Journal of Vocational Behavior 66 (2005) 26–44

www.elsevier.com/locate/jvb

Flow among music teachers and their
students: The crossover of peak experiencesq
Arnold B. Bakker*
Utrecht University, Department of Social and Organizational Psychology,
and Research School Psychology and Health, Utrecht, The Netherlands
Received 13 May 2003
Available online 4 February 2004

Abstract
This study among 178 music teachers and 605 students from 16 different music schools examined the peak experience of flow (Csikszentmihalyi, 1990). On the basis of the literature, it
is hypothesized that job resources, including autonomy, performance feedback, social support, and supervisory coaching have a positive influence on the balance between teachersÕ
challenges and skills, which, in turn, contributes to their experience of flow (absorption, work
enjoyment, and intrinsic work motivation). In addition, using emotional contagion theory, it
is hypothesized that flow may crossover from teachers to their students. The results of structural equation modeling analyses offer support for both hypotheses. These findings are discussed in light of theories about crossover and emotional contagion.
Ó 2003 Elsevier Inc. All rights reserved.
Keywords: Job resources; Flow; Crossover; Emotional contagion

1. Introduction
Flow is a state of consciousness where people become totally immersed in an activity, and enjoy it intensely. According to Csikszentmihalyi (1997), such a peak experience can emerge in any situation in which there is activity, and researchers have
q
I thank Korine Scheeres for her help with data collection, and Joy Oliver for her valuable comments
on a previous draft of this paper.
*
Fax: +31-30-2537584.
E-mail address: a.bakker@fss.uu.nl.

0001-8791/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.jvb.2003.11.001

Csikszentmihalyi & Csikszentmihalyi. and intrinsic work motivation. 1989. 1994).g. Csikszentmihalyi (1990). . 383).B. Kowal & Fortier. and Morris (1994) define flow as an optimal experience that is the consequence of a situation in which challenges and skills are equal. 1990). & Whalen. Jackson & Marsh. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 27 indeed found evidence for flow during the execution of a large number of different activities. or activity. Absorption refers to a state of total concentration. Employees who enjoy their work and feel happy make a very positive judgment about the quality of their working life (cf. 1996). Lutz and Guiry (1994) describe flow as: ‘‘. whereby employees are totally immersed in their work. work enjoyment. These three elements are indeed the core components that are usually included in studies by many flow-researchers (e. enjoyment. 1996).. Finally. Csikszentmihalyi. object. In the present study. and playing music (Catley & Duda. . 1997. and intrinsic motivation. This enjoyment or happiness is the outcome of cognitive and affective evaluations of the flow experience (cf. 2. Furthermore. Flow at work What is flow? The literature reveals a laundry list of definitions of the concept. 1993. The main research questions addressed whether job resources facilitate flow at work. Csikszentmihalyi & LeFevre. There is an optimum level of challenge relative to a certain skill level. it can be defined as a short-term peak experience at work that is characterized by absorption. such a situation facilitates the occurrence of flow-related phenomena. who coined the term flow. . 1988. Diener & Diener.’’ In addition to the pleasure in the activity and the intrinsic motivation to continue doing it. 2000. 1996. Veenhoven. According to these researchers. a state of mind sometimes experienced by people who are deeply involved in some event. 1984. For example. the phenomenon of flow is investigated among music teachers and their students. 337). Csikszentmihalyi. Indeed. Csikszentmihalyi. . work. . Ellis. time may seem to stand still and nothing else seems to matter while engaged in the consumption event. Ghani and Deshpande (1994) particularly emphasize the total concentration and the enjoyment people experience when in flow: ‘‘The two key characteristics of flow are (a) total concentration in an activity and (b) the enjoyment which one derives from an activity. Csikszentmihalyi & Csikszentmihalyi. such as positive affect. and whether this experience may crossover from music teachers to their students. Rathunde. 1999). Voelkl.’’ (p. the experience itself is so enjoyable that people will do it even at great cost.A. 1988. . The most prominent definitions of flow seem to have three elements in common. 1997. the total immersion in an activity seems to be a central aspect of the flow-experience. describes the concept as: ‘‘The state in which people are so intensely involved in an activity that nothing else seems to matter. . . and they forget everything else around them (cf. arousal. for the sheer sake of doing it. and intrinsic motivation (p. including sports. when flow is applied to the work situation. .’’ These scholars emphasize the idea that time flies during a flow experience. Accordingly. Diener. Larson & Richards. Time flies. namely absorption (‘‘the total immersion in an activity’’). they are completely and totally immersed in it.

Deci & Ryan. 1998). Csikszentmihalyi and Csikszentmihalyi (1988.g. theoretically.’’ Applied to the work situation.. 1985).g. Does this also apply to music teachers? According to Klausmeier (1978). Moreover.. What are the antecedents of flow? Csikszentmihalyi (1990) discovered that artists and athletes are particularly prone to experiencing flow. including work. 1994. As soon as the player feels confident with the scales. Ellis et al. The identification with the music allows the musician to feel at one with and become absorbed in the music (Palmer.B. self high skills). In addition. for example. research has shown that people need challenges to facilitate flow in a range of activities. For example. Karasek. There is indeed some empirical evidence for this pattern of experiences (e. and a high probability of stress if the opponent is much better (high challenge. 1988). for example. Clarke & Haworth. the practice. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 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. low skills). that boredom is more likely when people spend their leisure time on activities such as watching television. 1997. 1990. what are the factors in music teachersÕ working environment that contribute to flow experiences? CsikszentmihalyiÕs (1997) experience sampling studies have shown that people more often experience flow during their work than during free time. researchers generally agree that the occurrence of flow is most likely when people perceive a balance between the challenge of a situation and their own skills to deal with this challenge (e.28 A. 1989). There is balance when. We may therefore expect that musicians frequently experience flow. 261) have argued that ‘‘a beginning piano player will see learning the keys corresponding to the various notes as challenging. they are fascinated by the tasks they perform (Csikszentmihalyi. These studies show. Csikszentmihalyi. new challenges need to be found or he or she will get bored. a tennis player experiences a balance when he is confronted with an opponent who is approximately equally skilled. 1979. 1997). On the contrary. Edwards. p. 1996. however. This suggests that one has to invest time and energy to experience flow. Employees who are motivated by the intrinsic aspects of their work tasks want to continue their work. Intrinsically motivated employees are continuously interested in the work they are involved in (Harackiewicz & Elliot. an exciting game may develop in which the tennis players have to do their utmost to beat the opponent. Csikszentmihalyi. However. Massimini & Carli.. the high level of effort that is required to perform well coincides with high concentration. 1994. Indeed. and transference of music offers many reasons to become totally immersed in the activity. Relevant to the activity examined in the present study. 3. 1988). this means that employees are more likely to experience flow when their job demands or challenges match their professional skills. there is a higher probability of boredom if the opponent has fewer skills (low challenge in the situation. and might feel in flow simply by running the scales on the keyboard. experienced music teachers succeed in teaching . performance. In this situation. Massimini & Carli.

