MIHALY CSIKSZENTMIHALYI and JEREMY HUNTER
HAPPINESS IN EVERYDAY LIFE: THE USES OF
(Received 29 January 2003; Accepted 17 February 2003)
ABSTRACT. This paper uses the Experience Sampling Method data drawn from a
national sample of American youth. It examines the proximal environmental factors
as well as behaviors and habits that correlate to personal happiness. Momentary-
level scores show that reported happiness varies signiﬁcantly both by day of week
and time of day. Furthermore, particular activities are associated with varying degrees
of happiness. School activities rate below average scores in happiness, while social,
active and passive leisure activities are above average. Particular companions also
correlate to differing level of happiness. Being alone rates the lowest levels of hap-
piness, while being with friend corresponds to the highest. Person-level averages of
happiness suggest that both higher social class and age correlate with lower levels
of happiness, while gender and race do not. Paradoxically, youth who spend more
time in school and social activities are happier than those who spend less. Unex-
pectedly, students who spend more time pleasure reading report lower levels of
happiness. Finally, feeling good about the self, excited, proud, sociable, active as
well as being in the conditions for ﬂow experience are the strongest predictors of trait
KEY WORDS: experience sampling, happiness, usual daily activities
Current understanding of human happiness points at ﬁve major effects
on this emotion. These are, moving from those most impervious to
change to those that are most under personal control: genetic determi-
nants, macro-social conditions, chance events, proximal environment
and personality. It is not unlikely that, as behavioral geneticists insist, a
“set level” coded in our chromosomes accounts for perhaps as much as
half of the variance in self-reported happiness (Lykken and Tellegen,
1996; Tellegen et al., 1988). These effects are probably mediated by
temperamental traits like extraversion, which are partly genetically
determined and which are in turn linked to happiness (Myers, 1993).
Cross-national comparisons suggest that macro-social conditions such
as extreme poverty, war and social injustice are all obstacles to hap-
piness (Inglehart and Klingemann, 2000; Veenhoven, 1995). Chance
events like personal tragedies, illness, or sudden strokes of good for-
tune may drastically affect the level of happiness, but apparently these
∗This study was made possible by a grant from the Alfred P. Sloan
Journal of Happiness Studies 4: 185–199, 2003.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
186 MIHALY CSIKSZENTMIHALYI AND JEREMY HUNTER
effects do not last long (Brickman et al., 1978; Diener, 2000). One
might include under the heading of the proximal environment the social
class, community, family and economic situation – in other words, those
factors in the immediate surroundings that may have an impact on a
person’s well-being. And ﬁnally, habits and coping behaviors devel-
oped by the individual will have an important effect. Hope, optimism
and the ability to experience ﬂow can be learned and thus moderate
one’s level of happiness (Csikszentmihalyi, 1997; Seligman, 2002).
In this paper, we present a method that allows investigators to study
the impact of momentary changes in the environment on people’s hap-
piness levels, as well as its more lasting, trait-like correlates. Research
on happiness generally considers this emotion to be a personal trait.
The overall happiness level of individuals is measured by a survey
or questionnaire, and then “happy” people – those who score higher
on a one-time response scale – are contrasted with less happy ones.
Whatever distinguishes the two groups is then assumed to be a condi-
tion affecting happiness. This perspective is a logical outcome of the
methods used, namely, one-time measures. If a person’s happiness level
is measured only once, it is by deﬁnition impossible to detect intra-
individual variations. Yet, we know quite well that emotional states,
including happiness, are quite volatile and responsive to environmental
Of course both common sense and psychological research suggests
that when positive events happen in a person’s life, happiness increases.
For instance Schwartz and Strack (1999) have shown that even such
trivial events as one’s home team winning a soccer match, or the infor-
mation that the weather in one’s hometown is better than the weather in
surrounding areas, will raise happiness levels. However, they warn that:
“...subjective well-being cannot be predicted on the basis of objective
circumstances, unless one takes the construal process into account”
(p. 61). In other words, the impact of external events on happiness is
mediated by the person’s system of values and cognitive interpretive
It is to detect variations in emotional states over time that the
Experience Sampling Method (ESM) was developed. This method
relies on subjects’ responses to an electronic pager that signals at ran-
dom times during the waking hours of the day, yielding up to ﬁfty
measures of happiness at speciﬁc moments during an average week.
