Content uploaded by Marisa Salanova
Author content
All content in this area was uploaded by Marisa Salanova
Content may be subject to copyright.
MARISA SALANOVA, ARNOLD B. BAKKER and SUSANA LLORENS
FLOW AT WORK: EVIDENCE FOR AN UPWARD SPIRAL
OF PERSONAL AND ORGANIZATIONAL RESOURCES*
ABSTRACT. The present 2-wave study among 258 secondary school teachers
investigates the relationship between personal and organizational resources on
the one hand, and work-related flow on the other hand. On the basis of
Hobfoll
’
s (1988) conservation of resources theory, Bandura’ social cognitive
theory (1997; 2001), and Fredrickson’s (1998) ‘‘broaden-and-build’’ theory of
positive emotions, we formulated two hypotheses: (1) personal resources (i.e.,
self-efficacy beliefs) and organizational resources (including social support
climate and clear goals) facilitate work-related flow (work absorption, work
enjoyment, and intrinsic work motivation); and (2) work-related flow has a
positive influence on personal and organizational resources. The results of a
series of structural equation modeling analyses offer clear support for both
hypotheses. The theoretical and practical implications of these findings are
discussed.
KEY WORDS: flow, organizational resources, personal resources, positive
psychology, teachers.
INTRODUCTION
What is a good work life is a basic topic for workers and organi-
zations nowadays, being of fundamental relevance for psycholog-
ical research as well. Recently, Luthans (2002a, b) has noted the
need for Positive Organizational Behavior (POB) research, de-
fined as the study and application of positively oriented human re-
source strengths and psychological capacities that can be measured,
developed, and effectively managed for performance improvement
in today’s workplace (Luthans, 2003, p. 179). Wright (2003) has
argued that the mission of POB must also include the pursuit of
* This research was supported by a grant from the Bancaixa Foundation
(#11232.01/1) and the Spanish Ministry of Science & Technology (CICYT
#SEC2000-1031).
Journal of Happiness Studies (2006) 7:1–22 Springer 2006
DOI 10.1007/s10902-005-8854-8
employee happiness and health as viable goals in themselves. The
present two-wave study among school teachers will focus on the
experience of flow at the workplace. We concur with Wright
(2003) that the pursuit of flow is a viable goal in itself, and will
examine its dependence on and value for the quality of employ-
ees’ working environment. More specifically, the current study
will examine whether personal and organizational resources facil-
itate flow at work, and whether employees who experience flow
mobilize more resources over time.
Flow at Work
Csikszentmihalyi (1997, p. 29) has described the flow experience
as: ‘‘one that many people have used to describe the sense of
effortless action they feel in moments that stand out as the best
in their lives.’’ In this state, people are intensely involved in an
activity and so nothing else seems to matter. In addition to the
pleasure in the activity and the intrinsic interest to continue
doing it, the total immersion in an activity seems to be a central
aspect of the flow experience (Csikszentmihalyi et al., 1993; Ellis
et al., 1994; Ghani and Deshpande, 1994; Larson and Richards,
1994). On the basis of these previous studies, Bakker (2005)
applied the concept of flow to the work situation, and defined
flow as a short-term peak experience at work that is character-
ized by absorption, work enjoyment and intrinsic work motiva-
tion. Absorption refers to a state of total concentration, whereby
employees are totally immersed in their work. Time flies, and
they forget everything else around them (cf. Csikszentmihalyi,
1990). Employees who enjoy their work and feel happy make a
very positive judgment about the quality of their working life
(cf. Veenhoven, 1984). This enjoyment or happiness is the out-
come of cognitive and affective evaluations of the flow experi-
ence (cf. Diener, 2000; Diener and Diener, 1996). Finally,
intrinsic work motivation refers to the need to perform a certain
work-related activity with the aim of experiencing the inherent
pleasure and satisfaction in the activity (cf. Deci and Ryan,
1985). Intrinsically motivated employees are continuously inter-
ested in the work they are involved in (Harackiewicz and Elliot,
1998). Employees who are motivated by the intrinsic aspects of
MARISA SALANOVA ET AL.
2
their work tasks want to continue their work; they are fasci-
nated by the tasks they perform (Csikszentmihalyi, 1997).
Do Resources Lead to Flow at Work?
Bakker’s (2005) study among music teachers showed that
organizational resources can be important antecedents of flow
experiences among teachers and their students. He found that
teachers working at schools with high levels of autonomy, social
support, supervisory coaching, and feedback were most likely to
experience flow at work. Demerouti et al., (2001) defined job
resources as those physical, psychological, social, or organiza-
tional aspects of the job that either/or: (1) are functional in
achieving work goals; (2) reduce job demands and the associ-
ated physiological and psychological costs; (3) stimulate per-
sonal growth and development. Examples of job and
organizational resources are social support from colleagues, per-
formance feedback, skill variety, job control, and supervisory
coaching. These resources have motivational potential because
they make employees’ work meaningful, hold them responsible
for work processes and outcomes, and provide them with infor-
mation about the actual results of their work activities (cf.
Bakker et al., 2003; Hackman and Oldham, 1980). Although we
could not find other studies relating organizational resources to
work-related flow, several recent studies have provided addi-
tional evidence for the motivational potential of resources. In
their study among human service professionals (including con-
sultants, nurses, and teachers), Bakker et al. (2004) have shown
that resources foster work engagement, which, in turn, is pre-
dictive of (colleagues’ ratings of) organizational citizenship
behavior. Similarly, Bakker et al. (2003) have shown that pro-
duction workers’ organizational resources foster organizational
commitment, which, in turn, causes reduced absence frequency.