have a positive influence on the balance between their challenges and skills. Noe. Hackman & Oldham. 1979). work enjoyment. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 29 jazz music to inexperienced students who only learned to play rudimentary and specific pieces of classical music. 1996). and supervisory coaching. Bakker. Such a goal will particularly be reached when employees have certain professional skills (cf. Loher. Research has indeed shown that such job resources make a positive contribution to the motivation. On the basis of these previous findings. Moeller. & Schaufeli. This balance. in press). including autonomy. 1985. 1979. Demerouti.B. 2001. & Fitzgerald. the Person-Environment fit model. 1980. the following hypothesis is formulated (see also Fig. in turn. De Jonge. and may buffer the undesirable influence of job demands on stress reactions (Karasek. performance feedback. 1. Fig. Van Vegchel. and feedback about performance makes clear to what extent work-related goals have been reached (Locke. Schaufeli & Bakker. 2003a. Demerouti. and autonomy (Bakker. 2002). social support from colleagues. and performance of individuals (De Jonge. & Schaufeli. commitment. and intrinsic work motivation). Music teachersÕ job resources. contributes positively to the experience of flow (absorption. Possibilities for self-growth. or have sufficient resources in their work—such as support from colleagues. De Boer. Orpen. 1979. for example. & Schaufeli. performance feedback. will enable employees to cope better with the demands of their work. . see Hatfield. 1): Hypothesis 1. The Flow Model. Karasek. good material. 4. Nachreiner. Bakker. 1995. Is flow contagious? Several field and experimental studies have shown that positive and negative emotions can crossover from one person to another (for an overview. 1968).A. Edwards.

several ongoing interactive (verbal as well as non-verbal) processes take place between the teacher and his or her students. to converge emotionally’’ (Hatfield et al. Recent organizational studies have shown that employee well-being may take contagious forms as well. Likewise. Masters. the teacher plays an important role in designing the lessons. p.g. based on emotional contagion theory. In addition to seeing each other and talking to each other. cf.. & Schaufeli.30 A. evidence for positive crossover in the workplace. Campos and Sternberg (1981) have called this social referencing. and that the tone height and speed of the speaker influences the way he or she feels (Hatfield et al. & Bosveld. we may expect that flow experiences of music teachers have a positive influence on those of their students. Demerouti. and McHugo (1985). For intonation and speed of speech it has been shown that partners mutually influence and adjust to each other. Adelmann & Zajonc. defined as: ‘‘The tendency to automatically mimic and synchronize facial expressions. general practitioners (Bakker. Westman. consequently. 1988). 2001). During the teaching. For example.. Taken together. Kim. in a series of crossover studies among couples. Thus. 1998) has shown that burnout and depression may transfer from one spouse to another. burnout. showed that supporters shared his enjoyment when he was telling something happy. 2003) caught the burnout symptoms of their colleagues. Although opponents reported negative feelings during the whole speech. the emotions of mothers can be influenced by the facial expressions of their newborn children (Frodi et al. since stressful demands or a bad day at work have a negative impact on colleaguesÕ well-being. However. Woodson. Stiff. The perception and (unconsciously) adoption of facial expressions seems important (e. This phenomenon has also been described as emotional contagion. and whether we want to or not. 1995. Westman & Vinokur. Schaufeli. 5). in which individuals were filmed and questioned while they were watching a speech of President Reagan.. Sixma. In addition. Bakker and Schaufeli (2000) showed that teachers. Dillard. Somera. 2001.B. Cohen.. Sullivan. 1978). vocalizations. An experiment of Lanzetta. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 Cacioppo. strain. and when they frequently discussed work-related stress. 1989). their facial expressions were congruent with those of Reagan. 1994). 1994). and reported tension when he was telling something fearful. Note that the evidence found in field research for crossover relates solely to negative emotions (e. and Greenberg (1982) showed that the facial expression of a newborn child resembles the facial expression of the mother. these findings indicate that our emotions are influenced by those of others whether we like the others or not.. Field. & Sleight. & Rapson.g. 2001). It is still unclear which different mechanisms play a role in the process of crossover or emotional contagion. and employees working for an insurance company (Bakker. 1994. Westman & Etzion. if any. and depression). Garcia. The mood of the teacher thus influences many . For example. particularly when they were susceptible to the emotions of others (an individual difference variable. positive experiences at work may also crossover and have a positive effect on othersÕ well-being (cf. There is little. assessments of their galvanic skin responses showed that both supporters and opponents were more relaxed during the happy messages than during the disturbing announcements. (Westman. In addition. postures and movements with those of another person and.