Each time the pager signals, the respondents rate their experiential
states, including their levels of happiness (e.g. Csikszentmihalyi et al.,
HAPPINESS IN EVERYDAY LIFE 187
1977; Kubey et al., 1996; Csikszentmihalyi and Schneider, 2001;
a handbook for using the ESM is in preparation, see Hektner, in press).
This method not only accounts for momentary states, but can also yield
trait-like measure by adding up for each person the separate momentary
Daniel Kahneman (1999) has described this approach as measuring
“point-instant utility”, and argued for its theoretical importance: “An
assessment of a person’s objective happiness over a period of time can
be derived from a dense record of the quality of experience at each
point” (p. 3).
Thus repeated measures taken over a representative segment of a
person’s life can be used in two ways: (a) as indicators of momen-
tary happiness, which can help us understand the effect of Immediate
environmental circumstances; and (b) as personal traits derived from
aggregating the repeated responses over a week’s time, to derive a
trait-like measure of personal happiness.
The ﬁrst comparison of state-like and trait-like characteristics of sub-
jective experience using the ESM was a doctoral dissertation by Ronald
Graef (1978). In that work Graef found that while all the emotions were
more trait-determined than state-determined, this was particularly true
of happiness. In other words, a person’s average level of happiness
explained more of the variance in his or her responses over the week
than was explained by what that person was doing, where he or she
was, or whom he or she was with. This “set level” (cf. Tellegen et al.,
1988) explained about twice the variance in happiness compared to
other mood states. Longitudinal studies suggest a somewhat different
conclusion. In a 2-year follow-up of 455 high school students, the aver-
age ESM happiness scores correlated 0.55, more or less at the same
level as other mood variables. But a 4-year follow-up of a subset of
187 of these students showed only a correlation of 0.22 for happiness,
while r’s for all the other variables ranged from 0.34 (being in control)
to 0.56 (being relaxed), suggesting that self-reported happiness is less
stable than other dimensions of experience (Moneta et al., 2001; Patton,
1998; Hektner, in press).
In any case, there is obviously a great deal of variance unexplained by
a “set level” of happiness. In this paper we are going to use ESM data on
a group of over 800 adolescents to explicate the contributions of some of
the momentary conditions on intra-individual reports of happiness, and
then look from a trait-like perspective at how demographic variables
and patterns of behavior relate to over-all levels of happiness.
188 MIHALY CSIKSZENTMIHALYI AND JEREMY HUNTER
The participants of this study are primary school students from the
Alfred P. Sloan Study of Youth and Social Development, a national
multi-year study involving 6th, 8th, 10th and 12th graders from
33 elementary and secondary schools from 12 communities across the
country. These sites were chosen to create a nationally-representative
sample based on the variation in labor force composition and partic-
ipation, ethnicity, urbanicity, geographic location, and student ability
(see Csikszentmihalyi and Schneider, 2000 for a fuller description).
The 828 students included here are part of a focal group of 1215 youth.
The group here represents those who provided the minimum amount
of Experience Sampling Data and include 342 males (41.3%) and 486
females (58.7%), 491 Whites (59.3%), 54 Asians (6.5%), 131 Latinos
(15.8%), 145 African Americans (17.5%) and a small number (7) of
Native Americans (0.8%). Two-hundred and thirty-three, 6th graders
represented 28.1% of the sample, while the remainder were 236 Eighth
graders (28.5%), 196 Tenth graders (23.7%) and 163 Twelth graders
(19.7%). Social Class was measured on the community-level (rather
than through household income) and consisted of 118 students (14.4%)
from Poor communities, 133 (16.2%) from Working, 271 (33%) from
Middle, 212 (25.8%) from Upper Middle and 87 (10.6%) Upper classes.
Measures of subjective experience and time use are drawn from the
ESM, where each participant was given a programmable wristwatch
set to signal at random moments eight times a day from 7:30 am to
10:30 pm for one week. Upon hearing the signal, participants completed
a form containing open-ended questions about what they were doing
at that moment as well as multiple-choice items regarding whom they
were with and close-ended scales addressing a wide range of feelings
and conditions associated with that moment. The data included here
are from those students who completed at least 15 responses over the
course of the week.