Furthermore, in a series of studies in several occupational set-
tings, Salanova and her colleagues have shown that organiza-
tional resources can be important predictors of work
engagement, which, in turn, is predictive of important organiza-
tional outcomes including service climate (Salanova et al., in
press), and group performance (Salanova et al., 2003). Finally,
Houkes (2002) included several job resources in her longitudinal
RESOURCES AND FLOW AT WORK 3
research among bank employees and teachers, and found evi-
dence for a causal relationship between the ‘‘motivating poten-
tial score’’ (an additive index, including skill variety, task
identity, task significance, autonomy, and job feedback) and
intrinsic work motivation. By contrast, a lack of organizational
resources has a detrimental effect on workers’ motivation and
performance (e.g., Wong et al., 1998), since it precludes actual
goal accomplishment, and undermines employees’ learning
opportunities (e.g., Kelly, 1992).
This latter reasoning is consistent with Conservation of
Resources (COR) theory (Hobfoll, 1989, 1998, 2002). Accord-
ingly, people seek to obtain, retain, and protect resources and
stress occurs when resources are threatened with loss or lost, or
when individuals fail to gain resources after substantive
resource investment. Thus, COR-theory also places the acquisi-
tion and facilitation of resources as a central motivational con-
struct. It is assumed that individuals seek to acquire and
maintain resources, including objects (e.g., a home, clothes,
food), personal characteristics (e.g., self-esteem), conditions
(e.g., being married or living with someone who provides social
support), and energies (e.g., time, money, and knowledge).
However, the two principal types of resources that have been
examined within the COR framework are personal and psycho-
social resources (Hobfoll et al., 2003). The present study will fo-
cus specifically on self-efficacy (i.e., a personal resource), in
addition to organizational resources.
Personal resources are aspects of the self that are generally
linked to resilience. The best studied of these pertain to individ-
uals’ sense of their competence to successfully control and
impact their environment (Hobfoll et al., 2003). Self-efficacy has
shown its power as a buffer in stress processes, and it has also
been related to better health, better self-development and great-
er social integration (Bandura, 1997, 1999, 2001). In addition,
there is considerable evidence regarding the positive effects of
self-efficacy on performance and well-being in different domains
such as the workplace, school, and sports (Bandura, 1999,
2001). For example, research in the domain of work shows that
high levels of efficacy beliefs have a positive impact on employ-
ee well-being (Grau et al., 2001) and work engagement
MARISA SALANOVA ET AL.
4
(Salanova et al., 2003), and can buffer the negative impact of
job demands on burnout (Salanova et al., 2000; 2002). Accord-
ing to Bandura (1997, 2001), efficacy beliefs contribute to moti-
vation in several ways. Efficacy beliefs influence (a) the
challenges people pursue, (b) the effort they expend, and (c)
their perseverance in the face of obstacles.
So far, self-efficacy (as a personal resource) can be seen as
one of the resources people want to protect (a very important
one, which has received considerable research attention). The
organizational resources we included are yet other examples of
resources people are motivated to protect. In other words: all
resources we studied are part of the resource reservoir people
possess. This also means that COR theory can be seen as the
overall framework we used for our study.
Therefore, we hypothesize that, in addition to organizational
resources, self-efficacy beliefs may also enhance the experience
of flow at work:
Hypothesis 1: Organizational resources (e.g., social support
and clear goals) and personal resources (i.e., self-efficacy beliefs)
have a positive influence on the experience of work-related flow
(i.e., absorption, work enjoyment and intrinsic work motiva-
tion) (causation hypothesis)
Does Flow at Work Lead to an Accumulation of Organizational
and Personal Resources?
Is it also possible that flow experiences lead to more organiza-
tional and personal resources? Some recent studies indeed found
evidence for reversed causal relationships between organizational
resources and employee psychological well-being. In her study
among bank employees and teachers, Houkes (2002) found evi-
dence for a reversed causal relationship between the motivating
potential score and intrinsic work motivation. Furthermore,
Wong et al. (1998) reported that time 1 job satisfaction pre-
dicted several organizational resources assessed at time 2
(2 years later), including autonomy, task identity, skill variety,
task significance and feedback. In a similar vein, in a cohort
study among a heterogeneous sample of Dutch employees,
De Lange (2003) found a reversed causal relationship between
job control and job satisfaction.
RESOURCES AND FLOW AT WORK 5
Taken together, these findings suggest that work motivation
and job satisfaction can both be outcomes as well as predictors
of organizational resources, such that enhanced well-being
results in more favorable organizational resources over time.
This is consistent with Hobfoll’s (1989, 2002) claim that because
resources are valued either in their own right or because they
enable the acquisition or preservation of other valued resources,
people are motivated to create resources. Once resources are
obtained, people are motivated to protect them. In addition, the
hypothesis that flow at work may lead to more personal and
organizational resources over time is consistent with Fredrick-
son’s (1998, 2002) ‘‘broaden-and-build’’ theory of positive
emotions. Accordingly, positive emotions broaden people’s
momentary thought-action repertories and build their enduring
personal resources. Research has indeed shown that positive
emotions such as joy, happiness, and interest have long-term
adaptive benefits because broadening builds enduring personal
resources including physical, intellectual, social and psychologi-
cal resources (Fredrickson and Branigan, 2005; Fredrickson,
1998, 2001). Moreover, research with the broaden-and-build the-
ory showed that momentary experiences of positive emotions
can build enduring psychological resources and trigger upward
spirals toward emotional well-being. Thus, positive emotions not
only make people feel good at the moment, but also feel good in
the future (Fredrickson and Joiner, in press). The current study
will focus on the building effect (i.e., the second hypothesis in
the theory). Accordingly, positive experiences (such as flow at
work) build people’s enduring personal resources. Indirect evi-
dence consistent with this building hypothesis can be drawn
from correlational and experimental studies that link positive
states, traits and behaviors with physical, intellectual and social
resources (Boulton and Smith, 1992; Bryan, Mathur & Sullivan,
1996; Caro, 1988; Hazen and Durrett, 1982).
