to the ninth grades (ages 9 to
15). The emotionality hypothesis predicts an
increase in the statistical variance of hour-tohour mood states across this age period. We
also examined age differences in average
states and the frequency of extreme states.
Following the procedure used by Larson
et al. (1980) these students rated their moods
at random times during the day in response to
signals from a pager. In order to discriminate
findings from trends that might be attributed
to response style, we also asked a subsample
of the students to rate the emotional states
they perceived in six simple drawings of faces
representing a range of emotions. Given the
quantity of evidence showing sex differences
during this age period, all analyses were performed separately for girls and boys.
Method
Subjects.-Participants in the study were
473 randomly selected children and young
adolescents in the fifth to ninth grades (ages 9
to 15) from four suburban neighborhoods in
the Midwestern United States. Two neighborhoods were generally lower middle class,
the third was middle class, and the fourth was
upper middle class. These communities were
composed almost exclusively of people with
European ancestry. To diversify possible
time-of-year effects and make the data collection manageable, the participation of the students was spread across eight waves of data
collection over 2 years. Sample selection was
stratified to obtain balanced representation by
sex, community, time of year, and grade, with
approximately equal numbers of students representing each quarter from the autumn of
fifth grade to the winter of ninth grade.
The final sample includes 68.8% of the
randomly selected students initially invited to
take part. Twenty-four percent of those invited declined to take part or failed to obtain
permission from their parents, 4.4% began the
study but failed to complete the requisite
number of self-report forms, and 2.8% completed the study but provided data that were
unusable. Sample loss was approximately
equal among boys and girls (32.2% vs. 30.2%,
respectively) and was more frequent in the
higher grades for boys (fifth to ninth: 24%,

23%, 33%, 42%, 37%; tau b = .13, p < .005)
but not for girls (tau b = -.01, N.S.). A survey of the entire school population in two of
the six schools (Larson, 1989) indicated that
the students who declined participation did
not differ in social class, as reflected by Hollingshead ratings of their parents' occupations
(Hollingshead & Redlich, 1958), or selfesteem, as measured by a six-item version of
the Rosenberg scale (Rosenberg, 1965). They
were more likely than participants to live
with a stepparent; however, separate analyses
showed little relation between parents' marital status and the properties of emotional
states considered here. Sociometric ratings
obtained in one school indicated no significant relation between nonparticipation and an
individual's popularity among peers. This
general absence of differences between participants and nonparticipants, particularly in
self-esteem, suggests that the higher sample
loss among the older boys does not introduce
sampling bias.
Separate analyses showed that the small
fraction of students who participated in the
study but were not included in the final sample (6.2% of those initially invited) were rated
as significantly less mature by their teachers,
received lower grades in school, and came
from families with significantly lower SES
characteristics. They also rated themselves as
slightly more depressed, with marginal significance, on a standard depression inventory
(Larson, 1989). However, since these 39 individuals were distributed across age and sex
categories, their exclusion does not create
bias in the final sample.
Procedures.-Following the procedures
of the Experience Sampling Method (ESM)

(Csikszentmihalyi & Larson, 1987; Hormuth,
1986; Larson & Csikszentmihalyi, 1983), participants carried electronic pagers for 1 week
along with a booklet of self-report forms.
Seven signals were sent to the pagers each
day between 7:30 A.M.and 9:30 P.M.,with one
signal at a random time within each 2-hour
block. The students' instructions were to
carry the pager with them at all times and to
fill out one self-report form immediately after
each signal was received. The pager could be
turned off if they took a nap, went to bed before 9:30 P.M.,or planned to sleep late. Otherwise the stated goal was to obtain descriptions of their experience at each of the
designated times.
The students together responded to a total of 17,752 of the ESM signals by filling out
self-report forms, an average of 37.5 per per-