The open-ended items about the student’s current activity were coded
into several dozen speciﬁc categories, that can also be converted into
much more generalized groupings such as School (eg. studying, listen-
ing to lecture), Active Leisure (playing games, sports), Passive Leisure
HAPPINESS IN EVERYDAY LIFE 189
(watching tv, listening to music), Maintenance (grooming, eating, trans-
portation) and Work activities (after school jobs). In addition, two vari-
ables used for assessing the activity’s conditions for ﬂow experiences
are (1) the amount to which they found the current activity Challenging
(a 1–9 scale, where 1 is the lowest and 9 the highest value) and (2) the
student’s level of Skill in the activity (using the same 1–9 scheme).
Mood variables include a 1–7 scale (1 being the most negative and 7,
the most positive value) asking the student if they felt Happy (vs. Sad),
Strong (vs. Weak), Proud (vs. Ashamed), Sociable (vs. Lonely), Excited
(vs. Bored), Active (vs. Passive) and a 1–10 scale (where 1 is the most
negative and 10, the most positive) asking “Did you feel good about
yourself?”. These variables can be used to refer to speciﬁc moments in
time, for example what is the level of happiness when watching televi-
sion versus doing sports? Furthermore, an individual’s total responses
can also be combined to form a Person-level variable. Such variables
can be used to compare people who rank happier than others overall.
A third way these variables can be used is to combine the contextual and
the personal. For example, using happiness as referent, a Person-Level
contextual variable tells the amount of happiness a particular individual
experiences in a speciﬁc activity.
Momentary Changes in Happiness
Days of the Week
There is a widely held belief that people are more sad on certain days
of the week than on others. “Blue Mondays” in particular are held to
be depressing. In this sample variation in happiness (using “z” scores
calibrated on individual means) was very slight, although signiﬁcant.
An ANOVA produced an Fvalue of 3.4 (p<0.002). The lowest
happiness was reported on Sundays, and each day afterwards happiness
increased slightly, reaching its peak on Saturdays (see Figure 1).
Post-hoc Bonferroni tests indicated that respondents were signiﬁ-
cantly happier on Saturdays than they were on Mondays, Tuesdays and
Wednesdays (Sunday responses were fewer and had a greater variance
in happiness, thus yielded no signiﬁcant differences).
Clearly, the social structure of time has an impact on happiness: The
early part of the weekend, with its freedom from work or school, is
experienced as liberating. The effect is probably greater on adults, for
whom the working week is presumably even more constraining than it
is for teenagers.
190 MIHALY CSIKSZENTMIHALYI AND JEREMY HUNTER
Figure 1. Happiness (beep-level z-score) by day of week.
Figure 2. Happiness (beep-level z-score) throughout the weekday.
Times of Day
During the weekdays, time is structured by work or school requirements
according to a circadian pattern. The ﬁrst part of the day, spent at work
or school, tends to be less happy, except for a peak at lunch-time.
There is a dip after lunch, followed by higher reports of happiness in
the afternoon when one is again free (see Figure 2).
If we contrast afternoon reports with those obtained before noon, the
difference in happiness is striking (F=56.5, p<0.00001).
What one happens to be doing at the moment of the signal has an
even more speciﬁc effect on happiness. There are ten main activities
that teenagers do during the week, each taking up 2% or more of their
waking time. For seven of these ten, the average level of happiness is
HAPPINESS IN EVERYDAY LIFE 191
Happiness (aggregated Person–Level z-score) by top ten most frequent
Happy (z-score) T-value P< N
TV 0.03 1.24 NS 666
Talking with Friends 0.35 9.87 0.000 325
Eating a Meal 0.19 5.78 0.000 524
Unspeciﬁed Homework −0.30 −8.21 0.000 409
Individual Work −0.11 −2.99 0.003 358
Listening to Lecture −0.21 −5.36 0.000 381
Chores −0.21 −4.44 0.000 343
Fun Reading/Writing −0.01 −0.14 NS 324
Mathematics −0.25 −5.27 0.000 327
Talking with Family −0.03 −0.53 NS 281
∗Activities representing at least 2% of time during the week (1% is roughly
equal to 1 h)
signiﬁcantly higher or lower than it is on the average (see Table I).
The highest level of happiness is reported when talking with friends
(Mean z=0.35, t=9.87, p<0.00001), and the lowest when doing
school-related homework (Mean z=−0.30, t=−8.21, p<0.00001).