the processes of emotional contagion may take effect. they were teaching 12 h (SD ¼ 7:1 h) per week at their music school and organizational tenure was on average 13 years (SD ¼ 10:2 years). by referring to flow experiences when playing their musical instrument. indirectly. The accompanying letter explained that the goal of the study was to examine Ôteacher well-being. In addition. the second hypothesis is: Hypothesis 2. 4 students filled out the flow questionnaire. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 31 factors to which students are exposed during the lesson. This may determine the quality of the lesson and. There exists a positive relationship between the experience of flow (absorption.A. of which 16 participated (57% of the schools). guitar (12%). and in this way.’’ and ‘‘Participation is not anonymous if I fill out my age and instrument.’’ ‘‘We already have to fill out too many forms. it also contains the components of intrinsic work motivation and absorption. Regarding genre. 36% only taught individual students. The items were somewhat adjusted for the students.1.’’ For 67% of all teachers. Reasons not to fill out the questionnaire were: ‘‘I am too busy. for 80% of all teachers at least 3 students filled out the questionnaire. the enjoyment of the student. The average age of the students was 19 years (SD ¼ 13:2). On average. The remainder taught both groups and individuals.B. enjoyment. flute (12%). work enjoyment. 40% performed other activities as well. and intrinsic work motivation) of music teachers and the flow experiences of their students. violin (10%). The 405 music teachers who were employed at these schools received questionnaires in their mailboxes at work.Õ In addition. which may influence the student. On average. The sample included 75 men (42%) and 103 women (58%). The teaching concerned several instruments. Method 5. In total. such as providing private lessons and playing in an ensemble or professional orchestra (or both). Therefore. In addition to questions about demographics. Procedure and participants In consultation with the Royal Dutch Society for Music Artists (Koninklijke Nederlandse Toonkunstenaars Bond). In total. The mean age of the participants was 41 years (SD ¼ 9:8). as well as items about work-related flow experiences. Of all teachers. 57% of the . the questionnaire included items to assess job resources and the balance between challenges and skills. and keyboard (9%). particularly piano (17%). 5. and for 92% of all teachers at least 2 students filled out the questionnaire (for all teachers at least 1 student). 178 teachers (44%) and 605 students filled out the questionnaire. and intrinsic motivation) to four of their students during one of the lessons. the flow experience of a teacher not only includes a component of work enjoyment. Teachers and students turned in the questionnaires separately at the reception desk of their music school. all teachers were asked to randomly distribute a short flow questionnaire (including the three dimensions of absorption. they had received music lessons for five years and generally studied between 2 and 4 h per week. 28 music schools were approached.

Performance feedback was assessed with a three-item scale developed by Bakker et al. A sample item is: ‘‘Can you decide yourself how you execute your work?’’ Items are scored on a five-point Likert scale (1 ¼ never. Therefore. A sample item is: ‘‘I receive sufficient information about the goal of my work’’ (1 ¼ never. 5 ¼ always). Autonomy was measured with a short scale developed by Bakker. Preliminary analyses showed that none of the demographics was related to the model variables. and not from the rewards for it’’ (intrinsic work motivation). 4 ¼ often. 1994). (2003c). always in combination with classical music or pop music. work enjoyment (4 items) and intrinsic work motivation (5 items). Hackman & Oldham.e. and Harrison (1982). and Schreurs (2003c).’’ Flow was assessed with a recently developed instrument named the WOrk-reLated Flow scale (WOLF for short. 6 ¼ always). Examples are: ‘‘When I am working. 27% classic and pop music. A sample item is: ‘‘Can you ask your colleagues for help if necessary?’’ (1 ¼ never. 2001). Bakker. 3 ¼ regularly. Sample items are: ‘‘I am well able to meet the demands of my work. ‘‘When I am working very intensely.96. Autonomy or job control is considered a job resource since many studies have shown that this work characteristic enables employees to cope with job demands (Van der Doef & Maes.75 to .88 to . Social support was also assessed with a three-item scale developed by Bakker et al. I forget everything else around me’’ (absorption).32 A. 5. The WOLF includes 13 items measuring absorption (4 items). and . The participants were asked to indicate how often they had each of the experiences during the preceding week (0 ¼ never. Coaching by the supervisor was measured using a Dutch adaptation of Graen and Uhl-BeinÕs (1991) Leader-Member exchange scale (Le Blanc. based on French. Demerouti.2. I feel happy’’ (work enjoyment). The Balance between challenges and skills was assessed with a five-item scale. and has motivational potential (Fried & Ferris. 5 ¼ always). such as: ‘‘My supervisor uses his/her influence to help me solve my problems at work’’ (1 ¼ never.63 to . All responses were coded such that higher scores referred to more job resources. 2 ¼ sometimes. 5 ¼ very often. Caplan. freedom of action in accomplishing the formal work task). work enjoyment and intrinsic work motivation ranged from . respectively. and ‘‘I get my motivation from the work itself. 5 ¼ always). Measures Job resources. 2001). 1 ¼ almost never.. and have good reliabilities (Bakker. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 teachers taught only classical music. these factors will not be included in the following analyses. Taris.86. 1980). In addition. Four job resources were included in the questionnaire. Schaufeli. 20% of the teachers included jazz music in the curriculum. The scale includes five items. CronbachÕs a-coefficients for absorption. 5 ¼ always). Seven previous studies among a total of 1346 employees from different organizations and occupations have shown that the three factors can be empirically distinguished. (2003c). a longitudinal study among 248 consultants who filled . 1999). The scale includes three items referring to decision authority (i. 1987.B. .’’ and ‘‘I have sufficient skills to carry out my work tasks properly.82. In addition. and 16% only pop music.

77.79.. The scale for assessing ÔbalanceÕ was split in two reliable halves (cf. The intercorrelations between the three flow components are relatively high. All items that refer to work were changed by referring to playing music. for example: ‘‘When I am playing music.08 indicate a reasonable fit between the model and the data (Browne & Cudeck. Descriptive statistics Table 1 shows the means. the incremental fit index (IFI).B. and with teachersÕ flow experiences. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 33 out the WOLF twice with 6 weeks between both waves showed that the test–retest reliabilities of the subscales are satisfactory (Bakker. The test–retest correlations were . and the comparative fit index (CFI) were also examined. reliability coefficients. the goodness-of-fit index (GFI) and the root mean square error of approximation (RMSEA) were calculated. Table 1 shows that the four job resources correlate significantly with each other. and Hau (1996)—the non-normed fit index (NNFI). p < :001). the traditional v2 value. As can be seen. and (2) flow among music teachers.3. as recommended by Marsh. 1997).74. Results 6. structural equation modeling (SEM) analyses were carried out with the AMOS software package (Arbuckle. the items were slightly adjusted. the scales introduced before). . Because these indices are dependent on sample size. Balla. work enjoyment. respectively. and correlations between all study variables. consists of hypothetical constructs or latent variables that are all estimated by two or more manifest variables that are directly observed (i. and performance feedback. 1995). Analyses To test the two hypotheses simultaneously. The latent factor Ôflow among music teachersÕ was indicated by absorption.90 or higher (Hoyle. Finally. and intrinsic work motivation. In addition. all constructs that were assessed demonstrated good internal consistencies. 6. the three respective flow scales indicated the latent endogenous factor Ôflow among studentsÕ. As a rule of thumb. a GFI P . In order to test the fit between the model and the data. work enjoyment.A. To determine flow among students. the structural model includes two latent mediating variables: (1) balance between challenges and skills. 5.’’ The scores on each of the three flow dimensions were calculated for groups of students by using the average of all students for each teacher. 1993).76 (r ¼ :61. manifest variables. namely social support. Bakker et al. In addition. The model. as displayed in Fig. autonomy. . The second subscale includes three items and has a reliability of . These indices should have values of . The latent exogenous factor Ôjob resourcesÕ was operationalized by four observed. and intrinsic work motivation. 2000) that served as the indicators of the latent variable Ôbalance. 2001)..71 for absorption.1. standard deviations.Õ The first subscale includes two items and has a reliability of . 1. I forget everything else around me.90 and a RMSEA P . supervisory coaching. and .e.