son. After discounting signals that were
missed due to sleep and mechanical failure of
the pager, this represents a response to 84% of
the signals that were sent. Information obtained from the debriefing interview suggests
that the missed signals occurred during a
wide range of activities and do not represent
one particular category of experiences (Larson, 1989). The final sample of self-reports
thus provides a close-to-representative sample of their daily lives.
Instruments.-Students
rated their emotional states or moods at the time of each signal using six 7-point semantic-differential
scales. These items dealt with two basic dimensions of emotional experience that have
been identified and found to be stable for
children this age (Russell & Ridgeway, 1983).
Three of these items dealt with the dimension of pleasure-displeasure or affect (happyunhappy, cheerful-irritable, friendly-angry);
three dealt with the dimension of arousal
(alert-drowsy, strong-weak, excited-bored).
Values for these items are reported below on
a metric from - 3 to + 3, with negative values
corresponding to more dysphoric states, and
positive values corresponding to positive
states.
The analyses in this paper are based on
aggregated scores, computed for all of the
self-reports provided by each person. Hence
our unit of analysis is the person (N = 473),
not the individual time sample (N = 17,752).
The primary scores that were examined are
the standard deviation of each person's responses to each item-which serve as indices
of emotional variability-and
the means of
each person's responses to these itemswhich serve as indices of baseline emotional
state. The correlation matrix for these scores,
presented in Table 1, shows correlations
among the means ranging from r = .89 to .31
and correlations among the standard deviations ranging from r = .80 to .37. The correlations between the means and standard deviations range from r = .26 to -.30.
To evaluate the reliability of these scores,
the sequence of self-reports for each individual was divided at the middle point in the
week and the scores computed separately for
the first half of the person's week and the second half. The correlations between these
scores ranged from r = .57 to .46 for the standard deviations of the six items, and r = .73 to
.66 for the means (Larson, 1989). Given that
emotional states vary as a function of the ongoing events in people's lives, these values
are within an expected level of intra-subject
consistency. Findings demonstrating the va-
1253
lidity of these mood measures are presented

in an article by Csikszentmihalyi and Larson
(1987).
Evaluation of response style.-In order
to assess age differences in the tendency to
respond at the extremes on these scales, a
subset of 106 fifth to eighth graders from the
final two waves of the study were asked at the
end of the week of paging to use the same set
of semantic differential items to rate each of
six simple line drawings of faces. Ekman and
Friesen's (1975) guide to emotions in the human face was used to select features for these
faces such that they would provide a balanced
representation of the emotional dimensions of
affect and arousal. Two of these faces were
meant to be neutral, involving "blends" of
features; the others were intended to represent the four combinations of positive and
negative affect and high and low arousal. The
faces were ambiguous with regards to age and
sex. The students were asked, "How would
these people rate themselves, if they were
beeped?"
Each student's ratings for these six faces
were aggregated, similar to the aggregations
carried out with the ESM self-reports. Correlations between the self-report scores and the
ratings of faces showed little consistent pattern among the girls, with substantial correlations for the item strong-weak (r = .41 for the
standard deviations, r = .36 for the means)
and correlations for the other items ranging
between r = .26 and -.23. Among the boys,
correlations for the standard deviation scores
were positive, ranging from r = .42 to .29, and
for mean scores, slightly negative, ranging
from r = -.27 to .09.
Results
The analyses involved a series of parallel
multivariate analyses of variance with grade
as the independent variable and the self-reports and ratings of faces as the dependent
variables. All of the dependent variables
were subjected to homogeneity-of-variance
tests prior to the main analyses, which indicated that MANOVA would be an acceptable
method of handling these data. Each significant MANOVA was followed up with univariate tests for polynomial trends. Finally, all
analyses were performed separately for girls
and boys.
To test differences in emotional variability between age groups, we performed a
MANOVA with grade as the independent
variable and students' standard deviations on
each of the six self-ratings scales as the de-