Another way to observe the effects of activities is by combining them
into six major categories, which together account for 21,631 responses,
or 93% of the total. Four of the six categories are signiﬁcantly different
from the average (p<0.00001). Whenever students are involved with
School-related activities, their happiness level is below average (Mean
z=−0.19); when Socializing with friends, when involved in Active
Leisure, or in Passive Leisure it is above average (Mean z=0.28,
0.19 and 0.11, respectively). Some of the happiest experiences reported
in the Active Leisure category are Sports (Mean z=0.50), Music
(z=0.29) and visual Art (z=0.27). The other two major categories,
which are indistinguishable from the average in terms of happiness,
are Working and Maintenance activities such as doing chores, eating,
dressing, and so on.
Who one happens to be with companions, it also impacts signiﬁ-
cantly on the level of reported happiness. In terms of companionship,
youth experience the lowest levels of happiness when they are Alone
(Mean z=−0.12, p<0.0001), with Teachers (Mean z=−0.09,
192 MIHALY CSIKSZENTMIHALYI AND JEREMY HUNTER
p<0.0001), and with Classmates (Mean z=−0.07, p<0.0001)
while being with friends corresponds to the highest level (Mean
z=0.21, p<0.0001). Being with Parents is at the average for hap-
piness, which is lower than being with a Sibling (Mean z=0.03,
p<0.016). Spending time with a Relative, however, is associated with
more happiness (Mean z=0.09, p<0.002) than either of these two
Person-Level Correlates of Happiness
The analysis thus far focused on how happiness is experienced at the
moment – how situational context relates to shifting levels of happiness
within the individual. The ESM data can be also analyzed at the person
level, making it possible to answer the question, what differentiates
young people who on the average report higher levels of happiness
from those who during the week report being less happy?
General traits of the person have rather strong relationships to hap-
piness. The largest difference reﬂects the Social Class of Community
(SCC) in which the teenagers live. SCC was computed on ﬁve levels
of increasing afﬂuence: Poor (mostly single-parent, unemployed),
Working Class, Middle Class, Upper-Middle Class and Upper Class.
Contrary to expectations, the highest level of happiness was reported
by young people living in Working Class communities, then by those
in Middle Class, Poor, Upper Class and ﬁnally Upper Middle Class
environments. An ANOVA in which all the demographic variables
(i.e. age, gender, SCC, Ethnic background) were entered showed the
strongest effect for SCC (F=8.09, p<0.0001).
Age was the second most important factor (F=6.45, p<0.0001)
Happiness decreases through the teenage years; it reaches its lowest
point by age 16, and then shows a small recovery by age 18 (see also
Moneta, 2001). Gender and Ethnic background did not show signiﬁcant
effects, even though African-American and Hispanic youth had higher
levels of happiness than Caucasians and Asians – but these differences
appear to be due more to social class than to ethnicity.
Boys and girls generally did not differ in terms of happiness. How-
ever, the ANOVA showed one signiﬁcant interaction (F=2.92,
p<0.02) between gender and SCC. Poor girls (5.5) experience more
happiness than Poor boys (5.0) (t=−2.51, p<0.014).
HAPPINESS IN EVERYDAY LIFE 193
In the previous section, we have seen that teenagers are happier when
they do certain things (e.g. in leisure) than when they do others (e.g.
study). Here we are looking at the issue from a trait-like, rather than
a state-like perspective: in other words, are teenagers who spend more
time in leisure activities during the week happier than those who spend
more time studying?
Contrary to what one might expect, the amount of time spent in
school-related activities during the week is positively related to happi-
ness (multiple regression (MR), t=2.25, p<0.024), indicating that
those teenagers who study more are in fact happier, even though study-
ing is lower in happiness than most other activities. This apparently
paradoxical ﬁnding is one of the important ways in which the ESM can
reveal the fact that relationships that are negative at the state level can
at time be positive at the trait level. The percent of time students spend
socializing is also positively related to happiness (t=2.61, p<0.009).
In this case, both momentary and Person–Level relationships point in
the same direction. Young people feel happier when they interact with
peers, and those who do so more often are on the average happier than
those who interact less.
One unexpected ﬁnding was that of the smaller activity categories
the one that showed the strongest relation to happiness at the person
level was Reading a book for pleasure. The relationship was negative
(t=−2.09, p<0.04), suggesting that teenagers who spend more
time during the week are also generally less happy. This result could
be due to the fact that young people who read more are less often in the
company of their peers. There is a slight negative correlation (−0.09,
p<0.08, n=825) between the amount of time spent reading and the
percent of time spent with friends.