67 3.B.90) .34 Variable 1 Social support by colleagues 2 Supervisory coaching 3 Autonomy 4 Performance feedback 5 Balance (indicator 1) 6 Balance (indicator 2) 7 Absorption teacher 8 Work enjoyment teacher 9 Intrinsic work motivation teacher 10 Absorption Students 11 Enjoyment students 12 Intrinsic motivation students * p < :05.02 (.18 (.01 ).89 4.98 .20 3.08 .33 .49 .10 .91) A.56 .10 . on the diagonal).18 .83 .09 ).03 ).08 .07 2 3 4 5 6 7 8 9 (.76) .20 .02 .66 (.21 (.40 11 12 (.09 .75) .07 (.09 .07 (.98 .07 ).11 .18 .30 .17 .12 ).70 2.75 .88) .01 .74 1.02 ).78 4.28 4. p < :01.62 .04 ).14 .05 .76) .20 .69 (. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 Table 1 Means.03 .06 .13 3.28 .39 .50 (.44 .12 .39 .27 10 (.15 .93 1.10 . and correlations between the variables.11 .64 1.07 .06 .90) .79) .18 .71) .10 .91) .73 4.27 . reliability coefficients (CronbachÕs a. N ¼ 178 music teachers and 605 students .15 .09 .34 .23 .33 .17 3.37 1.77) .11 .28 .86) 3. standard deviations.22 ).08 .05 .72 . ** M SD 1 3.24 (.30 .62 .02 .76 4.18 .18 .09 ).

10 (t0 s < 1).50 111.08 .s. The first row in Table 2 shows that the hypothesized model fits reasonably well to the data. M2. and . and . The factor loadings for absorption. SEM analyses were carried out with the AMOS software package (Arbuckle. and .63.90 . Test of the flow model According to Hypothesis 1. Only the NNFI is. both paths were included in the model. all indicators had significant loadings on the intended factors.08 .90.91 .2. and IFI are all . This did also not result in a better fit of the model to the data (Delta v2 ð1Þ ¼ :27.90.87 .87. GFI. CFI.87 . M3.94 51 49 50 66 . non-significant value of . N ¼ 178 music teachers and 605 students Model v2 df GFI RMSEA NNFI CFI IFI M1.A. Furthermore.80.90 . Bakker / Journal of Vocational Behavior 66 (2005) 26–44 35 for both the teachers and the students. and—together with a RMSEA of . NNFI. RMSEA. for flow among teachers. 6. To test these hypotheses simultaneously. the path from balance to flow among students was added to M1. root mean square error of approximation. chi-square. 1). non-normed fit index. and intrinsic work motivation).08 . work enjoyment.63. . and from job resources to flow among students. contributes to explaining variance in music teachersÕ experiences of flow (absorption. In addition. somewhat lower than the criterion value of .08—this indicates an acceptable fit between the model and the data.76 113.54 to . in turn. CFI.59 . n.23 .95. The results (see Table 2) showed that this alternative model (M2) does not fit better to the data than the proposed hypothetical model. n.90 . In a third model (M3). job resources have a positive relationship with the balance between challenges and skills. which. Moreover.90 – Flow Model Alternative Model Alternative Model Null Model Note: v2 .87 – . respectively. alternative model. both path coefficients had the same. IFI. and the coefficient of the additional path proved to be non-sig- Table 2 Results of SEM-analyses: Fit indices of the flow model and the alternative models. .90 . To test the alternative hypothesis that job resources also have a direct relationship with flow among teachers and flow among students.71 for flow among students. incremental fit index.90 . see also Fig. goodness-of-fit index.23 673. (work) enjoyment and intrinsic (work) motivation were . The GFI. standardized maximum likelihood estimates.90 – . at . including the path from job resources to flow among teachers. For job resources.74. teachersÕ flow experiences partly correlate with those of the students. M2. . 1997). 113.B. comparative fit index. alternative model.90 .59. In addition. Hypothesis 2 states that there is a positive relationship between teachersÕ flow and that of their students (contagion hypothesis. Delta v2 ð2Þ ¼ 1:74.s.). these factor loadings ranged from . df . degrees of freedom. and M3. including the path from balance to flow among students. Another noticeable finding is that social support reported by teachers correlates significantly with studentsÕ levels of absorption and enjoyment when playing music. M0.