""
CO)
in
0
cqcl
cqlCO)
c qm n
0
Cq
,v
vOQ
p
"00Po4
)-,C)0CO
C0
CO)
cd
00m CO)m
0C)~O CO)
>
W0
C ..,
...
C)) ? r-00
C)ItN00in?lcoCO)
0 m0o
0-0
14
z
cj
.I
-4CIO
UQ o~m~
m
ncI 0000i
..0
E-
>
>
ZZ
z~
cd
0n - 4-1~
0 li C)CO
00CO
bC
z CA
~4
a
T~l
L4
(I)
"I
"
k
It
CO)
0 nIT
C)
.. .C)
o . 0....
-4
.iQn
. ...0.It. oc
.cq
.C)
. N.. C.1Z
P..
.
"0"~0
1
O S.1
bo
C) )
ItfCO)f0C)
1
..
...c. .
"
..
>
IIC-1
-
0(
m"in
'I0-.?coC?-I
)co
t-0rco
?-I
?I'm
Cco
C)tP-4
u a D0 Ecd?I -I~0~I?00
in00inIt0) C C ;2
d>
aa o
"c
*..
P.
..
4.
CO)
cd
. . . . .D
.
U
1$
P.
cd
r
rA
A) f*
bCbO
t. PO
m
~?m
a)

1255
TABLE 2
AGE TRENDS IN THE STANDARD DEVIATION OF REPORTED DAILY EMOTIONALSTATES
GRADE
Girls:
Happy ..............
Cheerful ............
Friendly ............
Alert ................
Strong ..............
Excited .............
N ...................
Boys:
Happy ..............
Cheerful ............
Friendly ............
Alert ................
Strong ..............
Excited .............
N ...................
NOTE.-The
t FOR
LINEAR
TREND
CORRELATION
WITH
GRADE
1.21
1.08
1.12
1.44
1.05
1.46
49
1.18
1.20
1.15
1.44
1.12
1.49
54
1.27
1.27
1.22
1.48
.98
1.58
52
1.25
1.22
1.27
1.52
1.19
1.46
48
1.36
1.21
1.23
1.50
1.12
1.40
37
1.70
2.27
1.68
.85
.81
- .65
.09
.03
.01
.40
.42
.52
.11
.15
.17
- .06
.05
- .03
1.24
1.27
1.24
1.37
1.12
1.44
49
1.18
1.17
1.15
1.43
1.11
1.40
54
1.28
1.25
1.24
1.56
1.17
1.49
52
1.11
1.21
1.16
1.50
1.06
1.42
48
1.12
1.12
.99
1.42
1.00
1.27
37
-1.48
- 1.06
-1.94
.62
-1.09
- 1.23
.14
.29
.05
.53
.28
.22
-.10
-.06
-.06
.05
-.07
-.07
body of the table shows mean standard deviations for each grade level.
pendent variables. This analysis showed an

association between grade level and the standard deviation scores for girls that was just
beyond significance, F(24,914) = 1.49, p =
.06, but no association for boys, F(24,886) =
.93, N.S. Univariate analyses revealed weak,
but significant linear trends for two scales
dealing with affect (cheerful, friendly). Table
2 shows the mean values for each grade level
and correlation coefficients indicating the
magnitude of these linear trends. Since the
quadratic and higher order polynomial terms
were not significant in these (and subsequent)
analyses, they are not shown.
We also performed multivariate analyses
using the standard deviations of the ratings of
faces as the dependent variables. There was
no relation between grade and these scores
among the girls, F(18,173) = .33, N.S., or for
the boys, F(18,155) = .32, N.S. Given that students at the different grade levels attributed
the same range of emotional states to these
standard faces, the age differences among
girls in the variance of self-reported daily
emotions cannot be explained as a function of
response style but rather appear to reflect true
differences in the affective states the older
girls experienced in their daily lives.
Significant age trends in average daily
emotional states were reported by both girls
and boys (see Table 3). MANOVAs were performed using the means of each person's daily
self-reports on the six rating scales as the dependent variables. These analyses revealed a
significant relation between grade level and