The social context affects happiness in complex ways. Those young
people who spend more time alone are in general less happy (MR,
t=−3.85, p<0.0001). Those who spend more time with relatives
during the week tend to be happier (MR, t=2.24, p<0.01). Although
being with friends is related to happiness it is not signiﬁcantly so,
because older teenagers spend more time with friends, while being
less happy than younger ones. Therefore, the age effect cancels out the
beneﬁcial effect of spending time with friends.
194 MIHALY CSIKSZENTMIHALYI AND JEREMY HUNTER
Standardized regression coefﬁcients from multiple stepwise regres-
sion of mean (Person–Level) happiness on Person–Level mood
variables (controlling for demographic variables)
Independent variables Mean (Person–Level) T-score P<
Strong 0.099 2.5 0.012
Feel Good About Self 0.093 6.62 0.000
Sociable 0.160 4.5 0.000
Excited 0.230 11.74 0.000
Proud 0.230 5.75 0.000
Active −0.050 3 0.003
Grade Level in School −0.050 −5.06 0.000
Constant 1.650 9.2 0.000
F-value 135.3 0.000
The Relation of Happiness to Other Moods
What other dimensions of subjective experience differentiate a happy
young person from one who is less so? To answer that question, we did
a regression in which the dependent variable was a person’s average
happiness score for the week, and the predictors included all the other
mood variables. Such a MR explained 55% of the variance in happiness
The strongest predictor of trait happiness was how Excited
(vs. Bored) a person felt, followed by the variables Feeling Good about
Self, Proud (vs. Ashamed), Sociable (vs. Lonely), feeling Active and
Strong (vs. Weak). The correlation coefﬁcients of these variables with
Happiness (and controlling for age), were 0.58, 0.59, 0.47 and 0.53,
respectively (with N=799, all p<0.0001).
Happiness and the Conditions for Flow
It was expected that young people who spend more time in situations
they perceive as being conducive to ﬂow would be on the whole happier.
To measure whether a person was more likely to be in a Flow condition
we calculated the percent of time spent in situations that were above
the mean level of challenge and the mean level of skill at the same time.
When a person was above the mean of skills but below mean challenge,
the condition was considered conducive to Relaxation. High challenges
and low skills were counted as Anxiety, and low challenges with low
skills as Apathy.
HAPPINESS IN EVERYDAY LIFE 195
Standardized regression coefﬁcients from multiple stepwise regression
of mean (Person–Level) happiness on ﬂow conditions (controlling for
Independent variables Mean (Person–Level) T-score P<
Flow Condition 0.013 6.05 0.000
Relaxation Condition 0.008 4.93 0.000
Grade Level in School −0.097 −7.01 0.000
Social Class of Community −0.080 −3.1 0.002
Constant 5.560 31 0.000
F-value 29.56 0.000
Table III shows the ﬁnal regression model, which includes Age and
the Gender by SCC interaction as well as the four Flow-related vari-
ables. The full model explains 12.4% of the variance in happiness. The
frequency of time spent in the Flow condition is a very strong predictor
of happiness (t=6.05, p<0.0001) even after taking the signiﬁcant
demographic variables into account.
The Final Model
To see if combining all the correlates of happiness in one model would
enhance understanding of the phenomenon, we created a ﬁnal regres-
sion model that included the most promising variables form previous
analyses – excluding, however, the mood variables which as we have
seen above (Table II), explain 54% of the variance in happiness.
The resulting model is the one reported in Table IV. The combined
predictive value is not much higher than that of some of the demo-
graphic variables taken singly, as it attains only 15% of the variance in
happiness. Nevertheless, the pattern is suggestive.
