Content uploaded by Amy L Collins
Author content
All content in this area was uploaded by Amy L Collins on Dec 06, 2019
Content may be subject to copyright.
RESEARCH PAPER
Flow and Happiness in Later Life: An Investigation
into the Role of Daily and Weekly Flow Experiences
Amy Love Collins ÆNatalia Sarkisian ÆEllen Winner
Springer Science+Business Media B.V. 2008
Abstract Fifty-four older adults ranging in age from 70 to 86 years old (M=77.54)
reported daily levels of positive and negative affect, life satisfaction and daily activities for
seven consecutive days. Hierarchical linear modeling (HLM) was used to investigate inter-
and intra-individual effects of flow experiences on affect. Higher quality of flow was
positively associated with high arousal positive affect (i.e., feeling peppy, enthusiastic,
happy), negatively associated with low arousal negative affect (i.e., feeling sad and dis-
appointed), and positively associated with life satisfaction. However, more frequent flow
experiences throughout the week predicted lower average levels of positive affect and life
satisfaction. Overall, the results demonstrate that flow is linked to the affective experiences
of older adults, and that an individual’s overall propensity to experience flow may be
influential beyond the immediate effects of a given flow experience.
Keywords Happiness Life satisfaction Flow Older adults Aging
In light of research demonstrating that happiness is linked with better coping, lower
morbidity, and lower mortality (Fredrickson 2001; Pressman and Cohen 2005; Salovey
et al. 2000), older age is a time when happiness is particularly important. Fortunately,
despite stereotypical notions that people get depressed as they age, there is little evidence
to support a link between aging and diminished happiness (Blazer 2003). Numerous studies
demonstrate that there is either no decline or a small decline in average levels of happiness
in old age (Baltes and Mayer 1999; Brandtstaedter and Wentura 1995; Cantril 1999;
Carstensen et al. 2000; Mroczek and Kolarz 1998; Ryff 1989; World Values Study Group
1994). Many studies report that happiness increases with age, particularly after midlife
(Cantril 1999; Diener and Suh 1997; Ryff 1989; World Values Study Group 1994).
A. L. Collins (&)
Office of Population Research, Princeton University, 263 Wallace Hall, Princeton, NJ 08544, USA
e-mail: alc@princeton.edu
N. Sarkisian E. Winner
Boston College, Chestnut Hill, Boston, USA
123
J Happiness Stud
DOI 10.1007/s10902-008-9116-3
These findings may depend on the aspect of happiness under consideration, however, as
happiness is a multi-faceted construct (Diener et al. 2003b). In later life, adults may
experience slightly less positive affect than younger individuals, but that is accompanied
by a decline in negative emotions, especially the higher arousal emotions such as anger and
fear (Gross et al. 1997; Kunzmann et al. 2000; Labouvie-Vief et al. 1989; Stacey and Gatz
1991). A decline in negative emotions suggests that older adults may be better at regulating
their emotions than younger adults (Carstensen 1995; Labouvie-Vief et al. 1989).
However, while depression and negative affect decline with age on average, inter-
individual variation in happiness increases with age (Headey and Wearing 1989; Lawton
et al. 1995; Lucas et al. 2003). The differences between ‘‘advantaged’’ and ‘‘disadvan-
taged’’ individuals become more pronounced in terms of their cognitive, physical, social,
and financial resources as they age (Dannefer 1987). Therefore, the discrepancy between
older adults who age successfully and those who do not is perhaps larger than at any other
point in the life span (Dannefer 1987). Research exploring this variation finds that cog-
nitive and personality factors such as intelligence and extraversion, social capital factors
like marital status and social network size, and economic resources and physical health can
account for some of the variation in levels of happiness in older adults (Isaacowitz and
Smith 2003; Mroczek and Kolarz 1998; Pavot and Diener 2004). Importantly, these factors
remain relatively stable for a given individual over time and therefore cannot account for
intra-individual variation in feelings of happiness. Hence, we argue that it is important to
examine how older adults’ daily activities and experiences may contribute to the variation
in levels of happiness—both for a given individual over time and across individuals.
Specifically, in this study we examine one factor that may contribute to higher positive
affect, reduce the levels of negative affect, and boost life satisfaction in later life—the
presence of flow.
Flow is an intrinsically rewarding or optimal state that results from intense
engagement with daily activities (Csikszentmihalyi 1990). Because the ability to be
highly engaged in daily activities is a characteristic of successful aging (Rowe and Kahn
1997), the capacity and opportunity to experience flow may increase positive affect and
life satisfaction and protect against negative affect in late adulthood. Structured and
meaningful activities are particularly important for subjective well-being in individuals
who are faced with new constraints on their daily lives, like those who are retired
(Hendricks and Hendricks 1986). While retirement may initially result in an increase in
well-being (Kim and Moen 2002), continued retirement may lead to an increase in
depressive symptoms for some individuals, particularly for those deficient in economic
and social resources, those with physical health problems, and those lacking structured,
meaningful activities (Kim and Moen 2002; Pinquart and Schindler 2007; Wang 2007).
In contrast, those engaged in flow activities may experience higher well-being in
retirement. The present study tests this hypothesis by examining relationships between
daily and weekly levels of happiness and experiences of flow in the lives of retired older
adults.
In Csikszentmihalyi’s (1982,1990) view, happiness is not the result of good fortune or
chance. Rather, it is achieved by the cultivation of and control over one’s inner experience
so that one can have optimal experience, or flow. Csikszentmihalyi (1988,1990) posits that
flow experiences do not occur in passive moments but at times when one is intensely
engaged in a motivating activity, intently focused, and challenged. This engagement creates
a feeling of exhilaration, satisfaction, and happiness. Flow experiences tend to derive from a
balance between challenges and skills: the experience must contain enough challenge to
stimulate the person, but not so much as to create anxiety (Csikszentmihalyi 1988,1990).
A. L. Collins et al.
123
According to Csikszentmihalyi (1988,1990), flow-generating activities involve intense
concentration, time distortion (i.e., time is reported to speed up or slow down during the
flow activity), increased confidence, and a loss of self-awareness (Csikszentmihalyi 1988).
What constitutes a flow activity varies greatly. Flow activities studied and reported in the
literature range from playing golf to composing music to engaging in housework (Asakawa
2004; Csikszentmihalyi 1988,1990; Han 1988).
