- Email: [email protected]
The athlete burnout syndrome: possible early signs
The athlete burnout syndrome: possible early signs SL Cresswell & RC Eklund School of Human Movement and Exercise Science, The University of Western Australia, Australia.
Burnout has been identified as a concern in elite sport. The aim of the current study was to examine relationships among proposed early signs and the athlete burnout syndrome. Quantitative methods (ie, questionnaire) were used to identify the level of burnout and perceptions of proposed early signs among New Zealand male semi- and fully-professional Rugby Union players (n= 199) aged 19 to 33 years (M= 25.19, SD= 2.98). Results indicated significant associations among proposed early signs and athlete burnout. Perceptions of adequate social support, competence and perceptions of control were negatively correlated with key characteristics of burnout. Perceived rugby and money hassles were positively correlated with key characteristics of burnout. This research provides support for the proposed early signs of burnout. Further research is needed to confirm these findings and investigate the nature of the relationship between these early signs and the burnout syndrome.
(J Sci Med Sport 2004;7:4:481-487)
Introduction B u r n o u t is currently a concern in rugby 1. While the term b u r n o u t is widely invoked in colloquial language, it is often misunderstood. B u r n o u t is a n experiential syndrome with three central characteristics including a n enduring sense of: (a) reduced accomplishment, (b) sport devaluation and (c) physical and emotional exhaustion 6. Recently we developed a practitioner's guide to the athlete b u r n o u t syndrome 1. Within the practitioner's guide several early signs to the b u r n o u t syndrome were proposed. The aim of the c u r r e n t research was to evaluate empirically the validity of these proposed early signs by assessing their associations with the key characteristics of the b u r n o u t syndrome. Establishing criterion validity for these proposed early signs of b u r n o u t could enable a proactive approach to the m a n a g e m e n t of the syndrome.
Early signs of burnout Early signs are proposed to indicate that a n athlete may be at risk of experiencing burnout. In contrast, proposed symptoms signify that a n athlete may actually be experiencing burnout. Symptoms refer to indicators of the key characteristics of b u r n o u t that athletes m a y experience. For example, nagging 481
The athlete burnout syndrome...
disappointment might be an early sign warning of the risk of impending b u r n o u t whereas enduring negative mood shifts would be classed as a symptom of emotional exhaustion. Five proposed early signs were included in the current study 1. These proposed early signs included competence related perceptions, perceptions of rugby-related hassles, perceptions of control over rugby career, satisfaction with social support and concerns a b o u t money-related hassles. We anticipated that perceptions of hassles (rugby-related or money-related) would be positively associated with b u r n o u t 5. In contrast past research suggests that perceptions of satisfaction with social support, a high degree of c o m p e t e n c e and perceptions of control should be associated with low levels of b u r n o u t 2,a.
Method Participants Semi- and fully-professional Rugby Union players (n= 199) aged 19 to 33 years (M= 25.2, SD= 3.0) were recruited from division one of the New Zealand National Provincial Competition. These men reported training a weekly average of 14.76 (+6.40) in-season hours, 16.46 (+7.8) pre-season h o u r s and 10.4 (+6.2) off-season hours. Players reported an average of 2.0 (+1.7) injuries that had prevented some training or playing in the current season. Instruments Athlete Burnout Questionnaire (ABQ) The Athlete B u r n o u t Questionnaire (ABQ) 6 has three 5-item subscales designed to measure: (a) reduced sense of accomplishment (eg, "It seems that no matter what I do, I don't perform as well as I should"); (b) devaluation (eg, "I have negative feelings towards sport"); and (c) emotional/physical exhaustion (eg, "I am exhausted by the mental and physical demands of my sport"). Item responses are collected on five-point Likert scales anchored by descriptors of "almost never" (1), "rarely" (2), "sometimes" (3), "frequently" (4) and "most of the time" (5). Acceptable internal consistency, test~retest reliability and construct validity have been previously reported 6. Internal consistency estimates observed in the present investigation (alphas between .77 and .82) were consistent with previous reports.