The pattern can be summarized as follows: Happier teenagers tend
to be younger, from lower socio-economic circumstances. They spend
less time alone and less time reading books. They spend more time
either in high challenge/high skill Flow producing situations, or low
challenge/high skill Relaxing situations. These are also the young peo-
ple who feel more Excited, Proud, Sociable, Strong, Active and Good
196 MIHALY CSIKSZENTMIHALYI AND JEREMY HUNTER
Standardized regression coefﬁcients from multiple stepwise regression
of mean (Person–Level) happiness on ﬂow conditions and time usage
(controlling for demographic variables)
Independent variables Mean (Person–Level) T-score P<
Percent of Time
Spent Alone −0.010 −4.60 0.000
In Flow Condition 0.013 6.2 0.000
In Relaxation Condition 0.009 5.22 0.000
Spent Reading/Writing for Fun −0.014 −2.08 0.037
Grade Level in School −0.080 −5.8 0.012
Social Class of Community −0.065 −2.5 0.000
Constant 5.590 31.46 0.000
F-value 24.77 0.000
The ESM makes it possible to separate the immediate context of
happiness from more long-term conditions. In terms of momentary
effects, it is clear that what one does and whom one is with will modify
a person’s base-line of happiness. Freely chosen activities and the com-
pany of peers raise the level of happiness, while obligatory activities
like schoolwork and the condition of solitude lowers it. The social struc-
ture of time affects happiness in a similar way: young people are much
happier in the afternoons and evenings of weekdays, when they are
free of requirements imposed by adults, and on weekends. But by the
end of the weekend, on Sunday afternoons, their happiness decreases
in anticipation of the school-day to come.
The demographic analyses provide rather counterintuitive sugges-
tions. That happiness decreases during the conﬂicted teenage years is
not surprising, and the recovery around age 18 has been documented
before (Moneta, 2001). What is surprising is the lack of positive corre-
lation between happiness and ﬁnancial afﬂuence. That teenagers from
working-class, and even impoverished backgrounds should be happier
than upper-middle-class teenagers living in exclusive suburban com-
munities is difﬁcult to explain. It is possible that some selection bias
is responsible for this result: perhaps relatively more students from
lower class backgrounds who were happy volunteered and completed
HAPPINESS IN EVERYDAY LIFE 197
the ESM compared with more afﬂuent students. But the rates of volun-
teering had been high in all schools, including the ones in the inner city
neighborhoods, so this explanation could not account entirely for the
ﬁndings. Perhaps in the afﬂuent suburban sub-culture it is not “cool”
to admit to being happy. Or perhaps material well-being is in fact an
obstacle to happiness. Recent research on materialism suggests that
excessive concern with consumer goods and material possessions is
inversely related with positive developmental outcomes (Schmuck and
Sheldon, 2001). In any case, this ﬁnding clearly deserves further study.
Aggregating responses over a week’s time suggests that happiness
is strongly related to an extraverted lifestyle. Not being alone, feeling
excited, proud, being in high-challenge, high skill situations are all
related to how happy a young person feels. It seems that at least at
this stage of life an experience of what we may call “vitality,” or eros,
is the most distinctive feature of happiness (Csikszentmihalyi, 1990;
1999; see also Ryan and Frederick, 1997, for recent studies dealing
At the same time, it is important to notice that studying, which pro-
duces an experience of sadness as it is occurring, helps young people
feel happier in the long run. This is an example of how building “ psy-
chological capital” involves the transformation of potentially negative
experiences in positive experience over time (Csikszentmihalyi, 2003).
For example, in a longitudinal study of talented teenagers we found that
only those who learned to enjoy practicing their talent (i.e. mathemat-
ics, music, science, art, athletics) were able to continue developing it
through the high school years. Those who became bored or stressed
when working on their talent sooner or later gave up, while those
who experienced ﬂow in their work continued to perfect their talent
(Csikszentmihalyi et al., 1997).
These results suggest that momentary happiness, at least for young
people, is a function of the ability to express their potential vitality
as fully as it is possible given the socialization demands the adult
world places on them. Teenagers ascribe “happiness” to their moods
when they are in situations of relative freedom, in the company of
age-mates, able to engage in ﬂow activities that stretch their skills and
makes them feel alive and proud. The same conditions are implicated
in more enduring, trait-like happiness. Here, however, happiness is
also affected by preparation for the future: young people who study
more are on the whole happier, presumably because they realize that
198 MIHALY CSIKSZENTMIHALYI AND JEREMY HUNTER
by building psychological capital the range of opportunities and hence
their freedom will increase in the future.
If this is the case, the results have important implications for edu-
cation and social policy. Happiness will increase to the extent that
individuals are provided with the means to learn skills that can be
deployed to meet reasonable challenges; that they are given freedom
to express themselves within bounds of responsibility; that they are
allowed to experience the joy of interaction with peers of one’s choice
and with adults that care for their well-being. These requirements for
happiness presumably operate at every level of societal complexity,
from the macro-level of the economy and political structure to the
meso- and micro-levels of community, school and family. There are
clear trends in contemporary life that militate against such conditions.