t ¼ 1:52). the following rule of thumb was used: teachers and students have a peak experience if they score higher than or equal to the 75e percentile of each of the three flow dimensions. GFI ¼ :96. The model explains 5% of the variance in Ôbalance between challenges and skillsÕ.B. RMSEA ¼ :05. 2. However. In this model. are summarized in Fig.36 A. The Flow Model: Standardized solution (maximum likelihood estimates). These results of the alternative models offer additional support for the hypothetical model. N ¼ 178 teachers/605 students. NNFI ¼ :96. 6. IFI ¼ :97. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 Fig.7%) and 20 groups of students (11. nificant (b ¼ :17. one could argue that the peak experience of flow is an all or nothing phenomenon: an experience that one either has or does not have (dichotomous variable). 19 teachers (10. CFI ¼ :97. and the latent factor flow was indicated by each of these variables in the SEM-analyses. but now flow was included in the model as an observed. The results of the final flow model. (work) enjoyment.3. 2. manifest variable (participants scored 0 or 1). 16% of the variance in flow among teachers. and 6% of the variance in flow among students. which is identical to the hypothetical model. According to this criterion. Flow as a peak experience In all previous analyses. p ¼ :12. and intrinsic (work) motivation). To examine whether this alternative operationalization of flow would lead to different results. This model also turned out to fit the data well.2%) reported experiences of flow (and were assigned the number Ô1Õ in the analyses). In other words. The flow model was tested again. and it was assumed that individuals could score low or high on each of them. the relationship between ÔbalanceÕ and flow . the three flow dimensions were operationalized as continuous variables. an additional SEM-analysis was performed. Because externally validated criteria for cut-off scores on each dimension do not exist. flow was conceptualized as a phenomenon with three underlying dimensions (absorption. v2 ð19Þ ¼ 26:31.

Csikszentmihalyi. (CR ¼ 1:98. on the basis of previous research. Clarke & Haworth. work enjoyment and intrinsic work motivation). the first hypothesis was that job resources would have a positive relationship with music teachersÕ balance between their challenges and skills. Bakker et al. p < :05). the findings showed that job resources—a combination of autonomy. Jones. Discussion The central aim of this study was to answer two questions.17 (CR ¼ 2:23. It should be noted that maximum likelihood (ML) estimation assumes that variables are continuous instead of dichotomous. 2001.B. 1990. The results of this analysis confirmed that the balance between challenges and skills is significantly related to flow among music teachers (B ¼ 1:15. Resources in the working environment would facilitate this balance. because job characteristics such as autonomy. 1994.. In addition. and to what extent flow may crossover from teachers to their students. 1997). and that flow among music teachers is significantly related to flow among students (B ¼ 1:23. These findings confirm and expand CsikszentmihalyiÕs (1997. because of the relative small sample size (cf.. 1994. Individuals who experience flow forget everything else around them. Hoogland & Boomsma. 2003c.g. p < :05). 2. and intrinsic work motivation (cf. in turn. 1997. More specifically. I therefore checked the appropriateness of the findings by comparing the ML estimations with the results of logistic regression analyses (cf. performance feedback. Unfortunately. social support. work enjoyment. To answer both questions. Ellis et al. would contribute to explaining flow (absorption. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 37 among teachers was 17.A. They . p < :05). Demerouti et al. 2001). 1994). 1988). and that this balance. and perform certain work-related activities with the aim to experience enjoyment and satisfaction that is inherent to these activities. 7. Bakker. and the relationship between flow among teachers and flow among students was also . and supervisory coaching can foster personal growth and the realization of goals (cf. 1990) flow theory. Massimini & Carli. namely whether job resources are possible antecedents of flow experiences among music teachers. p < :05). although the relationships in the model are less strong. The results of SEM-analyses supported this hypothesis. flow at work was first defined as a short-term peak experience that is characterized by absorption. had predictive value for the frequency of flow among music teachers. these findings are consistent with those in Fig. performance feedback. it was predicted that flow is most likely when individuals perceive a balance between the challenges of a situation and their own skills to cope with these challenges (e. Taken together. in turn. Therefore... which. Ellis et al.. evaluate the quality of their working life very positively. 1998). Asymptotically Distribution Free (ADF) estimation cannot be used to circumvent this problem in the present study. Csikszentmihalyi. Logistic regression is similar to linear regression but is suited to models where the dependent variable is dichotomous. social support from colleagues and supervisory coaching—had a positive relationship with the balance between challenges and skills.

. but also the more conscious crossover of teachersÕ dedication to their work (cf. In previous studies. in particular. The correlational analysis suggested that. the higher the frequency of comparable experiences among their students.. and demands in excess of resources may yield increased distress. work enjoyment. and intrinsic work motivation) and the experience of flow among their students.B. and is one of the first demonstrations in field research that positive emotions may crossover from one person to another (Westman. there is another issue at stake here. autonomy. French et al. Some evidence for such a pattern of responses has indeed been found in previous job stress studies with the demand-control model (De Jonge & Kompier. Resources in excess of demands may yield increased boredom. The second hypothesis was that there exists a positive relationship between music teachersÕ flow (absorption. teachersÕ intrinsic work motivation was related to flow experienced by students. This hypothesis was also confirmed by the findings. This positive relationship should hold until equilibrium between job demands and resources is achieved. Hatfield et al. but more generally. 1997) and the job demands—resources model (Bakker. one may specify that it involves equilibrium between job demands (challenges) and skills. Demerouti. 1987). researchers have applied various methods to assess flow. and job resources. music teachersÕ and studentsÕ absorption and enjoyment also showed positive relationships. 2003b). This means that if job demands are held constant. 1982). 2001).38 A. This flow dimension showed the strongest positive relationships with studentsÕ intrinsic motivation. these results suggest that students ÔcaughtÕ the flow experience of their teachers in a partly conscious and partly unconscious way. However.. presumably mitigating stress along the way. and was also related to their absorption and enjoyment. As resources increase. the more flow experiences music teachers reported. supervisory coaching and autonomy may also contribute to such a balance. including performance feedback. Csikszentmihalyi). The crossover process may include the automatic imitation of a cheerful and happy teacher. motivated teachers probably put more effort and energy in the search for nice and suitable music for the students. This finding is in line with emotional contagion theory (Hatfield et al. In terms of balance. 1994).g. and self-esteem. and enjoyment had the highest loading on the latent variable ÔflowÕ in the SEM-analyses. social support from colleagues. the results show that besides the facilitating role of information about the goals and results of oneÕs work (cf. then the likelihood of a flow state may decline because boredom will become the more likely state. and coaching). In addition. skills. resources and the likelihood of flow will take on a quadratic function (cf. including e. Thus. In addition. self-efficacy. they presumably have a more positive attitude toward work that motivates their students to concentrate during the music lesson and to perform well. Warr. & Euwema. Csikszentmihalyi (1990) usually applies the experience sampling method . Bakker / Journal of Vocational Behavior 66 (2005) 26–44 emphasize the importance of the fit between the skills of the person and the challenges he or she is exposed to at work (cf. As resources exceed the fixed job demands. one may argue that balance involves equilibrium between job demands and all accessible resources (personal resources. 1994). social support. the likelihood of flow will increase. In addition.