Potential early signs For this investigation, a 21-item inventory was constructed to m e a s u r e potential early signs of b u r n o u t (see Table 1). The i n s t r u m e n t contained five subscales designed to measure; (a) competence; (b) rugby hassles; (c) perceived control; (d) social support; and (e) money hassles. Items are responded to on a five-point Likert scale anchored by descriptors of "strongly disagree" (1), "disagree" (2), "neither agree nor disagree" (3), "agree" (4) and "strongly agree" (5}. Pilot work suggested that items had face validity among rugby players. Very low response variability for some competence and control subscale items was observed. These items were subsequently removed from further analyses. Principal-axis exploratory factor analyses (EFA) with direct oblim rotation were conducted specifying five factors. Cross loading items were removed to get a clean solution for s u b s e q u e n t analysis. A total of 16 items were retained in the final solution: Social Support (five items), Rugby Hassles (five items), Money
482
The athlete burnout syndrome...
M
DescriptiveD a t a SD Range
Money HasslesSubscale
2.74
1.06
1.0.5.0.
• Having enough money for basic needs (e.g., living expenses) was a hassle in the last month ° Financial security was a hassle in the last month
2.73
1.21
1.0- 5.0
2.75
1.11
1.0- 5.0
Rugby HasslesSubscale
2.72
0.63
1.0- 4.4.
, Not getting enough sleep/rest because of rugby commitments was a hassle in the last month • Too many responsibilities for rugby (eg, sponsorship duties) was a hassle in the last month , Too many extra rugby tasks (eg, watching video analysis) was a hassle in the last month ° Having too many rugby training sessions was a hassle in the last month ° Too many unexpected (extra) things to do for rugby was a hassle in the last month
2.73
1.00
1.0-5.0
2.67
0.89
1.0- 5.0
2.52
0.88
1.0- 5.0
2.80
0.82
1.0- 5.0
2,86
0.87
1.0- 5.0
Social Support Subscale
3.86
0.62
1.6- 5.0
, I am satisfied with the support people in my life give me when I am performing well (eg, winning, gaining selection in teams) ° I am satisfied with the support people in my life give me to console me when I am very upset or frustrated about rugby , I am satisfied with the support people in my life give me when I am under stress from rugby demands • I am satisfied with the support people in my life give me when my rugby performance is not going well • I am satisfied with the way support people in my life accept my highs (eg, wins) and lows (eg, losses, mistakes) in rugby
4.12
0.70
1.0- 5.0
3.84
0.84
1.0- 5.0
3.70
0.81
1.0- 5.0
3.75
0.84
1.0- 5.0
3.8
80.6
Correlations RA Exh Dev 21"*
.04
17"
.46** .26"*
-.42** -.17"
.04
.,30"*
1.0- 5.0
Competence Items ° I am able to make a contribution on the rugby field* • I have limitations in the rugby skills required for my position (reverse scored item) , In my position I can play well against anyone in my opposite number ^ ° [ am happy with my level of rugby skills
4.29 3.27
0.55 0.95
1.0- 5.0 1.0-5.0
-.10 +
.05 + -.10 +
4.13
0.65
1.0-5.0
-.26*
.02
-.18'
3.67
0.77
1.0- 5.0
-,23**
-.00
-.02
3.82 3.92
0.79 0,82
1.0- 5.0 2.0 - 5.0
-.42**
-.18'
-.29**
4.14 3.35
0.68 0.87
2.0-5.0 1.0- 5.0
4.30
0,62
2.0- 5.0
Control Items • I feel I have control of reaching my career goals in rugby ^ • I believe being successful at rugby is to do with luck (reverse scored item)* , My success on the rugby field will be determined by me • Despite my best efforts, my value as a rugby player will not be recognised (reverse scored item) ^ • I believe being successful at rugby is to do with hard work
-.11 -.17" -.16' ..28 +** -.30+**-.33 +** -.34**
-.13
-.28**
Note: RA = Reduced Accomplishment, Exh = Exhaustion, Dev = Devaluation. *items removed from subsequent analysis due to extremely low variability, ^ items removed following Exploratory Factor Analysis, +Correlations calculated after items were reversed, ** = significant correlation (p<.01), *=significant correlation (p<.05)
Table 1: Descriptive statistics and correlations for the proposed early signs of burnout.
483
The athlete burnout syndrome...
Hassles {two items), Competence (two items) and Control (two items). The r u g b y hassle, social s u p p o r t and m o n e y hassle subscales all displayed acceptable reliability (Cronbach's a l p h a coefficients between.77 and .85). The competence and control s u b s c a l e s did not display acceptable reliability (Cronbach's a l p h a coefficients of .48 and .50 respectively). Due to the low alphas correlations were calculated between b u r n o u t s u b s c a l e s and all of individual item in those subscales. The highest loading c o m p e t e n c e and control item on the respective factors was employed in the canonical analysis.