Flow is not easily quantifiable (Csikszentmihalyi 1982; Csikszentmihalyi and
LeFevre 1989). Many researchers have operationalized flow by assuming that flow is
always present when an activity involves a certain ratio of challenges and skills. Such
assumptions are problematic, however, especially given that different studies used
different ratios. Some studies have defined flow by the condition that the rating of the
challenges of an activity must equal the rating of skills one possesses (Csikszentmihalyi
1988; Voelkl and Ellis 1998); others have proposed that for older adults, flow occurs
when challenges are slightly above the competence of the individual (Lawton 1989);
yet others have defined flow as engagement in activities in which the level of chal-
lenges and skills are both above average for the individual (Csikszentmihalyi and
LeFevre 1989) or for the group studied (Asakawa 2004; Csikszentmihalyi and Rathunde
1993).
An alternative way to measure flow, one which we believe may be more authentic, is to
avoid such assumptions and to rely on individuals’ self-evaluations of their experiences of
flow. This method asks participants if they recognize descriptions of a flow experience as
something that they are currently experiencing or have experienced in the past
(Csikszentmihalyi 1988; Han 1988). Moreover, as Asakawa (2004) showed in a study of
flow in Japanese college students, this method makes it possible to assess the quality of the
flow experiences by asking respondents to rate the strength of their feelings of intense
concentration, confidence, and ‘‘losing oneself.’’ Hence, our study relies on subjective
ratings of the presence of flow and its quality.
Only about one-third of individuals in a survey of the United States and Germany never
or rarely experienced flow (Gallup Poll 1988, as cited in Asakawa 2004), but the frequency
and quality of flow experiences varies across individuals. Whereas individuals have dif-
ferent overall proclivity to experience flow, their experiences may vary day to day
(Asakawa 2004; Csikszentmihalyi 1990). To capture both the daily variation and the
overall personal predisposition, it is important to collect multiple measurements over time.
Such repeated measurement data should then be analyzed with techniques that allow
differentiating between the effects of daily fluctuations in flow experiences and the effects
of the overall propensity to experience flow.
Multiple studies connect flow with positive affect in younger populations (Asakawa
2004; Csikszentmihalyi 1988,1990; Csikszentmihalyi and Hunter 2003; Csikszentmih-
alyi and LeFevre 1989). Flow is also linked to self-esteem, life satisfaction, and
successful coping (Csikszentmihalyi 1990; Han 1988; Wells 1988). The causality of
these relationships remains uncertain. It is possible that flow provides momentary
positive affect, or that people seek out flow experiences when they are in a good mood.
It is likely that the relationship runs in both directions: flow experiences make one feel
happier and people are better able to experience flow when they feel happy. This is
consistent with evidence that positive affect seems to be related to less self-focus and
more outward directed behavior (Csikszentmihalyi and Figurski 1982; Greenberg and
Pyszczynski 1986; Ingram 1990).
Flow has been studied extensively as an optimal state for younger people (e.g.,
Csikszentmihalyi and Hunter 2003; Csikszentmihalyi and LeFevre 1989), but few studies
Flow and Happiness in Later Life
123
have examined flow and happiness in old age. Two exceptions are studies by Han (1988)
and Voelkl (1990), conducted on a sample of older, Korean immigrants and a sample of
nursing home residents, respectively. These studies demonstrate that older adults can
experience flow and that flow is positively related to feelings of happiness in some older
adults. However, the samples used in these studies make it difficult to generalize the
findings to other older populations. Moreover, the Han (1988) study was cross-sectional
and did not conduct multiple measurements over time. Voelkl’s (1990) study used the
experience sampling method (ESM) to collect measurements of flow and affect on 12
nursing home residents. ESM uses a palm pilot or beeper technology to prompt participants
to record their momentary thoughts and activities at random intervals throughout the day.
While this technology provides unparalleled immediate, event contingent sampling, it
presents problems when studying an older population, as many older adults may not be
accustomed to using handheld computer devices. This method may also fail to detect very
brief, momentary flow experiences that would be better reflected in a retrospective
assessment of emotions and flow activities collected at the end of the day (Kahneman et al.
2004). Furthermore, despite the availability of multiple measurements, Voelkl (1990) did
not separate the effects of the overall individual propensity to experience flow and the
effects of daily fluctuations in flow experiences.
Most previous studies of flow relied on affect to assess happiness, although some
utilized measures of life satisfaction (e.g., Han 1988). Studies focusing on affect usually
measured positive and negative affect on a single continuum (e.g., Voelkl 1990). Mea-
suring positive and negative emotions separately has been considered essential in
subjective well-being research, however, since Bradburn (1969) discovered that they are
related but distinct constructs (Diener et al. 2003b). Indeed, positive affect is not the
absence of negative affect, and the two types of affect have distinct physiological and
neurobiological correlates (Cacioppo and Bernston 1999; Cacioppo et al. 1999; Hamer
1996; Ryff et al. 2006). In addition, affect has both valence and arousal level, whereby
higher arousal emotions, like peppy and happy, are distinct from lower arousal emotions
such as calm and relaxed (Feldman Barrett and Russell 1998).
The present study conceptualizes happiness as a multi-faceted construct (Diener et al.
2003b) and utilizes five separate measures of happiness, including positive and negative
affect as separate constructs with both high and low arousal levels as well as the life
satisfaction scale. Furthermore, building on the Han (1988) and Voelkl (1990) studies, we
rely on detailed, retrospective assessment of happiness and flow collected at the end of
each day over a period of 1 week, and use such repeated measures of flow and happiness to
examine separately the effects of both the daily fluctuations in flow and the average
experiences of respondents.
The primary research question asks whether or not flow is a significant predictor of
happiness. Based on the findings of past research, we tested the following three
hypotheses: (1) Older adults will report higher positive affect and life satisfaction and
less negative affect on those days that they experienced flow (as measured by recog-
nizing descriptions of flow as similar to specific personal experiences); (2) Older adults
will report higher positive affect and life satisfaction and less negative affect on those
days that they reported a higher quality of flow (as measured by how strongly partici-
pants felt that certain aspects of flow were part of these recalled experiences); (3) Older
adults with more frequent flow experiences and with higher average quality of flow
experience will have higher average levels of positive affect and life satisfaction but
lower average levels of negative affect.