average daily mood states for both girls,
F(24,914) = 1.90, p < .01, and boys, F(24,886)
= 1.86, p < .01. Univariate analyses showed
significant linear relations with grade level for
five of the six variables both for girls and boys
(Table 3).
The ratings of the faces again indicated
that the age differences cannot be explained
in terms of variations in response style, at
least for most of these variables. For boys,
there was no significant association between
grade level and the average states attributed
to the drawings of the faces, F(18,155) = 1.18,
N.S. For girls, there was a significant association between grade level and the average ratings of the faces, F(18,173) = 1.69, p < .05.
Polynomial analyses, however, revealed no
univariate trends significant at the .05 level,
and only two of the six items suggested a linear relation with grade that might confound
interpretation of the trends for the self-ratings.
The correlation of grade with the mean happiness girls attributed to the faces was r =
-.15 (p = .23), and the correlation of grade
with the mean perceived cheerfulness of the
faces was r = -.23 (p = .052). Correlations
for the other four items were between r =
-.07 and .21. Therefore, with the possible
exception of the trends for happiness and
cheerfulness among the girls, the age trends
in self-reported daily moods cannot be explained as a difference in response style, but
rather indicate that the older participants experienced less positive average daily states.

1256
Child Development
TABLE 3
AGE TRENDS IN THE MEANS OF REPORTED DAILY EMOTIONALSTATES
Girls:
Happy ...........
Cheerful ..........
Friendly .........
Alert .............
Strong ...........
Excited ..........
Boys:
Happy ...........
Cheerful .........
Friendly .........
Alert .............
Strong ...........
Excited ..........
35
CORRELATION
WITH
GRADE
t FOR
LINEAR
TREND
5.61
5.34
5.60
4.45
4.46
4.77
5.43
5.25
5.41
4.58
4.68
4.64
5.10
4.92
4.98
4.24
4.28
4.25
5.14
5.11
5.12
4.31
4.34
4.44
4.98
4.83
5.11
4.48
4.42
4.42
-4.22
- 2.64
-3.19
- .42
- 2.07
- 2.07
.001
.009
.002
.68
.04
.04
- .28
-.17
- .21
- .04
-.14
-.14
5.27
4.94
5.26
4.67
4.92
4.53
5.18
4.90
4.93
4.77
4.99
4.40
4.96
5.01
4.88
4.47
4.89
4.38
4.86
4.60
4.78
4.26
4.52
4.26
4.90
4.46
4.75
4.12
4.58'
4.44
-2.81
- 3.03
- 3.04
- 3.30
- 2.62
- 1.08
.005
.003
.003
.001
.009
.48
-.19
-.19
- .20
- .21
-.17
- .06
GRADE
Percentage of Self-Reports
35
30
30
25
25
20
20
15
16
10
10
6
0
-3
-2
Negative
Neutral
+3
+2
+1
-1
Positive
Grade
5th
6th
7th
8th
9th
of self-reported states by grade: girls. (Note: Table shows the mean frequency
FIG. 1.-Frequency
with which students used each of the gradations on the scales to identify their daily emotional states.)
As a supplementary means of examining

these age trends, we evaluated the frequency
with which students at each grade level reported different gradations of positive and
negative mood. For each individual, the percentage of time was calculated that he or she
used each of the 7 points on the six scales to
identify his or her state. Figures 1 and 2 show
the means of these percentages at each grade
level.
It is apparent from this graph that the
greatest age differences occurred for the extreme positive and the mildly negative scale
points. For both girls and boys, the most positive scale point (+3) was used to describe
their states half as often in ninth grade as in
fifth grade. For girls, the linear correlation between grade and percentage in this category
was r = -.21 (p < .001) and for boys was r =
-.22 (p < .001). In place of these extreme
positive states, the older girls reported more
frequent experience of both mildly negative
and mildly positive states, as indicated by
significant linear correlations between frequency and grade level for the scale points
marked +1 (r = .20, p < .01), -1 (r = .21,
p < .001) and -2 (r = .14, p < .05). Older