and reduced professional efficacy. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 39 (Csikszentmihalyi & Larson. The question is. certain cause–effect relationships were assumed. In the present study. the frequency of flow will increase with an increase in each of these eight factors. although primarily assessing the three dimensions of absorption. Common method variance is a non-issue for the contagion hypothesis in this study. 1992). and that the relationships in the flow model were comparable with those between continuous variables (although weaker). Kimiecik & Stein. the conceptualization of flow as a combined measure is comparable with the approach of job burnout by Schaufeli and Van Dierendonck (2000). Participants are requested to react several times a day to a short questionnaire (experience sampling form) when they receive a random signal from an electronic device during a certain period of time (usually one week). but the existence of reversed causal effects cannot be excluded. has described several other ÔelementsÕ of flow. However. For example. Thus. concentration and focus. namely exhaustion. it is desirable that future studies examine such reciprocal effects. to what extent we really examined peak experiences among teachers and their students. Jackson and Marsh (1996) have argued that more research is needed to examine the reliability and validity of this method in field research. Using existing theories. 1992. namely music teachers and their students. Study limitations A limitation of the present study is its cross-sectional character. it is conceivable that teachers also experience more flow when their students have more of such peak experiences. transformation of time. Therefore. 1992). enjoyment. who define the risk of burnout as a combination of relatively high scores on the three dimensions measuring the syndrome. a strong point of this study is that two sources of information were used. and ask for attention to the practical problems that are inherent in this research strategy (see Kimiecik & Stein. . The results of the present study suggest that simple questionnaires may offer a reasonable alternative. loss of self-consciousness. the three flow dimensions were each assessed on a continuum. that Csikszentmihalyi (1990). According to Csikszentmihalyi. feedback. and an ‘‘autotelic’’ experience (the activity becomes a goal in itself). One could argue that individuals only experienced flow if they scored high on each of the three flow factors (for example. therefore. control. More research is needed to find out whether the three elements used in the current study sufficiently measure the experience of flow. (work) enjoyment and intrinsic (work) motivation). The results of additional analyses showed that approximately 11% of all teachers and students met this criterion. Note.B. a clear goal.A. P the 75e percentile). since one does not distinguish between the three underlying dimensions (absorption. although framed in terms of cause and effect. cynicism. the current findings. The disadvantage of this approach is that nuances in the experience of flow are lost.1. The advantage of a combined measure is that it may be very useful for organizational practice. and intrinsic motivation. including the balance between challenges and skills. did not demonstrate causality but merely relationships. however. 7. ranging from ÔneverÕ to ÔalwaysÕ. Interestingly.

g..e. it should be noted that the model explained only a limited amount of variance (16% of the variance in flow of teachers. although teachers were asked to randomly distribute short flow questionnaires among their students. regarding the crossover of flow. 2003) and from work to private life (e. this study is limited to the context of music teachers and their students. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 A second limitation is that only teachersÕ self-reports were used in the examination of relationships between job resources. These findings indicate that the subjectively reported job characteristics were anchored in the objective working situation. More specifically. On the other hand. sport teachers and their students). Third. 2001). enjoyment and intrinsic motivation during the lessons. 1988). and intrinsic work motivation were highest at those schools where many resources were available (particularly supervisory coaching and autonomy). which leads to an underestimation of the strength of relationships (cf. in some cases. Bakker et al. Because the study was conducted at 16 different schools. it should be noted that our strategy of analysis reduced the variance in studentsÕ flow experiences. It can be expected that both sources of information (in this study teachers and students) have their own unique causes of (statistically independent) error variance. Future studies with a similar design may circumvent this drawback by using multi-level analyses techniques. it would be interesting and relevant to examine this phenomenon in other teacher–student relationships. This can partly be explained by the use of different sources of information. work enjoyment. 2000. This implies that the variance in the student responses was restricted. If one uses only one method. Finally. the sources of error variance are the same. we could examine to what extent the flow experience per school was a function of the resources available at that school. Frese & Zapf. balance and flow. A limitation that we cannot overcome is that the average scores of the students on the three flow dimensions were.B. including flow. Several recent studies have shown that stress and strain can crossover at work (Bakker & Schaufeli. and 6% of the variance of the flow among students). It cannot be excluded that the teachers were selective. most teachers (67%) found four students who were willing to indicate the extent to which they experienced absorption. for each teacher. such as in non-traditional teacher–student relationships (i. Westman. Fourth. based on less than four students.40 A. it would be interesting to find out whether flow may also crossover from employees to their colleagues or partners.. The results thus offer specific starting-points for interventions aimed at mobilizing job resources and the promotion of flow experiences at music schools. Since the contagion hypothesis was confirmed regarding flow. From a practical and theoretical point of view it would be valuable to carry out more organizational studies to examine whether positive experiences. The consequence of this is inflated correlations. In addition. can crossover to others. In addition. . studentsÕ scale scores on each of the flow dimensions were summed for groups of students. Additional analyses showed that the scores on the flow dimensions absorption. it is unclear to what extent they complied with this request. Here the problem of common method variance may have played a role.