A. L. Collins et al.
123
1 Methods
1.1 Participants
Participants were volunteers recruited through senior centers and other community orga-
nizations in the greater Boston area. Eighty retired, older adults initially volunteered to
participate in the study. Seven volunteers were excluded based on the screening procedures
that assessed their ability to follow the directions of the study as well as the medications
they were taking. More specifically, we excluded those volunteers whose medications
indicated chronic mental or physical illness that might interfere with participation in this
study. Nineteen volunteers dropped out of the study for various reasons including illness,
relocation, and lack of interest. The remaining sample consisted of 13 men and 41 women
(N=54) with an age range of 70–86 years (M=77.54; SD =3.74). Women constituted
the majority of our sample, which reflects the trend at older ages in the United States
population (Smith 2003). Still, the disproportionally large number of women in our study
may have introduced a potential bias. The participants reported their ethnicity as White
with the exception of one woman who identified herself as Hispanic. Forty-three percent of
the sample was married and 44% was widowed. The remaining 13% were single, divorced,
or indicated ‘‘other.’’ The majority of the sample had had at least a college education, with
30% having attended or completed college and 31% having attended graduate school. Over
a third of the participants (39%) had no more than a high school level education. Most
participants rated their health as very good (50%) or excellent (41%), and only one par-
ticipant rated her health as poor. These self-reports of physical health suggest that this was
a high functioning group of older adults.
1.2 Procedure
Respondents provided repeated retrospective assessments of daily flow experiences and
happiness over the course of seven days. Interested participants were mailed a packet that
included a consent form, a clear description of the procedure, and the measures in the order
that they were to be filled out over the course of the 7 day observation period. All of these
measures were to be filled out at the end of each day. A few days after the study packet was
mailed, participants were called in order to make sure that they understood the instructions.
Importantly, a pilot study conducted at an earlier stage confirmed that older adults
understood the instructions and the measures and were successfully able to follow the
procedure. The initial packet also included addressed, stamped envelopes which partici-
pants used to mail back the informed consent form and packet of measures upon
completion. A total of 378 daily observations were collected from 54 participants.
1.3 Measures
1.3.1 Affect
In concordance with Diener et al.’s (2003a,b) and Feldman Barrett and Russell’s (1998)
recommendations, we utilized separate measures of positive and negative affect, both low
arousal and high arousal. Participants rated their emotions each day using a list of positive
and negative affect terms that refer to emotions characterized by varying degrees of arousal
(adapted from Feldman Barrett and Russell 1998). Participants indicated on an eight point
Likert-type scale the extent to which each adjective described their emotion for that day
Flow and Happiness in Later Life
123
overall (e.g., 0 =neutral to 7 =very calm). Positive Affect -Low Arousal was the total
score of three low arousal positive affect terms (satisfied, relaxed, calm) with a possible
range of 0–21. Positive Affect -High Arousal was the total score of three high arousal
positive affect terms (peppy, enthusiastic, happy) with a possible range of 0–21. Negative
Affect -Low Arousal was the total score of low arousal negative affect terms (sad, dis-
appointed) with a possible range of 0–14, and Negative Affect -High Arousal was the
sum of high arousal negative affect terms (nervous, afraid, aroused) with a possible range
of 0–21. In our analyses, we used the square root of all affect variables in order to correct
for heterogeneity of variance (Snijders and Bosker 1999; Raudenbush and Bryk 2002).
1.3.2 Life Satisfaction
Life satisfaction was measured at the end of each day using The Satisfaction with Life
Scale (Pavot et al. 1991). The Satisfaction with Life Scale has been shown to be a valid and
reliable measure of life satisfaction and is suitable for use in a wide range of age groups.
Participants were asked to rate on a five-point Likert scale how strongly they agreed with
five statements: (1) In most ways my life is close to my ideal; (2) The conditions of my life
are excellent; (3) I am satisfied with life; (4) So far I have gotten the important things I
want in life; (5) If I could live my life over, I would change almost nothing. The life
satisfaction variable was the total score of these items with a possible range of 5–25. We
used the square root of the life satisfaction variable in order to correct for heterogeneity of
variance (Snijders and Bosker 1999; Raudenbush and Bryk 2002).
1.3.3 Presence of Flow
Participants filled out a flow questionnaire (adapted from a questionnaire developed by
Csikszentmihalyi 1982) towards the end of each day of the study. The questionnaire asked
participants to read two descriptions of flow (i.e. ‘‘I am so involved in what I am doing. I
don’t see myself as separate from what I am doing.’’ and ‘‘My mind isn’t wandering. I am
not thinking of something else. I am totally involved in what I am doing…I don’t seem to
hear anything…I am less aware of myself and my problems.’’), and to indicate if they had
had a similar experience that day. If participants reported such an experience after reading
the two quotations, they were considered to be participants who experienced flow (1 =yes
and 0 =no). Those who reported flow experiences were then asked to describe which
activity during that day they most readily associated with that experience. The majority of
the sample understood the concept of flow and reported having had at least one activity
during which they had the experience over the 7 day observation period. They associated
flow with activities like working, reading and writing, watching or playing sports, spending
time with grandchildren, using the computer, and working on personal projects. These
types of activities are commonly reported to be sources of flow in various populations
(Csikszentmihalyi 1988,1990; Han 1988). We calculated a weekly measure of number of
days with flow (ranging from 0 to 7) by adding up the number of days during the seven day
observation period when participants reported the presence of flow.
1.3.4 Quality of Flow Experience
Those participants who reported flow on a given day were then asked about the degree of
intense concentration, increased confidence, and loss of self-awareness during their flow
A. L. Collins et al.
123
experiences on that day; this information was used to create the quality of flow experience
scale (alpha =0.74; 5 items). Specifically, participants were asked to consider the activity
or activities that made them experience flow that day and to rate on a five-point Likert scale
how strongly they agreed with five statements adapted from the Leisure Diagnostic Battery
(Witt and Ellis 1987): (1) When I was involved in the activity, I forgot about everything
else; (2) I paid very close attention to the activity I was involved in; (3) During the activity,
there were times when things were going so well, I felt I could do almost anything; (4) I
forgot my worries during the activity I was involved in; (5) I thought less about my
problems during the activity. These items were summed to create a scale with a possible
range of 5–25. The average (person level) quality of flow measure was calculated as the
mean of all available quality of flow scores for a given individual over the 7 day period.