35
Percentage of Self-Reports
35
30
30
25
25
20
20
15
15
10
10
5
5
0
1257
0
-3
-2
-1
Negative
+1
Neutral
+2
+3
Positive
FIG.2.-Frequency of self-reportedstates by grade:boys

boys reported more frequent experience only
at the mildly negative - 1 scale point (r = .22,
p < .001). There was no significant age difference in the frequency of extreme negative
states. Contrary to the emotionality hypothesis, the adolescents did not experience more
frequent extreme states but rather experienced more states in this middle emotional
range.
Again, these differences were not replicated in the ratings of faces by the subsample
of 106 students. For boys, the intermediate
2) were
points on the scales (-2,-1,+1,+
used 42.3% of the time by the fifth and sixth
graders and 41.7% by the seventh and eighth
graders; for the girls these intermediate
points of the scales were used for 35.5% of the
time by the younger group and 39.7% by the
older group. There also were no substantive
differences between age groups in the frequency with which positive or negative intermediate scale points were used to rate the
faces. The age differences in self-reported
daily states, therefore, do not reflect an age
difference in response style, but, rather, appear to represent a difference in daily emotional experience.
Discussion
These findings suggest that the onset of
adolescence is not associated with appreciable differences in the variability of emotional
states experienced during daily life. Partici-
pants in the study reported on their moods

when signaled at random times over a week.
Among boys there were no significant trends
across the age period from 9 to 15 in the
intraindividual variability of self-reported
moods. Among girls there was a slight and
only marginally significant age difference in
these standard deviations, accounted for by
positive linear age trends for two mood items
that dealt with affect. Neither boys nor girls
showed an increase in the frequency of extreme negative or positive states.1 Such
findings challenge the hypothesis that adolescence is a time of greatly increased emotionality. At the same time, the data do not suggest an age-related decrease in emotionality;
they do not show the beginnings of the decline in emotional variability that some studies have found between adolescence and
adulthood. Adolescence, then, may be a plateau period during which the emotional lability of childhood is sustained and is manifested within the more sophisticated and
adult-like prism of teenage experience.
While these findings indicated little age
difference in the variability of moods, they do
suggest that the onset of adolescence is associated with changes in average mood. Signaled during their daily lives, there were
fewer occasions when older participants reported extreme positive states and more occasions when they reported mildly negative
ones. For both boys and girls, the average
state-the
emotional baseline-was
lower
1 In separateanalyses we have examined these data using pubertalstatusratherthan gradeas

the independent variable (Richards& Larson,in process).Since many theories explicitly relatethe
emotionalityof adolescence to the biological events of puberty,one might expect pubertalstatusto
show strongerrelations with the dependent variablesconsidered here. The findings of these analyses, however, indicate that pubertalstatus is no strongera predictorof these mood variablesthan
grade.

1258
Child Development
among older than younger subjects. The absence of comparable age differences in ratings
of emotions in drawings of faces indicates that
these findings cannot be attributed to age differences in response bias.
Two explanations can be considered for
these differences between children's and adolescents' representations of their daily affective states. One explanation is that there are
differences in how younger and older youth
interpret what may be very similar internal
and external daily affective experiences. A
second explanation is that there are real
changes in these internal and external experiences.
The interpretive explanation posits an
adolescent who has become more critical and
discerning in reading internal and external
emotional cues, who may be less willing to
label his or her experience with naively positive superlatives, and who is more able to
detect cues indicative of mildly negative
states. Laboratory research indeed demonstrates that, with increasing age, children's
emotional understanding deepens (Nannis,
1988a, 1988b): their lexicon of emotional
terms expands, they make more complex differentiations of the emotions appropriate to
different situations (Harter, 1980; Schwartz &
Trabasso, 1984; Weiner & Graham, 1984), and
they more frequently consider internal processes in their attributions of emotion (Harris,
Olthof, & Terwogt, 1981; Wolman, Lewis, &
King, 1971). It is conceivable, then, that the
less frequent extreme positive ratings among
the adolescents and the slightly increased
variability with age among the girls reflects an
ability to make finer discriminations of emotional experience. This laboratory research,
however, does not easily account for the overall downward shift in average emotional
states.
The second explanation attributes the
age differences in reported states to real
changes in the affective composition of daily
life across the transition to adolescence. That
is, the downward shift in average states may
reflect an alteration in the balance sheet of
positive and negative affective cues that make
up the individual's daily experience. Such an
alteration might be related to a host of normative changes associated with early adolescence: increasing stressful events (BrooksGunn & Warren, 1989; Coddington, 1972),
the hormonal changes of pubescence (Nottelmann et al., 1987), complex interactions occurring with pubertal development (BrooksGunn & Warren, 1989; Petersen & Taylor,
1980; Simmons, Blyth, Van Cleave, & Bush,
1979), changes in the social environment