Procedures Permission was secured t h r o u g h telephone contact from t e a m coaches to a p p r o a c h players at t e a m training sessions. Players were a p p r o a c h e d on site immediately prior to or following t e a m training a n d invited to participate. Standardised introduction information and instructions were given to all participants, a n d written informed consent was obtained prior to administration of the questionnaire.
Statistical analysis These d a t a were analysed using bivariate correlations and a canonical correlation. A canonical correlation w a s performed between a key characteristic of b u r n o u t variate and an early signs variate using SPSS (version 9.0). Interpretation of significant canonical functions is best accomplished by examination of the canonical loadings-correlations between the variables a n d the canonical variates. Loadings above .30 are typically interpreted as meaningful contributors to the multivariate relationships 9.
Results Descriptive statistics and bivariate associations Descriptive statistics for ABQ s u b s c a l e s a n d proposed early signs m e a s u r e s as well as relevant bivariate correlations between these m e a s u r e s are p r e s e n t e d in Table 1. Observed m e a n s for the ABQ included Reduced Accomplishment 2.43 (SD=0.68), E x h a u s t i o n 2.32 (SD=0.72) and Devaluation 2.16 (SD=0.76). C o m p a r i s o n s with US collegiate athletes aged 17 to 23 from a range of different sports 6 reveal t h a t the college athletes h a d slightly higher levels of e x h a u s t i o n (M=2.62, SD-=0.86) and similar levels of reduced a c c o m p l i s h m e n t (M=2.37, SD=0.76) and sport devaluation (M=2.02, SD=-0.88). Significant correlations were present a m o n g proposed early signs and virtually all the key characteristics of b u r n o u t (see Table 1).
Canonical correlation A canonical correlation was p e r f o r m e d between the key characteristics of b u r n o u t and the proposed early signs. The canonical loadings and standardised canonical coefficients for the canonical variates are presented in Table 2. Of the three possible canonical functions, only two produced significant relationships. Dimension reduction analysis produced a significant Wilks L a m b d a = .55, F(15, 508.34)= 8.12, p< .001 with all three canonical correlations included and w h e n the first canonical correlation was removed, Wilks L a m b d a = .59, F(3, 184.00)= 42.18, 12< .001. S u b s e q u e n t steps in the dimension reductions analysis were not significant. The first canonical correlation
484
The athlete burnout syndrome...
Canonical Loadings
Standardised coefficient
Burnout
Reduced Accomplishment Emotional Exhaustion Sport Devaluation Percent of Variance Redundancy
-.69 -.86 -.72 .58 .18
-.38 -.64 -.26
Proposed Early Signs
Competence Rugby Hassles Control Social Support Money Hassles Percent of Variance Redundancy Correlation
.17 -.74 .32 .61 -.20 .29 .07 .56
.11 -.78 .20 .63 .18
Note: Control item number 3 (ie, "My success on the rugby field will be determined by me") and Competence item 4 (ie, '1 am happy with my level of rugby skills") were used in the canonical analyses as representative of the relevant constructs because of the low alpha coefficients observed in the subscales. Table 2: Canonical Ioadings and standardised canonical coefficients for burnout and early signs variates
was .58 (30% overlapping variance). Although significant, the second canonical correlation showed insignificant r e d u n d a n c y to be worthy of interpretation. The remaining canonical correlation was effectively zero. The canonical loadings for each variate are also presented in Table 2. Perceived control (.32) and satisfaction with social support (.61) were associated with a low frequency of b u r n o u t characteristics. Rugby hassles (.74) was associated with a high frequency of b u r n o u t characteristics. The competence and money hassles did not load above .3, b u t were associated with the early signs variate in the expected direction (positive and negative respectively).
Discussion T h e p u r p o s e of t h e c u r r e n t r e s e a r c h w a s to a s s e s s t h e a s s o c i a t i o n a m o n g
proposed early signs and the key characteristics of burnout. Results provide preliminary evidence of the relationship between some proposed early signs of b u r n o u t and the b u r n o u t syndrome itself. As predicted a high level of perceived rugby hassles was positively associated with key characteristics of burnout. Previous research within h u m a n care settings highlighted that work-related stressors predicted the experience of b u r n o u t in males a. Players' perceptions of money hassles were associated with a high level of reduced accomplishment. This finding supports previous research that has linked the perception of financial stress with indicators of illbeing (eg, anxiety and depression) 4. The m o n e y hassles subscale, however,
485
The athlete burnout syndrome...