M. B. (2001). Long (Eds.. W. Bakker.. How job resources buffer the impact of job demands on burnout.. After completion of the present study. 1–21. (1989). W. A. & Schaufeli. A multi-group analysis of the job demands .. Bakker. A. B. & Bosveld. J. Manuscript submitted for publication. B. but one could also consider skill variety.. If.). such resources seem not only important for teachers themselves.. De Boer. Advances in Psychology Research. Version 3. International Journal of Stress Management. E. Using histograms. B. The Netherlands: Department of Social and Organizational Psychology. Utrecht. appraisal and emotion: The onset of social referencing.6. for example. In K. (2003c). Burnout contagion processes among teachers. L.B. & Schreurs. W. Bakker. A.. B. Annual Review of Psychology. Lamb & L. Testing structural equation models. T. Campos. E. Since we know that music teachersÕ experiences of flow coincide with those of their students.g. Alternative ways of assessing model fit. Browne. A. 2003b. Journal of Social and Clinical Psychology. B. Demerouti. On the basis of this information. Bakker et al.. M. Demerouti. 2003c).. & Euwema. including information that has been collected in recent years in many different companies (e. Sixma. P.. (2001). 249–280.. The socially induced burnout model. & Zajonc. A. W.). B. Amos usersÕ guide. it was explained how teachers at each school scored relative to the research group as a whole and in comparison with an external benchmark. Bakker.). R. & Schaufeli. Practical implications It can be concluded that it is important for music teachers to have sufficient resources available in their work. C. social support from colleagues. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 41 7. References Adelmann. Schaufeli.. In F. 2003c). R. R. Autonomy. Perception.A. Burnout contagion among general practitioners. (2003a). Bakker. E. 2003b. 16–38.. and high quality communication (see also Bakker et al. B. but also for the happiness and motivation of their students.B. Monterey. Columbus (Ed. E. B. Utrecht University. and performance feedback are examples of such resources. 1–84). it may be desirable to offer additional training to supervisors regarding leadership styles. 2003a. B. K. Schaufeli. Sage Publications. (2000). it is possible to optimize the working environment. S. . and by regularly informing music teachers about the results of their work. & Schaufeli. Bakker... Chicago: Smallwaters. (in press). CA: Brooks/Cole. A. one schoolÕs scores are relatively low on supervisory coaching and the flow scores of the teachers and students are inferior to those of teachers in other schools. (1997). & Sternberg.2. Demerouti. (2003). 10. 341–356. Journal of Vocational Behavior. Bollen & J. Journal of Applied Social Psychology. 82–98. Feedback could be mobilized as a resource by making job evaluations. Arbuckle.resources model in four homecare organizations. (1993). B. P.. Facial reference and the experience of emotion. supervisory coaching.. 30.. E. possibilities for self-growth. A. W. Bakker. Taris. 20.. Demerouti. J. In M.. New York: Nova Science Publishers. W. W. Job demands and job resources as predictors of absence duration and frequency. Sherrod (Eds. A. 40. Emotion (pp. (1981). Vragenlijst voor het meten van werkgerelateerde flow: De WOLF [Questionnaire for the assessment of work-related flow: The WOLF]. J.. H. B. the participating schools received feedback about their most important job resources. & Cudeck. participation in decision making. 62.

. E.. B. M. J. W. E. French. & Deshpande. E. Developmental Psychology. Lamb. J. An examination of competing versions of the person-environment fit approach to stress. J. B. (1978). Y. (1991). 55. Graen. L. 287–322. (1995). A. The job demands—resources model of burnout. Reading. Voelkl. A. R. Garcia. Demerouti. The joint effects of target and purpose goals on intrinsic motivation: A mediational analysis. M.. The mechanisms of job stress and strain. Work redesign. 511–523. Ellis. Journal of Personality and Social Psychology. & Rapson. coping and consequences of stress at work (pp. B. A. 235–258.. Frese. Fried. 256–337. (1998). (2001). L. J. 128. P. D. Discrimination and imitation of facial expressions by term and preterm neonates. G. D. S. W. (1993).. Optimal experience: Psychological studies of flow in consciousness.42 A. Flow: The psychology of optimal experience. 815–822. D. Psychological antecedents of the frequency and intensity of flow in golfers. F. Task characteristics and the experience of optimal flow in humancomputer interaction.. In C. Ghani. E. Validity and reliability of the experience sampling method. New York: HarperCollins. Personality and Social Psychology Bulletin. Measurement and analyses issues with explanation of variance in daily experience using the flow model. Finding flow: The psychology of engagement with everyday life. (1994). 490–498. J. The transformation into professionals into self-managing and partially self-designing contributors: Toward a theory of leadership making. & Sherry. & Haworth. R. the science of happiness and a proposal for a national index. Journal of management Systems. (Eds. American Psychologist. G. The Journal of Psychology. Harackiewicz. (2000). & Morris.. well-being and health. Clarke. The Netherlands: University of Maastricht. R. Infant Behaviour Development. L. & Schaufeli. S. British Journal of Psychology. D. 24. (1994). (1982).. 34–43.. Neff. T. R. Most people are happy. G. & Larson. J.. & Duda. Talented teenagers: The roots of success and failure. de Vries (Ed. M. Optimal experience in work and leisure. New York: Plenum Press. Personnel Psychology.. Journal of Leisure Research. (1990). Nachreiner. Job autonomy. (1988). M.. T. Caplan. Edwards. (1997). Psychological Science.). & Zapf.. New York: Cambridge University Press. D. Methodological issues in the study of work stress: Objective vs. In M. Diener. A. M.). Csikszentmihalyi. 181–185. 3. Causes. Payne (Eds. I. New York: Cambridge University Press. Field. R. (1982). E. New York: Wiley. J. 28. New York: Cambridge University Press.. & Uhl-Bein.. De Jonge. Jr. Chichester: Wiley. 86. R.& Csikszentmihalyi. J. L. & Ferris. 485–490.. (1989). Csikszentmihalyi.. The experience of psychopathology. & Harrison.. 25–39. 7. M. Maastricht. M. 14. 56. Bakker. 6. L. (1994). De Jonge. C. E. (1985). Csikszentmihalyi. S. A critical examination of the Demand-Control-Support Model from a work psychological perspective. 309–322. J. & Oldham. FathersÕ and mothersÕ responses to the faces and cries of normal and premature infants. (1997). subjective measurement of work stress and the question of longitudinal studies. Subjective well-being.. & Kompier. New York: HarperCollins. Unpublished Doctoral Dissertation. (1987). Cacioppo. MA: Addison Wesley. (1992). J.. M. J. & Diener. . Cohen. S. Csikszentmihalyi. 40. & Greenberg. International Journal of Stress Management. J. Donovan. Csikszentmihalyi. 657–689. M. Academy of Management Journal. & Ryan. V.. D. J. 381–391..... G. M.. Diener. 39. 499–512. K.. Hackman. R. & Whalen. R.B. ‘‘Flow’’ experience in the daily lives of sixth-form college students. P. (1997). & Elliot. G. (1996). Rathunde. A. (1994). (1996). Bakker / Journal of Vocational Behavior 66 (2005) 26–44 Catley. C. M.. 375–411). & LeFevre. Csikszentmihalyi. Journal of Applied Psychology. Deci.. Hatfield. R. Cooper & R. J. The validity of the job characteristics model: A review and meta-analysis. R. Woodson. L. 85. R. Leavitt. E. T. (1980). International Journal of Sport Psychology. (1988)..). Frodi. C. M. New York: Cambridge University Press. 292–339. Intrinsic motivation and self-determination in human behaviour. Emotional contagion. R. 26.. 4. A. R.. M...