Furthermore, research on Social Cognitive Theory (Bandura,
1997, 1999, 2001) shows that experiencing positive emotions
increases self-efficacy as a personal resource. When people judge
their capabilities, they rely partly on bodily information con-
veyed by emotional states. According to Bandura, emotions are
especially relevant when it comes to stress and health: negative
MARISA SALANOVA ET AL.
6
mood diminished perceived efficacy, whereas positive mood
enhances it. As is a tradition in psychological research, most
studies investigated the impact of negative emotions (i.e., anxi-
ety) on self-efficacy (Bandura, 1997). Studies relating positive
states (including flow) to self-efficacy are sparse, but there is evi-
dence showing that positive states influence self-efficacy, Salano-
va et al. (2002) showed that positive states at work such as
enthusiasm, satisfaction and comfort influenced self-efficacy
through work engagement in different occupational groups. Re-
sults indicated that the more positive experiences, the more
engagement, which, in turn, predicted future self-efficacy.
On the basis of these theories and findings, we formulated
our second hypothesis. The two hypotheses are graphically
depicted in Figure 1.
Hypothesis 2: The experience of work-related flow (i.e.,
absorption, work enjoyment and intrinsic work motivation) has
a positive influence on organizational and personal resources
(reversed causation hypothesis).
METHOD
Participants and Procedure
A follow-up study with two waves was carried out among
Spanish secondary school teachers. At the beginning of the
Figure 1. The upward spiral model of resources and flow at work.
RESOURCES AND FLOW AT WORK 7
academic year, a letter was sent to 50 secondary schools
explaining the goal of the research. Self report questionnaires
including scales to measure the main variables of the current
study plus other scales related to psychological well-being were
distributed among 600 secondary teachers from these schools
and were sent back through surface mail to the university. In
total, 484 respondents from 34 schools returned the question-
naire (81% response rate). Eight months later, at the end of the
academic year, identical questionnaires were distributed among
the same schools. After deletion of missing cases, it turned out
that 258 teachers (57% women, 43% men) from 24 schools had
completed both questionnaires, and their scores could be used
in the longitudinal analyses. Thus, 57% of the teachers who
participated at T1 also participated at T2. The mean age of the
sample was 40 years (SD = 7.01).
In order to test whether the drop-outs differed from the
panel group, we compared the T1 background variables of both
groups (i.e., age, gender, type of school (private versus public),
teaching experience, and organizational tenure). Results from
ANOVAs and v
2
analyses showed that there were no significant
differences between the groups regarding background variables.
We therefore concluded that the panel group does not differ
from the drop-outs in terms of background variables.
Measures
Organizational Resources were measured with the short version
of the FOCUS Organizational Culture Questionnaire
(Gonza
´lez-Roma et al., 1995), including twelve items. Four re-
sources were assessed with three items each, namely: (1) social
support orientation or the extent to which there are kindly and
supportive relationships among organizational members (e.g.,
‘‘How many people with personal problems are helped?’’), inno-
vation orientation or the extent to which there is openness to
new ideas and projects (e.g., ‘‘How many people are trying out
new ways of working?’’), rules orientation or the extent to
which organizational members’ behavior is regulated by formal
norms and rules (e.g., ‘‘How often are jobs performed accord-
ing to defined procedures?’’), and goals orientation or the extent
to which activities and behaviors are oriented towards the
MARISA SALANOVA ET AL.
8
attainment of previously established objectives (e.g., ‘‘How of-
ten does management specify the targets to be attained?’’). The
answer categories ranged from 1(‘‘never/nobody’’) to 6 (‘‘al-
ways/most of the people’’). Cronbach’s alpha of each scale is
shown in Table I.
Personal Resources was measured as efficacy beliefs. It was
measured with Schwarzer’s (1999) 10-item generalized self-
efficacy scale. However, consistent with social cognitive theory,
the scale was slightly adapted for use in specific domain (i.e.,
the workplace, which is schools in our study). One example
item is: ‘‘I will be able to solve difficult problems in my work, if
I invest the necessary effort.’’ The anchors ranged from 1
(‘‘never’’) to 6 (‘‘always’’). Self-efficacy was included in the
structural equation model as a latent variable, indicated by two
reliable halves. Cronbach’s alpha of the scales as measured at
T1 and T2 is shown in Table I.
Flow at work was assessed with the WOrk-reLated Flow scale
(WOLF; Bakker, 2001). The WOLF includes thirteen items
measuring absorption (4 items), work enjoyment (4 items), and
intrinsic work motivation (5 items). Examples are: ‘‘When I am
working, I forget everything else around me’’ (absorption),
‘‘When I am working very intensely, I feel happy’’ (work enjoy-
ment), and ‘‘I get my motivation from the work itself, and not
from the rewards for it’’ (intrinsic work motivation). The partic-
ipants were asked to indicate how often they had each of the
experiences during the preceding week (0 = never, 6 = every
day).
Data Analysis
The data were analyzed with structural equation modeling tech-
niques using the AMOS software package (Arbuckle, 1997).