(Lewin, 1938), or increased emotional autonomy and decreased parental ego support
(Blos, 1961; Steinberg & Silverberg, 1986).
This explanation is also consistent with the
finding that rates of depressive feelings and
depressive disorders increase with the onset
of adolescence (Rutter, 1986).
The two explanations actually converge
in what they suggest about the daily lives of
children and adolescents. Whether one concludes that there is a real change in the affective composition of their lives or that adolescents just think there is, the outcome is the
same-the adolescent's conscious experience
includes many fewer occasions when the individual feels on top of the world and more
occasions of feeling mildly negative. While
the majority of the daily states they experience are still positive, the overall range of
their experience is markedly lower.
Further research will be required to investigate the causes and consequences of
these alterations in daily experiences. Are
they related to the normative changes associated with early adolescence: increased life
stress, hormonal changes, autonomy from parents? What is the association of these alterations in experience with subsequent development of adolescent problems, particularly
with the high rates of depression found
among girls beginning in mid-adolescence?
Lastly, it is important to consider how these
changes are related to, and might be modified
by, developmental changes in the understanding of emotions and the individual's acquisition of strategies for controlling and regulating them.
References
Adelson, J. (1979, February).Adolescence and the
generalizationgap. Psychology Today, 12, 3337.
Becker, H. S. (1964). Personal change in adult life.
Sociometry,27, 40-53.
Benedict, R. (1938). Continuities and discontinuities in cultural conditioning. Psychiatry, 1,
161-167.
Blos, P. (1961). On adolescence. New York:Free
Press.
Bradburn,N. (1969). The structureof psychological
well-being. Chicago: Aldine.
Brooks-Gunn,J., & Warren,M. P. (1989).Biological
and social contributions to negative affect in
young adolescent girls. Child Development,
60, 40-55.
Campbell, A. (1981). The sense of well-being in
America: Recent patterns and trends. New
York:McGraw-Hill.

1259
Coddington, R. D. (1972). The significance of life

events as etiologic factors in the diseases of
children: II. A study of a normal population.
Kohlberg, L., LaCrosse, J., & Ricks, D. (1974). The

predictability of adult mental health from
childhood behavior. In B. Wolman (Ed.), Man-
Journal of Psychosomatic Research, 16, 205-
ual of child psychopathology(pp. 1217-1234).
213.
Csikszentmihalyi, M., & Larson, R. (1987). Validity
and reliability of the experience-sampling
New York: McGraw-Hill.

Larson, R. (1989). Beeping children and adolescents: A method for studying time use and
daily experience. Journal of Youth and Adolescence, 18(4), in press.
Larson, R., & Csikszentmihalyi, M. (1983). The experience sampling method. In H. T. Reis (Ed.),
method. Journal of Nervous and Mental Disease, 175, 526-536.

Diener, E., Sandvik, E., & Larsen, R. (1985). Age
and sex effects for emotional intensity. Devel-
opmental Psychology,21, 542-546.