1.3.5 Control Variables
Socio-demographic variables including age (in full years), gender (1 =female and
0=male), education (1 =at least some college and 0 =high school or less), and self-
assessed health (1 =excellent and 0 =less than excellent) were utilized as control
variables.
1.4 Data Analyses
Hierarchical linear modeling (HLM) was used to analyze two levels of data, with seven
daily observations of happiness and flow nested within 54 individuals age 70 and older. In
this study, Level 1 (day level) variables contain information about individuals’ experiences
on specific days of study; they can be used to test Hypotheses 1 and 2. In contrast, Level 2
(person level) variables describe individuals’ socio-demographic position as well as
characterize their experiences over the entire week. Level 2 variables can be used to test
Hypothesis 3, that is, to determine if the overall propensity to experience flow in general
and quality flow in particular influenced one’s happiness.
First, preliminary analyses were performed to obtain descriptive statistics and correla-
tions among day level (Level 1) and person level (Level 2) variables. Here as well as in the
multivariate analyses, we used an alpha level of 0.10 to establish statistical significance
because of our small sample size on Level 2, as determined by power analysis using
Optimal Design software (Spybrook et al. 2006).
Second, HLM models with random intercepts were estimated. Such models utilize both
Level 1 and Level 2 predictors, and include two error terms, Level 1 (e
ij
) and Level 2 (u
j
),
in order to account for the multilevel structure of the unexplained variance. Two sets of
random intercept models were estimated. In both sets, the dependent variables were the
square roots of the happiness measures, and the controls included four Level 2 socio-
demographic variables. The first set estimated on the total sample of days aimed to test
Hypotheses 1 and 3 by assessing the effects of the presence of flow on positive and
negative affect and life satisfaction. In this model, the main independent variables of
interest were the presence of flow on Level 1 and the number of days with flow on Level 2:
Happinessij ¼b0þb1Presence of Flowij
þb2Number of Days with Flowj
þb3Agej
þb4Femalej
þb5Educationj
þb6Healthj
þujþeij:
Flow and Happiness in Later Life
123
The second set was estimated on a subsample of those days when flow was reported,
which included 46 respondents and 242 days. These models aimed to test Hypotheses 2
and 3 by evaluating the effects of the quality of flow. Therefore, the main independent
variables included the daily quality of flow on Level 1 and the average quality of flow on
Level 2:
Happinessij ¼b0þb1Quality of Flowij
þb2Average Quality of Flowj
þb3Agej
þb4Femalej
þb5Educationj
þb6Healthj
þujþeij:
In order to obtain estimates that are more robust and easier to interpret (Raudenbush and
Bryk 2002), we mean-centered all independent variables (both Level 1 and Level 2) by
subtracting their overall sample means, except for the dichotomous variables on both levels
(presence of flow, gender, health, and education) that were left uncentered.
2 Results
The means and standard deviations of all the variables are presented in Table 1. Within-
participant correlations for Level 1 variables are reported in Table 2and between-participant
correlations for Level 2 variables are reported in Table 3(affect and life satisfaction
variables were aggregated across time for each participant in this analysis). All of the within-
participant correlations among happiness measures are in the expected direction and
statistically significant. In contrast to our hypotheses, happiness measures demonstrate no
Table 1 Descriptive statistics
Variable Full sample Subsample of days with flow
NMean or
percent
SD NMean or
percent
SD
Daily (Level 1) characteristics
Positive Affect -Low Arousal
(square root)
374 3.89 0.52 238 3.86 0.51
Positive Affect -High Arousal
(square root)
371 3.79 0.74 237 3.73 0.81
Negative Affect -Low Arousal
(square root)
375 1.04 1.07 239 1.15 1.07
Negative Affect -High Arousal
(square root)
371 1.16 1.22 235 1.29 1.22
Life satisfaction (square root) 375 5.07 0.56 240 5.00 0.60
Presence of flow 377 64.19% – – – –
Quality of flow – – – 239 19.48 2.35
Individual (Level 2) characteristics
Number of days with flow 54 4.48 2.54 – – –
Average quality of flow – – – 46 19.29 1.76
Age 54 77.54 3.74 46 77.35 3.84
Female 54 74.07% – 46 71.74% –
College education 54 61.11% – 46 63.04% –
Excellent health 54 40.74% – 46 43.48% –
A. L. Collins et al.
123
Table 2 Level 1 (within participant) correlations (and the corresponding sample sizes)
Variable Positive affect Negative affect Life satisfaction Presence of flow Quality of flow
a
Low arousal High arousal Low arousal High arousal
Positive Affect -Low Arousal 1.00 (374)
Positive Affect -High Arousal 0.50*** (370) 1.00 (371)
Negative Affect -Low Arousal -0.39*** (374) -0.39*** (371) 1.00 (375)
Negative Affect -High Arousal -0.25*** (370) -0.10* (367) 0.50*** (371) 1.00 (371)
Life satisfaction 0.31*** (372) 0.33*** (369) -0.19*** (373) -0.12** (369) 1.00 (375)
Presence of flow 0.08 (373) 0.07 (370) 0.00 (374) 0.06 (370) 0.03 (374) 1.00 (377)
Quality of flow
a
0.13* (235) 0.22*** (235) -0.19*** (236) -0.01 (232) 0.17** (237) – 1.00 (239)
a
Subsample of days on which participants experienced flow (n=242)
*p\0.10; ** p\0.05; *** p\0.01
Flow and Happiness in Later Life
123
Table 3 Level 2 (between-participant) correlations (N=54)
Positive affect Negative affect Life
satisfaction
No. of days
with flow
Average
quality
of flow
a
Age Female College
education
Excellent
health
Low arousal High arousal Low arousal High arousal
Positive Affect -Low Arousal 1.00
Positive Affect -High Arousal 0.64*** 1.00
Negative Affect -Low Arousal -0.41*** -0.12 1.00
Negative Affect -High Arousal -0.17 0.12 0.79*** 1.00
Life satisfaction 0.47*** 0.29** -0.31** -0.05 1.00
No. of days with flow -0.21 -0.22 0.26* 0.20 -0.24* 1.00
Average quality of flow
a
0.14 0.30** -0.11 -0.06 -0.06 0.30** 1.00
Age 0.14 -0.07 -0.25* -0.23* 0.16 -0.32** -0.32** 1.00
Female -0.04 0.01 -0.21 -0.38*** -0.28** 0.05 0.32** -0.14 1.00
College education -0.16 -0.02 0.12 0.32** 0.03 0.09 0.00 0.02 -0.39*** 1.00
Excellent health 0.10 -0.08 -0.05 -0.11 0.01 0.08 0.16 -0.03 -0.03 0.04 1.00
a
Subsample of participants who experienced flow (N=46)
*p\0.10; ** p\0.05; *** p\0.01
A. L. Collins et al.
123
significant associations with the presence of flow. They are, however, significantly related to
the quality of flow measure, with the exception of the high arousal negative affect measure.