Each of the model components was included as a latent factor
in the model, and was operationalized by the subscales intro-
duces above as observed, indicator variables. Specifically, orga-
nizational resources was indicated by four variables (social
support orientation, innovation orientation, rules orientation
and goals orientation), flow at work was indicated by the three
flow scales of the WOLF (absorption, work enjoyment, and
RESOURCES AND FLOW AT WORK 9
TABLE I
Means, standard deviations (SD), Cronbach’s Alpha (on the diagonal), and correlations for the study variables, N= 258
Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Time 1 Variables
1. Support Orientation 3.71 1.07 0.86
2. Innovative
Orientation
3.47 1.20 0.65*** 0.87
3. Rules Orientation 3.93 1.09 0.29*** 0.38*** 0.68
4. Goal Orientation 3.50 1.21 0.51*** 0.63*** 0.49*** 0.82
5. Absorption 3.57 1.03 0.05 0.09 0.05 0.19** 0.78
6. Work Enjoyment 4.44 1.08 0.18** 0.19** 0.09 0.24*** 0.54*** 0.90
7. Intrinsic Work
Motivation
3.91 1.25 0.21** 0.20** 0.14* 0.32*** 55*** 0.83*** 0.89
8. Self-efficacy-
indicator 1
4.02 0.94 0.19** 0.19** 0.10 0.23*** 0.24*** 0.48*** 0.43** 0.88
9. Self-efficacy-
indicator 2
3.84 0.96 0.20** 0.19** 0.13* 0.22*** 0.28*** 0.45***. 0.41** 0.70** 0.90
Time 2 Variables
10. Support
Orientation
4.22 1.17 0.09 0.06 0.12* 0.18** 0.15* 0.21** 0.28*** 0.25*** 0.26*** 0.86
11. Innovative
Orientation
3.32 1.24 0.52*** 0.62*** 30*** 58*** 0.19** 0.23*** 0.26*** 0.21*** 0.24*** 0.15* 0.91
12. Rules Orientation 3.71 1.14 0.30*** 0.29*** 0.54*** 0.46*** 0.12* 0.14* 0.19** 0.13* 0.09 0.08 0.43*** 0.67
13. Goal Orientation 3.37 1.24 0.43*** 0.49*** 0.41*** 0.69*** 0.23*** 0.21*** 0.30*** 0.29*** 0.26*** 0.21** 0.74*** 0.54*** 0.88
14. Absorption 3.50 1.01 0.05 0.05 0.07 0.17** 0.72*** 0.45*** 0.45*** 0.14* 0.19** 0.16** 0.22*** 0.13* 0.24*** 0.82
15. Work Enjoyment 4.41 1.01 0.21** 0.25*** 0.12* 0.30*** 0.44*** 0.72*** 0.68*** 0.39*** 0.43*** 0.29*** 0.34*** 0.26*** 0.35*** 0.47*** 0.89
16. Intrinsic Work
Motivation
3.81 1.19 0.16** 0.21** 0.16* 0.29*** 0.44*** 0.64*** 0.75*** 0.35*** 0.35*
k
* 0.29*** 0.34*** 0.24*** 0.35*** 0.52*** 0.79*** 0.87
17. Self-efficacy-
indicator l
4.08 0.83 0.16** 0.16** 0.09 0.24*** 0.31*** 0.43*** 0.41*** 0.58*** 0.27*"* 0.29*** 0.33*** 0.19** 0.38*** 0.31*** 0.35*** 0.45*** 0.89
18. Self-efficacy-
indicator 2
3.92 0.89 0.18** 0.21** 0.11 0.21** 0.21** 0.37*** 0.36*** 0.54*** 0.61*** 0.28*** 0.32*** 0.14* 0.35** 0.22*** 0.53*** 0.43*** 0.84*** 0.93
Note. All correlations are significant at the ***p< 0.001 level, **p< 0.01, *p< 0.05.
MARISA SALANOVA ET AL.
10
intrinsic work motivation), and self efficacy was indicated by
two reliable halves of Schwarzer’s modified scale.
A number of competing structural equation models were
fitted to the data in several steps. First, a model without cross-
lagged structural paths but with temporal stabilities and syn-
chronous correlations (Model 1) was specified. The temporal
stabilities were specified as correlations between the constructs
for each possible pair of measurement waves. This model esti-
mates therefore the total stability coefficient between measure-
ment waves 1 and 2, without decomposing the variance into
constituent paths (direct and indirect effects) (see Pitts et al.,
1996). Second, this stability model was compared with three
more complex models that were nearest in likelihood to the
hypothesized structural model: (a) Model 2: a model that is
identical to Model 1 but also includes cross-lagged structural
paths from Time 1(T1) organizational resources and T1 per-
sonal resources to T2 work-related flow. This model is the cau-
sality model; (b) Model 3: a model that is identical to Model 1
but also includes cross-lagged structural paths from T1 flow to
T2 organizational resources and T2 personal resources. This is
the reversed causation model; (c) Model 4: a model that includes
reciprocal relationships between organizational resources, per-
sonal resources, and work-related flow. This model includes all
paths of Model 2 and 3, and is called the reciprocal model.
For all models, the measurement errors of the same indica-
tors (i.e., subscales) collected at different time points were
allowed to covary over time (e.g., a covariance is specified be-
tween the measurement error of work enjoyment as measured at
T1 and the measurement error of this scale as measured at T2).
While in cross-sectional data measurement errors should gener-
ally not covary, in longitudinal measurement models the errors
of measurement corresponding to the same indicator should co-
vary over time. According to Pitts et al. (1996), this specifica-
tion of covariance between errors of measurement accounts for
the systematic (method) variance associated with each specific
indicator.
The various nested models were compared by means of the
v
2
difference test (Jo
¨reskog and So
¨rbom, 1986). Besides the chi-
square statistic, the analysis assessed the goodness-of-fit index
RESOURCES AND FLOW AT WORK 11
(GFI) and the root mean square error of approximation
(RMSEA), Furthermore, AMOS provides several fit indices that
reflect the discrepancy between the hypothesized model and the
baseline, Null model. In the present analyses, the comparative
fit index (CFI) and the non-normed fit index (NNFI) are uti-
lized. Marsh, Balla and Han (1996) recommended the latter two
indices, because they are less dependent on sample-size than the
v
2
statistic and the GFI. In general, models with fit indi-
ces > 0.90 and RMSEA < 0.08 indicate a good fit (Hoyle,
1995). Preliminary analyses showed that the demographics
(included as covariates) were not systematically related to the
model variables, and did not modify the results of the model
testing. Therefore, to facilitate model estimation, the
demographics were excluded from all further analyses.