& Fortier. Kimiecik. H. Van der Doef. Motivational determinants of flow: Contributions from selfdetermination theory. Communication Monographs. Journal of Sport and Exercise Psychology. The Netherlands: Swets Test Services. and adolescents. Moeller. 32. In G. In R. (2000). Administrative Science Quarterly. (1978). & Richards. & Carli. P. (1988). Csikszentmihalyi (Eds. The Journal of Social Psychology. Work and Stress. Balla. 83. Intense consumption experiences: Peaks. Die Lust sich musikalisch aus zu dr€ ucken [The passion to express oneself with music]. M. & Hau. F. R. & Schaufeli. Lisse. New York: Basic Books. concepts. 198–213. (1994). H. De Jonge.. C. Mass media and political thought (pp. issues. (1992).. and performance: A field experiment. T. and mental strain: Implications for job design. Dillard.). (1999). In S. P. E. CA: Sage. Perloff (Eds. 26. (1985). The Netherlands: Kluwer. 315–353). Human Relations. S..B.. Dordrecht. Van Vegchel. (1999). J. (1985). A. A.). H. (1995).. W.A. Sullivan. 144–160. Advanced structural equation modeling. L. (2002). R. R. Hoyle. J. J. S. performances. S. & Fitzgerald. Development and validation of a scale to measure optimal experience: The flow state scale. (1979). Massimini. R. Examining flow experiences in sport contexts: Conceptual issues and methodological concerns. M. L. Karasek. Music therapy in gerontology: A review and a projection. and applications (pp. 80–89. Optimal experience: Psychological studies of flow in consciousness. G. J. H. B. 85–116)... T. Schaufeli. D.. Marsh. De UBOS. fathers. ViewersÕ emotional and cognitive responses to televised images of political leaders. 87–114. & Marsh. M. and goals on reaction-time performance. 189–217. F.. American Journal of Psychology. 6. (1996).. C. Loher. 52–56.. Kim. E. Amsterdam: Thesis Publishers. . Jackson. Locke. J. The Job Demand-Control (-Support) Model and psychological wellbeing: A review of 20 years of empirical research. Somera. A. The effects of job enrichment on employee satisfaction. M. NJ: Lawrence Erlbaum. W. C. Job demands. & McHugo. H. Sociological Methods and Research. & Guiry. Bakker. Kowal. Structural equation modeling. Klausmeier. (1994). F. M. Hoyle (Ed. M. G. In M. A. F. R. motivation. N. and flows. 566–574. G.). A meta-analysis of the relation of job characteristics and job satisfaction. B. B. R. 139. Mahwah. Testing global and specific indicators of rewards in the Effort-Reward Imbalance Model: Does it make any difference? European Journal of Work and Organizational Psychology. feedback in relation to standards.. 70. J. B.. 24. & Sleight. Utrecht Burnout Scale.. Utrechtse Burnout Schaal. Schaufeli. Empathy. & Maes. A. (in press).. 55. (1984)..). De steun van de leiding: Een onderzoek naar het Leader Member Exchange model in de verpleging [Leader’s support: A study of the Leader Member Exchange model among nurses]. Noe. S. W. Conditions of happiness. Thousand Oaks. & Boomsma. Le Blanc. Lutz. issues and techniques (pp. Robustness studies in covariance structure modeling: An overview and meta analysis.. Music Therapy Perspectives. and their relationship with burnout and engagement: A multi sample study. Job demands. A. Csikszentmihalyi & I. 4. 1–15). Cambridge: Cambridge University Press. 329–367.. 13. 355–368. D. & Van Dierendonck. communication. A. Divergent realities: The emotional lives of mothers. and prosocial behavior. Manual]. Stiff. D. (1998). T. P. 11. H. Kraus & R. (1979). The structural equation modeling approach: Basic concepts and fundamental issues. Jones. Schumacker (Eds. CA: Sage. Larson. Journal of Applied Sport Psychology. St. Effects of knowledge of results. N. involvement.. R. J. Paper presented at the Winter Marketing EducatorsÕ Conference. Palmer. Masters. 18. 475–482. 403–421. W. Handleiding [The UBOS. B. The systematic measurement of flow in daily experience.. Orpen. Petersburg. (1997). Beverly Hills. B. Pay procedures and voluntary turnover: Does procedural justice matter? Psychological Reports. Hamburg: Rowohlt Verlag. Lanzetta. Journal of Organizational Behavior. Marcoulides & R. R. C. job resources. J. Bakker / Journal of Vocational Behavior 66 (2005) 26–44 43 Hoogland. 285–308... 17–35. A. W. FL (February). B. An evaluation of incremental fit indices: A clarification of mathematical and empirical properties. Veenhoven. (1994). & Stein. 81. K. J.. (1996). job decision latitude. L. Journal of Applied Psychology. (1988). (1968). (1989).. J.. & Bakker.

Bakker / Journal of Vocational Behavior 66 (2005) 26–44 Veenhoven. (1995).B. M. Human Relations. (1996). & Etzion. Unraveling the relationship of distress levels within couples: Common stressors. (1998). or crossover via social interactions? Human Relations. Journal of Organizational Behavior. 169–181. M. & Vinokur. 51. Stress and strain crossover. Oxford: Oxford University Press. M. Work. Westman. 717–751. 16. Crossover of stress. strain and resources from one spouse to another. (1987). (2001). Warr. A. Universiteit van Utrecht.. B. Westman. D. . Rede uitgesproken bij aanvaarding van Bijzonder Hoogleraarschap. 54.44 A. 137– 156. empathic reactions. Leefbaarheid van landen. R. unemployment and mental health.. Westman. P.