should be interpreted with caution as it contained a minimal n u m b e r of items (n=2). Overall these associations s u p p o r t psychosocial theories t h a t propose b u r n o u t is one potential outcome of the stress process 8. While the results highlight a n association a m o n g these s t r e s s o r s and the key characteristics of b u r n o u t , they do not imply prediction or causation. While perceiving these hassles m a y lead to b u r n o u t , it is also possible t h a t athletes experience these hassles as a result of b u r n o u t . For example, money-related h a s s l e s could result in players experiencing enduring feelings of sport devaluation. Alternatively, players m a y perceive m o n e y h a s s l e s as salient b e c a u s e they have started to devalue their sport participation, losing intrinsic motivation and developing a n external focus. In addition it is possible t h a t b o t h b u r n o u t and the hassle m a y be a result of a third variable. For example, it is conceivable t h a t b o t h sport devaluation and m o n e y hassles m a y be the result of non-selection or injury. T h u s the fact t h a t there are significant correlations does not imply t h a t the early signs are in fact the c a u s e of b u r n o u t . Using self-determination theory, r e s e a r c h e r s have postulated t h a t the failure to m e e t needs of autonomy, c o m p e t e n c e and social affiliation m a y be related to individuals experiencing negative emotional states 7. In the c u r r e n t study, variables relating to the needs of autonomy, competence a n d social affiliation were related to key characteristics of b u r n o u t . For example, level of social s u p p o r t was negatively correlated with all key characteristics of b u r n o u t . This association is consistent with p a s t r e s e a r c h t h a t h a s identified social s u p p o r t as a buffer to the experience of b u r n o u t for m e n in general working situations 2. Preliminary evidence was also found for associations between players' perceptions of control, competence a n d the key characteristics of b u r n o u t . Results from the control and competence subscales, however, should be regarded as preliminary a n d interpreted with caution b e c a u s e single representative items were u s e d in the canonical correlation due to poor internal reliability of these subscales. For instance in the case of control, item 3 w a s chosen for the canonical correlation b e c a u s e it loaded the highest in the factor analysis a n d this m a y have b e e n conservative. Examination of Table 1 indicates t h a t item 3 was not correlated with b u r n o u t to the s a m e extent as other items (eg, item 5). These findings s u p p o r t previous r e s e a r c h t h a t h a s linked perceptions of a lack of perceived control to higher levels of b u r n o u t 3. Further r e s e a r c h is needed, however, to establish if perceptions of control are related to SDT's need for a u t o n o m y or other theoretical explanations of the b u r n o u t syndrome, s u c h as s t r e s s - b a s e d approaches. Overall, these results s u p p o r t elements within several possible theoretical explanations of b u r n o u t 8 and ill-being 7. Further examination of these theoretical contentions and those forwarded in the b u r n o u t guide proposed by Cresswell and Eklund 1 is warranted. Further r e s e a r c h could examine the guide's efficacy to predict key signs of b u r n o u t with athletes active in a range of sports and at different levels. Given limited specific a n d comprehensive inventories t h a t m e a s u r e the early signs a n d s y m p t o m s exist qualitative m e t h odology m a y be advantageous.
Authors' Note Scott Cresswell is now at the School of Sport, UNITEC New Zealand. This r e s e a r c h was funded by the New Zealand Rugby Union. The c u r r e n t report is
486
The athlete burnout syndrome...
based u p o n a data collection that also involved motivational constructs d o c u m e n t e d in Cresswell SL and Eklund RC. Motivation and b u r n o u t in professional rugby players. {Submitted but as yet unpublished}.
References 1. Cresswell SL. and Eklund RC. The athlete b u r n o u t syndrome: A practitioner's guide. New
Zealand Journal of Sports Medicine 2003;31:4-9. 2. Etzion D. Moderating effect of social support on the s t r e s s - b u r n o u t relationship. J Appl
Psychol 1984;69:615-622. 3. Glass DC, McKnight JD and Valdimarsdottir H. Depression, burnout, and perceptions of control in hospital nurses. J Consult Clin Psychol 1993;61:147-155. 4. Lange C and Byrd M. The relationship between perceptions of financial distress and feelings of psychological well-being in New Zealand university students. International Journal of
Adolescence & Youth 1998;7:193-209. 5. Price L and Spence SH. Burnout symptoms amongst drug and alcohol service employees: Gender differences in the interaction between work and home stressors. Anxiety, Stress and
Coping 1994;7:67-84. 6. Raedeke TD and Smith AL. Development and preliminary validation of an athlete b u r n o u t measure. Journal of Sport and Exercise Psychology 2001 ;23:281-306. 7. Ryan RM. and Deci EL. The darker and brighter sides of h u m a n existence: Basic psychological needs as a unifying concept. Psychological Inquiry 2000; 11:319-338. 8. Smith RE. Toward a cognitive-affective model of athletic burnout. Journal of Sport
Psychology 1986;8:36-50. 9. Tabachnick BC and Fidell LS. Using multivariate statistics (4th Ed.). Boston, USA. Allyn & Bacon. 2001.