RESULTS
Descriptive Statistics
Prior to the model testing, the means, standard deviations,
Cronbach’s alpha coefficients and bivariate correlations (includ-
ing test-retest correlations) were computed (see Table I). As can
be seen from the table, all variables had an alpha coefficient
higher than 0.70 (with the exception of rules orientation which
had an alpha coefficient of 0.68 and 0.67 at T1 and T2), and
test–retest reliabilities of at least 0.54. The highest test–
retest reliabilities resulted for self-efficacy followed by innova-
tion orientation. This means that self-efficacy and innovation
orientation are relatively stable over time.
Model Testing
Table II displays the overall fit indices of the competing models.
In general, all models indicate a good fit since all fit indices are
equal to or higher than 0.90, the RMSEA is smaller than 0.08,
and the ratio between the v
2
statistic and the number of degrees
of freedom is relatively low. We will first concentrate on the
model comparisons.
The causality model (M2) proved to be superior to the stability
model (Ml), Delta v
2
(2) = 35.29, p< 0.001. This suggests that
MARISA SALANOVA ET AL.
12
the inclusion of cross-lagged paths from organizational resources
and personal resources to flow is substantial. Additionally, the
reversed causality model (M3) fitted significantly better to the
data than the stability model, Delta v
2
(2) = 33.49, p< 0.001.
This indicates that the model with the cross-lagged path from T1
flow to T2 organizational resources and T2 personal resources
showed a better fit to the data than the model including only tem-
poral stabilities and synchronous correlations (M1).
The v
2
difference test regarding the stability model vis-a
`-vis
the reciprocal model (M4; see Table II) revealed that the addi-
tion of reciprocal effects significantly improved the stability
model, Delta v
2
(4) = 103.10, p< 0.001. Moreover, the model
with the cross-lagged reciprocal relationships among the vari-
ables (M4) resulted in a significantly better fit to the data than
the causality model (M2) and the reversed causality model
(M3). The results of the v
2
difference tests for both comparisons
(M2 versus M4 and M3 versus M4) are Delta v
2
(2) = 67.81,
p< 0.001, and Delta v
2
(2) = 69.61, p< 0.001, respectively.
This means that the theoretical model including cross-lagged
reciprocal relationships between organizational resources, per-
sonal resources, and work-related flow fits best to the data.
We will now discuss the specific structural relationships that
resulted from these models. First of all, it is important to note
that all manifest variables loaded significantly on the intended
TABLE II
Goodness-of-fit indices for the alternative resources – flow models, N= 258
Model v
2
df pGFI RMSEA NNFI CFI
M1. Stability Model 250.88 117 0.001 0.90 0.07 0.95 96
M2. Causality Model
(JR, PR fiFlow)
215.59 115 0.001 0.91 0.06 0.96 0.97
M3. Reversed Causality Model
(Flow fiJR, PR)
217.39 115 0.001 0.92 0.06 0.96 0.97
M4. Reciprocal Model 147.78 113 0.02 0.94 0.04 0.99 0.99
Null Model 3776.16 153 0.001 0.26 0.30 – –
Note: v
2
= chi-square; df = degrees of freedom; GFI = goodness-of-fit
index; RMSEA = root mean square error of approximation; NNFI = Non-
nonmed fit index; CFI = comparative fit index. JR = Job Resources,
PR = Personal Resources.
RESOURCES AND FLOW AT WORK 13
latent factors. Inspection of the output revealed that all indica-
tors of organizational resources had loadings on the intended
latent factor higher than 0.50, both at T1 and T2. Furthermore,
a both waves of measurement, the loadings of absorption, work
enjoyment and intrinsic work motivation on the flow factor
were higher than 0.67, whereas the loadings of the two self-
efficacy indicators were higher than 0.92. Second, the autocorre-
lations between the two waves were 0.71 for job resources, 0.60
for flow, and 0.47 for self-efficacy.
Hypothesis 1 asserted that job and personal resources would
have lagged positive effects on work-related flow. The model
that includes these causal relationships, the reciprocal model
(M4), resulted in significant lagged, and positive effects of TI
job resources on T2 flow (b= 0.26, t= 4.07, p< 0.001), as
well as a positive effect of T1 personal resources on T2 flow
(b= 0.41, t= 6.20, p< 0.001). These findings clearly support
our first hypothesis.
Hypothesis 2 stated that T1 flow would have a lagged posi-
tive effect on T2 organizational resources and on T2 personal
resources. The final reciprocal model also resulted in significant
reversed causal structural relationships. Specifically, the rela-
tionships were as follows: T1 flow – T2 organizational resources
b= 0.33 (t= 4.72, p< 0.001), T1 flow T2 personal resources
b= 0.47 (t= 7.17, p< 0.001). These findings clearly support
our second hypothesis. Thus, the results from Model 4 (includ-
ing the reciprocal relationships) showed that both causal and
reversed causal relationships were simultaneously active. The
significant paths of the reciprocal model are graphically dis-
played in Figure 2. The hypothesized predictors at TI explained
11% of the variance in T2 resources, 28% of the variance in T2
flow, and 22% of the variance in T2 self-efficacy.
DISCUSSION
In the current study, the phenomenon of flow at work was
investigated among teachers. The main research questions were
whether flow is facilitated over time by the presence of personal
and organizational resources, and whether flow has a positive
MARISA SALANOVA ET AL.
14
influence on the availability of resources over time. In this study,
flow was conceptualized as a short-term peak experience at work
that is characterized by absorption, work enjoyment and intrin-
sic work motivation. Absorption refers to a state of total con-
centration, whereby employees are totally immersed in their
work. Time flies, and they forget everything else around them.
Enjoyment is a positive emotion of feeling happy and endorsing
a very positive judgment about the quality of the working life.
Finally, intrinsic work motivation refers to the need to perform
a certain work-related activity with the aim of experiencing the
inherent pleasure and satisfaction in the activity.