At Level 2, number of days with flow shows two unexpected correlations, a negative cor-
relation with life satisfaction (r=-0.24), and a positive correlation with low arousal
negative affect (r=0.26). The relationship between high arousal positive affect and average
quality of flow is in the expected direction (r=0.30). High arousal negative affect and
levels of life satisfaction are lower on average among women than among men, and negative
affect is lower among older individuals than among younger individuals. Interestingly, high
arousal negative affect is higher among those with college education than among those less
educated. In terms of the relationships among the independent variables, number of days
with flow was significantly related to mean reported quality of flow (r=0.30). There were
significant negative correlations between age and both the number of days with flow
(r=-0.32) and the average quality of flow (r=-0.32). Females were also more likely to
experience higher average quality of flow than males (r=0.32).
The results of the HLM analyses are summarized in Table 4. The variance components
reveal that there was substantial variation both over time and between individuals for all
outcome measures. They also show that there was more variation in negative affect than in
positive affect or life satisfaction for this sample.
With regard to Hypothesis 1, our results indicate that presence of flow did not signif-
icantly predict any of the outcomes. With regard to Hypothesis 2, the results show that on
the days that participants experienced flow, the reported quality of flow was a significant
direct predictor of high arousal positive affect, t(228) =3.16, p\0.01, and life satis-
faction, t(230) =2.28, p\0.05. Flow quality was also inversely related to low arousal
negative affect, t(229) =-2.35, p\0.05. Hypothesis 3 was not supported. There were no
significant relationships between average flow quality and the outcome measures. Unex-
pectedly, a higher number of days with flow predicted lower average levels of both low
arousal and high arousal positive affect as well as of life satisfaction, t(48) =-2.24,
p\0.05, t(48) =-3.19, p\0.01, and t(48) =-2.15, p\0.05, respectively.
With regard to the socio-demographic variables, the models revealed that being older
was related to lower levels of negative affect, which is consistent with previous research
(Gross et al. 1997; Labouvie-Vief et al. 1989; Mroczek and Kolarz 1998). Female par-
ticipants reported significantly lower levels of high arousal negative affect than did males.
Generally, studies have shown that older women have higher negative affect than older
men (Mroczek and Kolarz 1998), but these studies do not distinguish low from high
arousal negative affect. Finally, self-assessed health and education did not significantly
predict any of these outcomes. This finding is consistent with previous literature demon-
strating that socio-demographic variables generally do not explain much of the variance in
individuals’ levels of happiness (Diener et al. 2003b).
3 Discussion
This study explored the relationship between happiness and flow in a sample of 54 older
adults. Specifically, we investigated whether or not the presence and quality of flow on a
given day was significantly related to low and high arousal positive and negative affect and
level of life satisfaction on that day. We also investigated whether the general propensity to
experience flow, as indicated by more frequent and higher quality flow experiences over
the course of the seven day study period, was associated with affective experiences and life
satisfaction. Our analyses revealed three major findings.
Flow and Happiness in Later Life
123
Table 4 Coefficients (and standard errors) of hierarchical linear models
Variable Presence of flow models Quality of flow models
Positive affect Negative affect Satisfaction Positive affect Negative affect Satisfaction
Low High Low High Low High Low High
Level 1
Presence of flow 0.084
(0.058)
0.094
(0.079)
-0.003
(0.128)
0.124
(0.126)
0.020
(0.032)
–– – ––
Quality of flow – – – – – 0.026
(0.019)
0.067***
(0.021)
-0.087**
(0.037)
-0.006
(0.032)
0.023**
(0.010)
Level 2
No. of days with flow -0.040**
(0.018)
-0.079***
(0.025)
0.063
(0.049)
0.037
(0.060)
-0.047**
(0.022)
–– – ––
Average quality of flow – – – – – 0.013
(0.016)
0.049
(0.064)
0.016
(0.078)
-0.012
(0.080)
-0.001
(0.045)
Age 0.009
(0.015)
-0.027
(0.022)
-0.047
(0.028)
-0.064
(0.038)
0.007
(0.020)
0.013
(0.041)
0.003
(0.025)
-0.062*
(0.032)
-0.069*
(0.038)
0.016
(0.020)
Female -0.074
(0.132)
0.016
(0.184)
-0.442
(0.259)
-0.789**
(0.304)
-0.348**
(0.130)
-0.127
(0.144)
-0.164
(0.208)
-0.388
(0.310)
-0.886**
(0.338)
-0.367**
(0.148)
College education -0.144
(0.107)
0.020
(0.189)
0.021
(0.210)
0.363
(0.250)
-0.070
(0.163)
-0.081
(0.123)
0.031
(0.198)
-0.058
(0.209)
0.283
(0.241)
-0.001
(0.181)
Excellent health 0.093
(0.104)
-0.074
(0.181)
-0.118
(0.201)
-0.296
(0.236)
0.027
(0.139)
0.112
(0.112)
-0.138
(0.227)
-0.114
(0.218)
-0.270
(0.246)
0.047
(0.149)
Variance components
Level 2 0.141*** 0.363*** 0.475*** 0.687*** 0.259*** 0.145*** 0.406*** 0.474*** 0.605*** 0.289***
Level 1 0.129*** 0.220*** 0.633*** 0.601*** 0.045*** 0.132*** 0.254*** 0.642*** 0.679*** 0.056***
Sample size N=54
n=373
N=54
n=370
N=54
n=374
N=54
n=370
N=54
n=374
N=46
n=235
N=46
n=235
N=46
n=236
N=46
n=232
N=46
n=237
*p\0.10; ** p\0.05; ***p\0.01
A. L. Collins et al.
123
First, the majority of the older adults demonstrated an understanding of the concept of
flow and reported having had at least one flow experience over the 7 day study period.