Dusek, J., & Flaherty, J. (1981). The development
of the self-concept during the adolescent years.
Monographs of the Society for Research in

Child Development, 46(4, Serial No. 191).
Eichhom, D. (1980). The school. In M. Johnson
(Ed.), Toward adolescence: The middle school,

years (pp. 56-73). Chicago: University of Chicago Press.
Ekman, P., & Friesen, W. V. (1975). Unmasking the
face. Englewood Cliffs, NJ: Prentice-Hall.
Fox, V. (1977). Is adolescence a phenomenon of
modem times? Journal of Psychohistory, 1,

271-290.
Freud, A. (1937). The ego and the mechanisms of

defense. New York: International Universities
Press.
Hall, G. S. (1904).Adolescence: Its psychology and

its relations to physiology, anthropology, sociology, sex, crime, religion, and education.
New York: Appleton.
Hamilton, D. L. (1965). Personality attributes associated with extreme response style. Psychologi-
cal Bulletin, 69, 192-203.

Harris, P. L., & Olthof, T. (1982). The child's concept of emotion. In G. Butterworth & P. Light
(Eds.), Social cognition: Studies in the development of understanding (pp. 188-209). Chicago: University of Chicago Press.
Harris, P. L., Olthof, T., & Terwogt, M. M. (1981).
Children's knowledge of emotion. Journal of
Child Psychology,22, 247-261.

Harter, S. (1980). A cognitive-developmental approach to children's understanding of affect
and trait labels. In R. Serafica (Ed.), Social-cog-
nitive development in context (pp. 27-61).

New York: Guilford.
Hess, R., & Goldblatt, I. (1957). The status of adolescence in American society: A problem in
social identity. Child Development, 28, 459468.
Hill, J. P., & Holmbeck, G. N. (1986). Attachment
and autonomy during adolescence. In G. J.
Whitehurst(Ed.),Annals of child development

(Vol. 3, pp. 145-189). Greenwich, CT: JAI.
Hollingshead, A., & Redlich, F. (1958). Social class
and mental illness: A community study. New

York: Wiley.
Hormuth, S. E. (1986). The sampling of experience
in situ. Journal of Personality,54, 262-293.
Naturalistic approaches to studying social interaction, new directions for methodology of

social and behavioral science (pp. 41-56). San
Francisco: Jossey-Bass.
Larson, R., Csikszentmihalyi, M., & Graef, R.
(1980). Mood variability and the psychosocial
adjustment of adolescents. Journal of Youth
and Adolescence, 9, 469-490.
Lewin, K. (1938). Field theory and experiment in

social psychology: Concepts and methods.
AmericanJournal of Sociology, 868-896.

Light, C. S., Zax, M., & Gardiner, D. H. (1965).
Relationship of age, sex, and intelligence level
to extreme response style. Journal of Personal-
ity and Social Psychology,2, 907-909.

Lipsitz, J. (1977). Growing up forgotten. London:
Transaction Books.
Malatesta, C. Z. (1981). Affective development over
the lifespan: Involution or growth? Merrill-
Palmer Quarterly,27, 145-173.

Malatesta, C. Z., & Kalnok, M. (1984). Emotional
experience in younger and older adults. Jour-
nal of Gerontology,39, 301-308.

Malmquist, C. P. (1975). Depressions in childhood
and adolescence. New England Journal of
Medicine, 284, 887-893.
Mead, M. (1928). Coming of age in Samoa. New
York: Morrow.
Montemayor, R. (1986). Family variation in parentadolescent storm and stress. Journal of Adoles-
cent Research, 1, 15-31.
Musgrove, F. (1963). Intergenerational attitudes.
British Journal of Sociology and Clinical Psychology, 2, 209-223.

Nannis, E. D. (1988a). Cognitive-developmental
differences in emotional understanding. In
E. D. Nannis & P. A. Cowan (Eds.), New direc-
tions for child development: Developmental

psychopathology and its treatment (no. 39).
San Francisco: Jossey-Bass.
Nannis, E. D. (1988b). A cognitive-developmental
view of emotional understanding and its implications for child psychotherapy. In S. R. Shirk
(Ed.), Cognitive development and child psychotherapy. New York: Plenum.