It is difﬁcult for a young person to be happy when living in a sterile sub-
urb that lacks opportunities for action, forced to attend schools where
there is little chance to express oneself except in abstract intellectual
terms, surrounded by a small nuclear family that is seldom together
and relaxed enough to interact freely. Understanding more clearly the
conditions that affect happiness is a prerequisite if social scientists are
to help improve the quality of life.
Brickman, P., D. Coates and R.J. Janoff-Bulman: 1978, ‘Lottery winners and accident
victims: Is happiness relative?’, Journal of Personality and Social Psychology, (36),
Csikszentmihalyi, M.: 1990, Flow: The Psychology of Optimal Experience
(HarperCollins, New York).
Csikszentmihalyi, M.: 1997, Finding Flow: The Psychology of Engagement with
Everyday Life (Basic Books, New York).
Csikszentmihalyi, M.: 2003, Good Business: Leadership, Flow, and the Making of
Meaning (Viking, New York).
Csikszentmihalyi, M and B. Schneider: 2001, ‘Conditions for optimal development
in adolescence: An experiential approach’, Applied Developmental Science 5(3),
Csikszentmihalyi, M., R. Larson and S. Prescott: 1977, ‘The ecology of adolescent
activities and experiences’, Journal of Youth and Adolescence 6(3), pp. 281–294.
Csikszentmihalyi, M., K. Rathunde and S. Whalen: 1993, Talented Teenagers: The
Roots of Success and Failure (Cambridge University Press, New York).
Diener, E.: 2000, ‘Subjective well-being: The science of happiness and a proposal
for a national index’, American Psychologist, 55(1), pp. 34–43.
HAPPINESS IN EVERYDAY LIFE 199
Graef, R.: 1978, ‘An analysis of the person by situation interaction through repeated
measures’, Unpublished doctoral dissertation, The University of Chicago.
Hektner, J.: ESM Handbook. (in preparation).
Inglehart, R. and H.D. Klingemann: 2000, ‘Genes, culture, democracy and
happiness’, in E. Diener and E.M. Suh (eds), Culture and Subjective Well-being
(The MIT Press, Cambridge, MA), pp. 165–183.
Kahneman, D.: 1999, ‘Objective happiness’, in D. Kahneman, E. Diener and
N. Schwartz (eds), Well-being: The Foundations of Hedonic Psychology
(Russel Sage, New York), pp. 3–25.
Kubey, R. Larson and M. Csikszentmihalyi: 1996, ‘Experience sampling method
applications to communication research questions’, Journal of Communication
46(2), pp. 99–120.
Lykken, D. and A. Tellegen: 1996, ‘Happiness is a stochastic phenomenon’,
Psychological Science 7, pp. 186–189.
Moneta, G., B. Schneider and M. Csikszentmihalyi: 2001, ‘A longitudinal study
of the self-concept and experiential components of self-worth and affect across
adolescence’, Applied Developmental Science 5(3), pp. 125–152.
Patton, J.D.: 2002, ‘The role of problem pioneers in creative innovation’, Creativity
Research Journal, 14(1), pp. 111–126.
Ryan, R. and C. Frederick: 1997, ‘On energy, personality, and health: Subjective
vitality as a dynamic reﬂection of well-being’, Journal of Personality, 65(3),
Schwartz, N. and F. Strack: 1999, ‘Reports of subjective well-being: Judgmental
processes and their methodological implications’, in D. Kahneman, E. Diener
and N. Schwartz (eds), Well-Being: The Foundations of Hedonic Psychology
(Russel Sage, New York), pp. 61–84.
Schmuck, P. and K.M. Sheldon: 2001, Life-goals and Well-being (Hogrefe & Huber,
Seligman, E.P.: 2002, Authentic Happiness (Free Press, New York).
Tellegen, A., D.T. Lykken, T.J. Bouchard, K.J. Wilcox, N.L. Segal and S. Rich: 1988,
‘Personality similarity in twins reared apart and together’, Journal of Personality
and Social Psychology 54, pp. 1031–1039.
Veenhoven, R.: 1995, ‘The cross-national pattern of happiness: Test of predictions
implied in three theories of happiness’, Social Indicators Research 34, pp. 33–68.
Address for Correspondence:
Claremont Graduate University
The Quality of Life Research Center
171 E, Tenth Street, CA 91711