Although what was deemed a flow activity varied by individual, the reported flow activities
were consistent with the types of activities that have been reported in previous studies of
flow (Csikszentmihalyi 1988; Csikszentmihalyi and LeFevre 1989; Han 1988). There were,
however, individual differences in the ability to experience flow. Bivariate correlations
demonstrated that the older participants had fewer days with flow and reported lower
quality of flow than did younger participants. In addition, women reported higher quality of
flow than did men.
Second, the HLM analyses demonstrated that experiencing flow on a given day was not
associated with either affect or life satisfaction. However, there was a relationship between
the quality of flow experiences and happiness. Higher quality of flow, defined by intense
concentration, loss of self-awareness, and rewarding outcomes, was positively associated
with high arousal positive affect (i.e., feeling peppy, enthusiastic, happy), negatively
associated with low arousal negative affect (i.e., feeling sad and disappointed), and posi-
tively associated with life satisfaction.
In contrast, on the individual level, we observed that a higher number of days with flow
exhibited a negative relationship with both low and high arousal positive affect as well as
with life satisfaction. This unexpected result demonstrates that person-level variables can
operate differently than their lower level counterparts (Snijders and Bosker 1999) and
therefore underscores the importance of examining both daily experiences and the overall
propensity to experience flow. These results suggest that people who had a higher overall
propensity to experience flow, as indicated by a higher number of days with flow, had
lower average levels of positive affect and feelings of life satisfaction. The causality of this
relationship is not clear. It is possible that flow provides more of a boost for positive
feelings when it is a rare occurrence rather than an everyday experience. It is also possible,
however, that individuals consciously or unconsciously use flow as an emotion regulation
technique. That is, seeking out flow experiences could be a strategy that less happy people
use to make themselves feel better. This is consistent with research demonstrating that
people are motivated to regulate their moods, particularly those with low self-esteem or
high self-focus, as they tend to have more negative mood overall (see Norris 1999, for a
review of this literature). Thus, in either case, we cannot argue that flow causes higher or
lower positive affect or life satisfaction; we can only make the observation that people who
reported a higher number of days with flow had lower average levels of positive affect and
life satisfaction, yet participants were happier on the days that they had higher quality of
flow.
The study has several limitations that should be addressed in future research. First, we
took the hedonic approach to psychological well-being and investigated the relationship of
flow to happiness, as measured by positive and negative affect as well as life satisfaction.
Other scholars, however, have advocated the eudaimonic approach to psychological well-
being, stating that actualization of human potentials is a more valid conceptualization of
well-being than subjective experiences of happiness (Ryan and Deci 2001). Future research
should examine the impact of flow experiences on measures of both hedonic and eudai-
monic psychological well-being.
A second potential limitation is that happiness and flow were daily global ratings that
were recalled at the end of each day. One alternative would be to obtain observations of
happiness and flow randomly throughout the day using the ESM. However, as mentioned
previously, older adults may find the ESM technology difficult to use. In addition, flow
experiences might well be fleeting; therefore, random observations would have to be
Flow and Happiness in Later Life
123
frequent in order to capture an experience that is infrequent or brief. A better alternative to
be utilized in future research might be retrospective assessments which tap specific events
or activities. For example, the Day Reconstruction Method (DRM), a retrospective account
of the previous day’s activities, has been shown to be comparable to ESM in assessing how
people spend their time (Kahneman et al. 2004), and might prove to be a good technique
for detecting flow experiences. A superior alternative might be to combine the ESM
technology with the DRM method.
Third, additional limitations stem from the nature of the sample and unmeasured
variables. The study relied on a sample of volunteers, and no data were collected on those
who chose not to participate in the study, so it is possible that the sample is biased by
disproportionally including individuals who were above (or below) average in their hap-
piness levels or in their propensity to experience flow. The sample used in our study was a
particularly high functioning group of White older adults and consisted primarily of
women. A larger, more heterogeneous sample would provide more predictive power that
might clarify the relationship between flow and happiness. Further, the causes and con-
sequences of happiness may vary according to cultural norms and expectations (Diener
et al. 2003a), but the homogenous sample made it impossible to investigate cultural dif-
ferences in the relationship between flow and happiness. In addition, although we screened
participants for medications indicating chronic physical or mental illness, we did not
control for potential confounders like physical functioning (beyond self-assessed health),
cognitive impairment, depressive symptoms, and anxiety, all of which may be related to
both happiness and flow.
Finally, another limitation involves reporting bias that may have resulted from the
structure of the questionnaire. First, as all the participants filled out the happiness measures
before the flow questionnaire, we cannot eliminate the possibility that reporting happiness
may have systematically influenced subsequent flow reports. Second, reporting the pres-
ence of flow may have taken more effort than reporting the absence of flow. Those
participants who were feeling positive enough to fill out the rest of the questionnaire may
have been more likely to report that they had a flow experience. In contrast, participants
who were in a bad mood may have perceived that reporting more information took too
much effort. It is possible that this reporting bias could account for a portion of the positive
association between the quality of flow and happiness. We consider this unlikely, however,
since all of our participants were eager and willing to participate and returned their data to
us reliably. Finally, it is possible that participants completed all the surveys at one time
rather than filling the questionnaires out on subsequent days. We also consider this pos-
sibility highly unlikely, considering the participants would have had to spend more than
two hours filling out the seven questionnaires at one time.
Despite the above limitations, our study provides a foundation for future research on
happiness and flow in late adulthood. Given the strong evidence on the role that happiness
plays in improving health and well-being (Fredrickson 2001; Pressman and Cohen 2005;
Salovey et al. 2000), it is pertinent to consider ways in which older adults can increase the
number of positive experiences in their daily lives. Our results point to the potential
importance of flow in generating happiness in older adults. Our findings suggest that
seeking flow experiences may be a strategy that older adults could use to regulate their
affect, and especially to maximize positive emotions and to minimize low arousal negative
emotions. Based on these findings, future investigations should explore what role flow may
play in self-regulation in old age and whether or not there is a profile of sociodemographic
and personality characteristics which distinguish those who are more likely to use flow to
regulate their emotions. Future research should also examine whether the links between
A. L. Collins et al.
123
flow and happiness vary depending on cultural context. Finally, there should be further
investigation into the relationships between flow, psychological well-being, and physical
functioning in late adulthood in order to provide a better understanding of how older adults
can improve the quality of their lives.