On the basis of a brief literature review, it was predicted that
flow develops over time when personal and organizational
resources are sufficiently available, because these positive char-
acteristics foster flow experiences. In addition, we hypothesized
a reversed causal relationship between resources and work-
related flow, i.e., that flow would predict future resources. The
results of SEM-analyses -- using a cross-lagged panel design
Figure 2. Standardized solution (maximum likelihood estimates) of the up-
ward spiral model of resources and flow at work, N= 258.
RESOURCES AND FLOW AT WORK 15
(N= 258 secondary school teachers) supported both hypothe-
ses. More specifically, the results showed that the theoretical
model including cross-lagged reciprocal relationships between
resources and work-related flow fits best to the empirical data.
This means that both causal and reversed causal relationships
were simultaneously active in the reciprocal relationship be-
tween (organizational and personal) resources and work-related
flow.
Organizational resources -- a combination of different types
of organizational climate orientation indicators such as social
support (e.g., people help each other mutually, there are a
good relationship between the co-workers), innovation (e.g.,
teachers can give suggestions to improve the quality of work,
and they can put forward new ideas to improve the work), rules
(e.g., the work that the teacher must do is plenty of norms
and the decisions about the work process is made by the
supervisors and goals, e.g., objectives are clearly defined) --
had a positive influence on the occurrence of flow among
teachers over the time. Simultaneously, personal resources --
operationalized as strong beliefs about one’s own competence
at work -- also fostered these flow experiences over time, So
far, our first hypothesis was confirmed. These findings are con-
sistent with previous research on the motivational power of re-
sources at the work place (Bakker et al., 2003, 2004;
Demerouti et al., 2002; Hackman and Oldham, 1980; Houkes,
2002; Salanova et al., 2003, 2004). Furthermore, the current
study expands research on COR theory, showing that people
seek to obtain, retain and protect resources in order to avoid
stress and be psychologically healthy, including positive experi-
ences (i.e., flow). COR theory evidences that the acquisition
and facilitation of resources is a central motivational con-
struct. In addition, this study also agrees with predictions
from Social Cognitive Theory, which assumes that self-efficacy
facilitates well-being (Bandura, 1997, 1999, 2003). In this
sense, we found that self-efficacy is a powerful personal
resource to build future positive experiences such as being
immersed in one’s work, to enjoy working, and to feel intrinsi-
cally motivated at work. Feeling competent in the present,
seems to predict being in flow in the future.
MARISA SALANOVA ET AL.
16
The second hypothesis was that the experience of flow
(absorption, work enjoyment and intrinsic work motivation) has
a positive influence on organizational and personal resources.
Here and based on the literature, we assumed a reversed causal
effect of work-related flow on organizational and personal
resources. The results confirmed our hypothesis. More specifi-
cally, experiences of flow at work at the present influenced the
gain of organizational and personal resources in the future.
These findings are consistent with previous research on COR
theory (Hobfoll, 1988), broaden-and-build theory (Fredrickson,
1998) and social cognitive theory (Bandura, 1997). For example,
as resources are valuable in their own right, people in one way
or another are motivated to create or increase their resources
(Hobfoll, 1989, 2002), and positive experiences are powerful
mechanism to increase and build organizational and personal
(i.e., self-efficacy) resources over time (Bandura, 1997; Fredrick-
son, 2001). In our study, we found that flow as a positive expe-
rience at work influenced the organizational and personal
resources among secondary school teachers over time. Further-
more, we found that flow predicted self-efficacy slightly more
strongly than self-efficacy predicted flow. These results fit with
what Hobfoll et al. (2003) found for sense of mastery as well.
Taken together, and perhaps even more interesting, our
results showed that there exist reciprocal relationships between
resources and flow. These results are supporting in a way the
predicted upward spiral in which positive emotions are building
resources that in turn influence positive emotions (Fredrickson,
2002). Findings of the current study showed that such positive
work-related experiences as flow build organizational and per-
sonal resources, and that sometimes these positive experiences
are reciprocally influenced by these resources.
Study Limitations and Future Research
A strong point of this study is its longitudinal character. Thus,
the current findings can be framed in terms of cause and effect
relationships because main variables are measured at different
points of time, and we used several theories to formulate our
hypotheses (Taris, 2000). However, we the results still need to be
interpreted with some caution because of the non-experimental
RESOURCES AND FLOW AT WORK 17
nature of this study. Another strong point is that the model ex-
plained an acceptable part of the variance in the dependent vari-
ables (11% of the variance in T2 resources, 28% of the variance
in T2 flow, and 22% of the variance in T2 self-efficacy). How-
ever, a limitation of this study is that only teachers’ self-reports
were used in the examination of relationships between resources
and flow. Here the problem of common method variance may
have played a role. Because the study was conducted at 24 dif-
ferent schools, we could examine to what extent the flow experi-
ence per school was a function of the resources available at that
school. Additional analyses showed that the scores on the flow
dimensions absorption, work enjoyment and intrinsic work
motivation were highest at those schools where many resources
were available. These findings indicate that the subjectively re-
ported resources were anchored in the objective working situa-
tion. Results thus offer specific starting-points for interventions
aimed at mobilizing organizational resources and the promotion
of flow experiences at schools.
Finally, this study is limited to the context of secondary
school teachers. Since the main hypotheses were confirmed
regarding reciprocal relationships between flow and resources, it
would be interesting and relevant to examine this phenomenon
in other occupational fields. In addition, it would be important
to test the upward spiral model of these relationships using a
cross-lagged panel design with more waves, in order to test the
long time effects of these reciprocal relationships over time.
Practical Implications
It can be concluded that it is important for teachers to have
sufficient resources available in their work. There are ‘‘good
ingredients’’ of organizational culture such as social support,
innovation, rules and goals orientation are examples of such
organizational resources. When these resources are present in
the organizational environment teachers are more likely to get
engrossed in their work, to enjoy their work activities, and to
be intrinsically motivated. They seem to form the basic ingredi-
ents of a ‘‘Good Working Life.’’ As Nakamura and
Csikszentmihalyi (2002) noted a good life is one that is charac-
terized by complete absorption in what one does (p. 89). When
MARISA SALANOVA ET AL.