Nottelmann, E. D., Susman, E. J., Blue, J. H., InoffGermain, G., Dorn, L. D., Loriaux, D. L., Cutler, G. B., & Chrousos, G. P. (1987). Gonadal
and adrenal hormone correlates of adjustment
in early adolescence. In R. M. Lemer & T. T.

1260
Child Development
Foch (Eds.), Biological-psychological interactions in early adolescence (pp. 303-323).

Hillsdale, NJ: Erlbaum.
Offer, D., & Offer,J. (1975). From teenage to manhood. New York:Basic.
Offer, D., Ostrov, E., & Howard, K. I. (1981). The
adolescent. New York:Basic.
Petersen, A. C. (1988). Adolescent development.
Annual Review of Psychology,39, 583-607.
Petersen, A. C., & Taylor,B. (1980). The biological
approach to adolescence: Biological change
and psychological adaptation. In J. Adelsen
(Ed.), Handbookof adolescent psychology (pp.
117-155). New York:Wiley-Interscience.
Rosenberg, M. (1965). Society and the adolescent
self-image. Princeton,NJ: PrincetonUniversity
Press.
Russell, J. A., & Ridgeway, D. (1983). Dimensions
underlying children's emotion concepts. Developmental Psychology, 19, 795-804.
Rutter,M. (1986). The developmental psychopathology of depression: Issues and perspectives.
In M. Rutter,C. E. Izard, & R. B. Read (Eds.),
Depression in young people (pp. 3-30). New
York:Guilford.
Rutter, M., Graham,P., Chadwick, O., & Yule, W.
(1976). Adolescent turmoil: Fact or fiction?
Journal of Child Psychology and Psychiatry,
17, 35-56.
Schwartz,R. M., & Trabasso,T. (1984). Children's
understandingof emotions. In C. E. Izard, J.
Kagan,& R. B. Zajonc(Eds.), Emotions, cogni-
tions, and behavior (pp. 409-437). Cambridge:

CambridgeUniversity Press.
Sharp,V. (1980). Adolescence. In J. Bemorad(Ed.),
Child development in normality and psychopathology (pp. 174-218). New York:Brunner/
Mazel.
Simmons,R. G., & Blyth, D. A. (1987).Movinginto
adolescence. New York:Aldine De Gruyter.
Simmons,R. G., Blyth, D. A., Van Cleave, E. F., &
Bush, D. M. (1979). Entry into early adolescence: The impactof school structure,puberty,
and early dating on self-esteem. American Sociological Review, 44, 948-967.
Spacks, P. M. (1981). The adolescent idea. New
York:Basic.
Steinberg, L. (1985). Adolescence. New York:
Knopf.
Steinberg,L., & Silverberg,S. B. (1986). The vicissitudes of autonomy in early adolescence.
Child Development, 57, 841-851.
Strober,M. (1986).Psychopathologyin adolescence
revisited. Clinical Psychology Review, 6, 199209.
Weiner, B., & Graham,S. (1984). An attributional
approachto emotional development. In C. E.
Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotions, cognition, and behavior (pp. 167-191).
Cambridge:CambridgeUniversity Press.
Wolman, R. N., Lewis, W. C., & King, M. (1971).
The development of the language of emotions:
Conditions of emotional arousal.Child Development, 42, 1288-1293.


Daily Emotional States As Reported by Childrens and Adolescents.

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Daily Emotional States As Reported by Childrens and Adolescents.

Hochgeladen von

Copyright:

Verfügbare Formate

Daily Emotional States as Reported by Children and Adolescents

Author(s): Reed Larson and Claudia Lampman-Petraitis

This content downloaded from 155.54.213.13 on Mon, 7 Apr 2014 07:13:21 AM