References
Asakawa, K. (2004). Flow experience and autotelic personality in Japanese college students: How do they
experience challenges in daily life? Journal of Happiness Studies, 5(2), 123–154. doi:10.1023/B:JOHS.
0000035915.97836.89.
Baltes, P., & Mayer, K. U. (1999). The Berlin aging study: Aging from 70 to 100. Cambridge, UK:
Cambridge University Press.
Blazer, D. (2003). Depression in late life: Review and commentary. Journal of Gerontology: Medical
Sciences, 58A(3), 249–265.
Bradburn, N. M. (1969). The structure of psychological well-being. Oxford, England: Aldine.
Brandtstaedter, J., & Wentura, D. (1995). Adjustment to shifting possibility frontiers in later life: Com-
plementary adaptive modes. In R. A. Dixon & L. Ba
¨ckman (Eds.), Compensating for psychological
deficits: Managing losses and promoting gains (pp. 83–105). Mahwah, NJ: L. Erlbaum Associates.
Cacioppo, J. T., & Bernston, G. G. (1999). The affect system: Architecture and operating characteristics.
Current Directions in Psychological Science, 8(5), 133. doi:10.1111/1467-8721.00031.
Cacioppo, J. T., Garnder, W. L., & Bernston, G. G. (1999). The affect system has parallel and integrative
processing components: Form follows function. Journal of Personality and Social Psychology, 76(5),
839. doi:10.1037/0022-3514.76.5.839.
Cantril, H. (1999). The pattern of human concerns. New Brunswick, NJ: Rutgers University Press.
Carstensen, L. L. (1995). Evidence for a life-span theory of socioemotional selectivity. Current Directions
in Psychological Science, 4(5), 151–156. doi:10.1111/1467-8721.ep11512261.
Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R. (2000). Emotional experience in everyday
life across the adult life span. Journal of Personality and Social Psychology, 79(4), 644–655. doi:
10.1037/0022-3514.79.4.644.
Csikszentmihalyi, M. (1982). Toward a psychology of optimal experience. In L. Wheeler (Ed.), Review of
personality and social psychology (Vol. 2, pp. 13–36). Beverly Hills: Sage.
Csikszentmihalyi, M. (1988). The flow experience and human psychology. In M. Csikszentmihalyi &
I. Csikszentmihalyi (Eds.), Optimal experience: Psychological studies of flow in consciousness
(pp. 15–35). Cambridge, England: Cambridge University Press.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper Collins.
Csikszentmihalyi, M., & Figurski, T. (1982). Self-awareness and aversive experience in everyday life.
Journal of Personality, 50(1), 15–28. doi:10.1111/j.1467-6494.1982.tb00742.x.
Csikszentmihalyi, M., & Hunter, J. (2003). Happiness in everyday life: The uses of experience sampling.
Journal of Happiness Studies, 4(2), 185–199.
Csikszentmihalyi, M., & LeFevre, J. (1989). Optimal experience in work and leisure. Journal of Personality
and Social Psychology, 56(5), 815–822. doi:10.1037/0022-3514.56.5.815.
Csikszentmihalyi, M., & Rathunde, K. (1993). The measurement of flow in everyday life. Nebraska Sym-
posium on Motivation, 40, 57–97.
Dannefer, D. (1987). Aging as intracohort differentiation: Accentuation, the Matthew effect, and the life
course. Sociological Forum, 2(2), 211–236. doi:10.1007/BF01124164.
Diener, E., & Suh, E. (1997). Measuring quality of life: Economic, social and subjective indicators. Social
Indicators Research, 40, 189–216. doi:10.1023/A:1006859511756.
Diener, E., Oishi, S., & Lucas, R. (2003a). Personality, culture and subjective well-being: Emotional and
cognitive evaluations of life. Annual Review of Psychology, 54, 403–425. doi:10.1146/annurev.psych.
54.101601.145056.
Diener, E., Scollon, C. N., & Lucas, R. E. (2003b). The evolving concept of subjective well-being: The
multi-faceted nature of happiness. In P. T. Costa & I. C. Siegler (Eds.), Advances in cell aging and
gerontology (Vol. 15, pp. 187–219). New York: Elsevier Science.
Feldman Barrett, L., & Russell, J. (1998). Independence and bipolarity in the structure of current affect.
Journal of Personality and Social Psychology, 74(4), 967–984. doi:10.1037/0022-3514.74.4.967.
Flow and Happiness in Later Life
123
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden-and-build
theory of positive emotions. The American Psychologist, 56(3), 218–226. doi:10.1037/0003-066X.
56.3.218.
Gallup Poll. (1988). Omnibus, III.
Greenberg, J., & Pyszczynski, T. (1986). Persistent high self-focus after failure and low self-focus after
success: The depressive self-focusing style. Journal of Personality and Social Psychology, 50(5),
1039–1044. doi:10.1037/0022-3514.50.5.1039.
Gross, J. J., Carstensen, L. L., Pasupathi, M., Tsai, J., Goetestam Skorpen, C., & Hsu, A. Y. C. (1997).
Emotion and aging: Experience, expression, and control. Psychology and Aging, 12(4), 590–599. doi:
10.1037/0882-7974.12.4.590.
Hamer, D. H. (1996). The heritability of happiness. Nature Genetics, 14(2), 125–126. doi:10.1038/ng1096-
125.
Han, S. (1988). The relationship between life satisfaction and flow in elderly Korean immigrants. In M.
Csikszentmihalyi & I. S. Csikszentmihalyi (Eds.), Optimal experience: Psychological studies of flow in
consciousness (pp. 138–149). New York, NY: Cambridge University Press.
Headey, B., & Wearing, A. (1989). Personality, life events, and subjective well-being: Toward a dynamic
equilibrium model. Journal of Personality and Social Psychology, 57(4), 731–739. doi:10.1037/0022-
3514.57.4.731.
Hendricks, J., & Hendricks, C. D. (1986). Aging in mass society: Myths and realities. Boston: Little Brown.
Ingram, R. E. (1990). Self-focused attention in clinical disorders: Review and a conceptual model.