18
work (re)design takes into account these organizational
resources, there may be a good start, Job (re-)design seems to
be a good organizational strategy to implement in schools in or-
der to better teaching activities performance. On the basis of
this job design, it is possible to optimize the working environ-
ment. On the other hand, feeling competent at work is other
important (personal) resource to optimize at work. Our study
showed once more the positive consequences of self-efficacy: the
power of believing you can...
REFERENCES
Arbuckle J.L.: 1997, Amos users’ guide. Version 4.0. Chicago, Ill: Smallwaters.
Bakker, A.B.: 2001, Questionnaire for the Assessment of Work-related Flow:
The WOLF (Department of Social and Organizational Psychology, Utrecht
University, The Netherlands).
Bakker, A.B.: 2005, ‘Flow among music teachers and their students: The
crossover of peak experiences’, Journal of Vocational Behavior 66, pp. 26–
44.
Bakker, A.B., E. Demerouti, De Boer and W.B. Schaufeli: 2003, Job demands
and job resources as predictors of absence duration and frequency, Journal
of Vocational Behavior 62, pp. 341–356.
Bakker, A.B., E. Demerouti and W.B. Schaufeli: 2003, ‘Dual processes at work
in a call centre: An application of the Job Demands – Resources model’,
European Journal of Work and Organizational Psychology 12, pp. 393–417.
Bakker, A.B., E. Demerouti and W. Verbeke: 2004, ‘Using the Job Demands –
Resources model to predict burnout and performance’, Human Resource
Management 43, pp. 83–104.
Bandura, A.: 1997, Self-efficacy: The Exercise of Control (Freeman, New
York, NY).
Bandura, A.: 1999, Social cognitive theory of personality, in L. Pervin
O. John (eds),Handbook of Personality 2nd edition, (Guilford, New York),
pp. 154–196.
Bandura, A.: 2001, ‘Social cognitive theory: An agentic perspective’, Annual
Review of Psychology 52, pp. 1–26.
Boulton, M.J. and P.K. Smith: 1992, ‘The social nature of play fighting and
play chasing: Mechanisms and strategies underlying cooperation and com-
promise’, in J.H. BarkowL. Cosmides and J. Tooby (eds), The Adapted
Mind: Evolutionary Psychology and the Generation of Culture (Oxford
University Press, New York), pp. 429–444.
Bryan, R., S. Mathur and K. Sullivan: 1996, The impact of positive mood on
learning, Learning Disabilities Quarterly 19, pp. 153–162.
RESOURCES AND FLOW AT WORK 19
Caro, T.M.: 1988, Adaptive significance of play: Are we getting closer?, Tree 3,
pp. 50–54.
Csikszentmihalyi, M.: 1990, Flow: The Psychology of Optimal Experience
(HarperCollins, New York).
Csikszentmihalyi, M.: 1997, Finding Flow: The Psychology of Engagement
with Everyday Life (HarperCollins, New York).
Csikszentmihalyi, M., K. Rathunde and S. Whalen: 1993, Talented Teenagers:
The Roots of Success and Failure (Cambridge University Press, New York).
Deci, E.L. and R.M. Ryan: 1985, Intrinsic Motivation and Self-determination
in Human Behaviour (Plenum, New York).
De Lange, A.: 2003, Three types of causation and longitudinal research. APA-
NIOSH conference, Toronto, Canada (March 2003).
Demerouti, E., A.B. Bakker, F. Nachreiner and W.B. Schaufeli: 2001, ‘The job
demands – resources model of burnout’, Journal of Applied Psychology 86,
pp. 499–512.
Diener, E.: 2000, Subjective well-being, the science of happiness and a proposal
for a national index, American Psychologist 55, pp. 34–43.
Diener, E. and C. Diener: 1996, Most people are happy Psychological Science
7, pp. 181–185.
Ellis, G.D., J.E. Voelkl and C. Morris: 1994, ‘Measurement and analyses issues
with explanation of variance in daily experience using the flow model’,
Journal of Leisure Research 26, pp. 337–256.
Fredrickson, B.L. and C. Branigan: 2005, ‘Positive emotions broaden the scope
of attention and thought-action repertories’, Cognition and Emotion 19,
pp. 313–332.
Fredrickson, B.L., and T. Joiner: (in press), ‘Positive emotions trigger upward
spirals toward emotional well-being’ Psychological Science.
Fredrickson, B.L.: 1998, ‘What good are positive emotions?’, Review of
General Psychology 2, pp. 300–319.
Fredrickson, B.L.: 2001, ‘The role of positive emotions in positive psychology:
The broaden-and-build theory of positive emotions’, American Psychologist
56, pp. 218–226.
Fredrickson, B.L.: 2002, Positive emotions, in C.R. Snyder and S. Lopez (eds),
The Handbook of Positive Psychology (Oxford University Press, Oxford),
pp. 715–730.
Ghani, J.A. and S.P. Deshpande: 1994, ‘Task characteristics and the experience
of optimal flow in human-computer interaction’, The Journal of Psychology
128, pp. 381–391.
Grau, R., M. Salanova and J.M. Peiro
´: 2001, Moderating effects of self-efficacy
on occupational stress, Psychology in Spain 5, pp. 63–74.
Gonza
´lez-Roma
´V., S. Lloret and J.M. Peiro
´: April, 1995, The Measurement
of climate: An abridged version of the FOCUS Questionnaire. 7th European
Congress on Work and Organizational Psychology, Gyoˆ r, Hungary.
Hackman, J.R. and G.R. Oldham: 1980, Work Redesign Reading (Addison
Wesley, MA).
MARISA SALANOVA ET AL.
20
Harackiewicz, J.M and A.J. Elliot: 1998, ‘The joint effects of target and pur-
pose goals on intrinsic motivation: A mediational analysis’, Personality and
Social Psychology Bulletin 24, pp. 657–689.
Hobfoll, S.E.: 1989, ‘Conservation of resources: A new attempt at conceptu-
alizing stress’, American Psychologist 44, pp. 513–524.
Hobfoll, S.E.: 1998, Stress, Culture, and Community: The Psychology and
Philosophy of Stress (Plenum, New York).
Hobfoll, S.E.: 2002, ‘Social and psychological resources and adaptation’,
Review of General Psychology 6, pp. 307–324.
Hobfoll, S.E., R.J. Johson, N. Ennis and A.P. Jackson: 2003, ‘Resource loss,
resource gain and emotional outcomes among inner city women’, Journal of
Personality and Social Psychology 84, pp. 632–643.
Houkes, I.: 2002, Work and Individual Determinants of Intrinsic Work
Motivation, Emotional Exhaustion and Turnover Intention. Doctoral thesis,
Maastricht University. Maastricht, The Netherlands: Datawyse.
Hoyle, R.H.: 1995, The structural equation modeling approach: Basic concepts
and fundamental issues, in R.H. Hoyle (ed.), , Structural Equation Model-
ing, Concepts, Issues, and Applications (Sage, Thousand Oaks, CA), pp. 1–
15.
Jo
¨reskog K.G. and D. So
¨rbom: 1986, LISREL User Guide Version VI (4th ed)
(Scientific Software International Mooresville, I11).
Kelly, J.E.: 1992, Does job re-design theory explain job re-design outcomes?,
Human Relations 45, pp. 753–774.
Larson, R. and M.H. Richards: 1994, Divergent Realities: The Emotional
Lives of Mothers, Fathers, and Adolescents (Basic Books, New York).
Luthans, F.: 2002a, ‘Positive organizational behavior: Developing and main-
taining psychological strengths’, Academy of Management Executive 16, pp.
57–72.
Luthans, F.: 2002b, The need for and meaning of positive organizational
behavior, Journal of Organizational Behavior 23, pp. 695–706.
Luthans, F.: 2003, Positive organizational behavior (POB): Implications for
leadership and HR development and motivation, in R.M. Steers L.W.
Porter and G.A. Bigley (eds), Motivation and Leadership at Work
(McGraw-Hill/Irwin, New York), pp. 178–195.
Marsh, H.W., J.R. Balla and K.T. Hau: 1996, An evaluation of Incremental
Fit Indices: A clarification of mathematical and empirical properties, in
G.A. Marcoulides and R.E. Schumacker (eds), Advanced Structural Equa-
tion Modeling, Issues and Techniques (Lawrence Erlbaum, Mahwah, NJ),
pp. 315–353.
Pitts, S.C., S.G. West and J.Y. Tein: 1996, Longitudinal measurement models
in evaluation research: Examining stability and change, Evaluation and
Program Planning 19, pp. 333–350.
Salanova, M., S. Agut and J.M. Peiro
´: in press, Linking organizational
resources and work engagement to employee performance and customer
RESOURCES AND FLOW AT WORK 21
loyalty: The mediating role of service climate. Journal of Applied Psychol-
ogy.
Salanova, M., R. Grau, E. Cifre and S. Llorens: 2000, Computer training,
frequency of use and burnout: The moderating role of computer self-efficacy,
Computers in Human Behaviour 16, pp. 575–590.
Salanova, M, S. Llorens, E. Cifre, I. Martinez and W.B. Schaufeli: 2003,
‘Perceived collective efficacy, subjective well-being and task performance
among electronic work groups: An experimental study’, Small Groups
Research 34, pp. 43–73.
Salanova, M., S. Llorens, J.M. Peiro
´and W. Schaufeli: July, 2002, How
Positive Emotions Predict Self-Efficacy Via Engagement: A Positive Occu-
pational Psychology Framework (Symposium Job Engagement, 25th ICAP
Congress, Singapore).
Salanova, M., J.M. Peiro
´and W.B. Schaufeli: 2002, ‘Self-efficacy specificity
and burnout among information technology workers: An extension of the
job demands-control model’, European Journal of Work and Organizational
Psychology 11, pp. 1–25.
Schwarzer, R.: 1999, General perceived self-efficacy in 14 cultures. Self-efficacy
assessment. http:/www.yorku.ca/faculty/academic/schwarze/world14.htm.
Taris, T.W.: 2000, A Primer in Longitudinal Data Analysis (Sage, Thousand
Oaks, CA).
Van Muijen, J.J., P. Koopman, K. De Witte, G. De Cock, Z Susanj, F.D.
Lemoine Bourantes, N. Papalexandris, I. Branyicski, E. Spaltro, J. Jesuino,
J. Gonzalez das Neves, H. Pitariu, E. Konrad, J. Peiro
´, V. Gonzalez-Roma
´
and D. Turnipseed: 1999, Organizational culture: The FOCUS question-
naire, European Journal of Work and Organizational Psychology 8, pp. 551–
568.
Veenhoven, R.: 1984, Conditions of Happiness (Kluwer, Dordrecht, The
Netherlands).
Wong, C.S., C. Hui and K.S. Law: 1998, A longitudinal study of the job
perception-job satisfaction relationship: A test of the three alternative
specifications, Journal of Occupational and Organizational Psychology 71,
pp. 127–146.
Wright : 2003, Positive organizational behavior: An idea whose time has truly
come, Journal of Organizational Behavior 24, pp. 437–442.
Address for Correspondence:
MARISA SALANOVA
Department of Psychology
Jaume I University
Campus Riu Sec, s/n 12017
Castello
´n, Spain
E-mail: Marisa_Salanova@uji.es
MARISA SALANOVA ET AL.
22