Psychological Bulletin, 107, 156–176. doi:10.1037/0033-2909.107.2.156.
Isaacowitz, D., & Smith, J. (2003). Positive and negative affect in very old age. Journal of Gerontology:
Psychological Sciences, 58B(3), P143–P152.
Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone, A. A. (2004). A survey method for
characterizing daily life experience: The day reconstruction method. Science, 306, 1776–1780. doi:
10.1126/science.1103572.
Kim, J. E., & Moen, P. (2002). Retirement transitions, gender, and psychological well-being: A life-course,
ecological model. Journal of Gerontology, 57(3), 212–222.
Kunzmann, U., Little, T. D., & Smith, J. (2000). Is age-related stability of subjective well being a paradox?
Cross-sectional and longitudinal evidence from the Berlin aging study. Psychology and Aging, 15(3),
511–526. doi:10.1037/0882-7974.15.3.511.
Labouvie-Vief, G., Hakim-Larson, J., DeVoe, M., & Schoeberlein, S. (1989). Emotions and self-regulation:
A life span view. Human Development, 32(5), 279–299.
Lawton, M. P. (1989). Environmental proactivity and affect in older people. In S. Spacapan & S. Oskamp
(Eds.), The social psychology of aging. Claremont symposium on applied social psychology (pp. 135–
163). Thousand Oaks, CA: Sage.
Lawton, M. P., DeVoe, M. R., & Parmelee, P. (1995). Relationship of events and affect in the daily life of an
elderly population. Psychology and Aging, 10(3), 469–477. doi:10.1037/0882-7974.10.3.469.
Lucas, R. E., Clark, A. E., Georgellis, Y., & Diener, E. (2003). Reexamining adaptation and the set point
model of subjective well-being: Reactions to changes in marital status. Journal of Personality and
Social Psychology, 84(3), 527–539. doi:10.1037/0022-3514.84.3.527.
Mroczek, D. K., & Kolarz, C. M. (1998). The effect of age on positive and negative affect: A developmental
perspective on subjective well-being. Journal of Personality and Social Psychology, 75(5), 1333–1349.
doi:10.1037/0022-3514.75.5.1333.
Norris, W. N. (1999). The mood system. In D. Kahneman, E. Diener, & N. Schwarz (Eds.), Well-being: The
foundations of hedonic psychology (pp. 169–189). New York: Russell Sage.
Pavot, W., & Diener, E. (2004). The subjective evaluation of well-being in adulthood: Findings and
implications. Ageing International, 29(2), 113–135. doi:10.1007/s12126-004-1013-4.
Pavot, W., Diener, E., Colvin, C. R., & Sandvik, E. (1991). Further validation of the Satisfaction with Life
Scale: Evidence for the cross-method convergence of well-being measures. Journal of Personality
Assessment, 57(1), 149–161. doi:10.1207/s15327752jpa5701_17.
Pinquart, M., & Schindler, I. (2007). Changes of life satisfaction in the transition to retirement: A latent class
approach. Psychology and Aging, 22(3), 442–455. doi:10.1037/0882-7974.22.3.442.
Pressman, S. D., & Cohen, S. (2005). Does positive affect influence health? Psychological Bulletin, 131(6),
925–971. doi:10.1037/0033-2909.131.6.925.
Raudenbush, S., & Bryk, A. (2002). Hierarchical linear models: Applications and data analysis methods.
Thousand Oaks, CA: Sage.
Rowe, J. W., & Kahn, R. L. (1997). Successful aging. The Gerontologist, 37(4), 433–440.
A. L. Collins et al.
123
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on hedonic and
eudaimonic well-being. Annual Review of Psychology, 52, 141–166. doi:10.1146/annurev.psych.52.1.
141.
Ryff, C. D. (1989). In the eye of the beholder: Views of psychological well-being among middle-aged and
older adults. Psychology and Aging, 4(2), 195–210. doi:10.1037/0882-7974.4.2.195.
Ryff, C. D., Dienberg Love, G., Urry, H. L., Muller, D., Rosenkranz, M. A., Friedman, E. M., et al. (2006).
Psychological well-being and ill-being: Do they have distinct or mirrored biological correlates?
Psychotherapy and Psychosomatics, 75(2), 85–95. doi:10.1159/000090892.
Salovey, P., Rothman, A. J., Detweiler, J. B., & Steward, W. T. (2000). Emotional states and physical health.
The American Psychologist, 55(1), 110–121. doi:10.1037/0003-066X.55.1.110.
Smith, D. (2003). The older population in the United States: March 2002. U.S. Census Bureau Current
Population Reports, P20-546. Washington, DC.
Snijders, T., & Bosker, R. (1999). Multilevel analysis. London: Sage.
Spybrook, J., Raudenbush, S. W., Liu, X., & Congdon, R. (2006). Optimal design for longitudinal and
multilevel research: Documentation for the ‘‘Optimal Design’’ software. http://sitemaker.umich.edu/
group-based/files/odmanual-20060919-v176.doc.
Stacey, C. A., & Gatz, M. (1991). Cross-sectional age differences and longitudinal change on the Bradburn
Affect Balance Scale. Journal of Gerontology, 46, 76–78.
Voelkl, J. E. (1990). The challenge skill ratio of daily experiences among older adults residing in nursing
homes. Therapeutic Recreation Journal, 24(2), 7–17.
Voelkl, J., & Ellis, G. (1998). Measuring flow experiences in daily life: An examination of the items used to
measure challenge and skill. Journal of Leisure Research, 30(3), 380–389.
Wang, M. (2007). Profiling retirees in the retirement transition and adjustment process: Examining the
longitudinal change patterns of retirees’ psychological well-being. The Journal of Applied Psychology,
92(2), 455–474. doi:10.1037/0021-9010.92.2.455.
Wells, A. J. (1988). Self-esteem and optimal experience. In M. Csikszentmihalyi & I. Csikszentmihalyi
(Eds.), Optimal experience: Psychological studies of flow in consciousness (pp. 327–341). Cambridge,
England: Cambridge University Press.
Witt, P., & Ellis, G. (1987). The leisure diagnostic battery users manual. State College, PA: Venture
Publishing.
World Values Study Group. (1994). World values survey, 1981–1984 and 1990–1993. Ann Arbor, MI:
Institute for Social Research, University of Michigan.
Flow and Happiness in Later Life
123