487
Burnout has been identified as a concern in elite sport. The aim of the current study was to examine relationships among proposed early signs and the athlete burnout syndrome. Quantitative methods (ie, questionnaire) were used to identify the level of burnout and perceptions of proposed early signs among New Zealand male semi- and fully-professional Rugby Union players (n= 199) aged 19 to 33 years (M= 25.19, SD= 2.98). Results indicated significant associations among proposed early signs and athlete burnout. Perceptions of adequate social support, competence and perceptions of control were negatively correlated with key characteristics of burnout. Perceived rugby and money hassles were positively correlated with key characteristics of burnout. This research provides support for the proposed early signs of burnout. Further research is needed to confirm these findings and investigate the nature of the relationship between these early signs and the burnout syndrome.
(J Sci Med Sport 2004;7:4:481-487)
Introduction B u r n o u t is currently a concern in rugby 1. While the term b u r n o u t is widely invoked in colloquial language, it is often misunderstood. B u r n o u t is a n experiential syndrome with three central characteristics including a n enduring sense of: (a) reduced accomplishment, (b) sport devaluation and (c) physical and emotional exhaustion 6. Recently we developed a practitioner's guide to the athlete b u r n o u t syndrome 1. Within the practitioner's guide several early signs to the b u r n o u t syndrome were proposed. The aim of the c u r r e n t research was to evaluate empirically the validity of these proposed early signs by assessing their associations with the key characteristics of the b u r n o u t syndrome. Establishing criterion validity for these proposed early signs of b u r n o u t could enable a proactive approach to the m a n a g e m e n t of the syndrome.
Early signs of burnout Early signs are proposed to indicate that a n athlete may be at risk of experiencing burnout. In contrast, proposed symptoms signify that a n athlete may actually be experiencing burnout. Symptoms refer to indicators of the key characteristics of b u r n o u t that athletes m a y experience. For example, nagging 481
The athlete burnout syndrome...
disappointment might be an early sign warning of the risk of impending b u r n o u t whereas enduring negative mood shifts would be classed as a symptom of emotional exhaustion. Five proposed early signs were included in the current study 1. These proposed early signs included competence related perceptions, perceptions of rugby-related hassles, perceptions of control over rugby career, satisfaction with social support and concerns a b o u t money-related hassles. We anticipated that perceptions of hassles (rugby-related or money-related) would be positively associated with b u r n o u t 5. In contrast past research suggests that perceptions of satisfaction with social support, a high degree of c o m p e t e n c e and perceptions of control should be associated with low levels of b u r n o u t 2,a.
Method Participants Semi- and fully-professional Rugby Union players (n= 199) aged 19 to 33 years (M= 25.2, SD= 3.0) were recruited from division one of the New Zealand National Provincial Competition. These men reported training a weekly average of 14.76 (+6.40) in-season hours, 16.46 (+7.8) pre-season h o u r s and 10.4 (+6.2) off-season hours. Players reported an average of 2.0 (+1.7) injuries that had prevented some training or playing in the current season. Instruments Athlete Burnout Questionnaire (ABQ) The Athlete B u r n o u t Questionnaire (ABQ) 6 has three 5-item subscales designed to measure: (a) reduced sense of accomplishment (eg, "It seems that no matter what I do, I don't perform as well as I should"); (b) devaluation (eg, "I have negative feelings towards sport"); and (c) emotional/physical exhaustion (eg, "I am exhausted by the mental and physical demands of my sport"). Item responses are collected on five-point Likert scales anchored by descriptors of "almost never" (1), "rarely" (2), "sometimes" (3), "frequently" (4) and "most of the time" (5). Acceptable internal consistency, test~retest reliability and construct validity have been previously reported 6. Internal consistency estimates observed in the present investigation (alphas between .77 and .82) were consistent with previous reports.
Potential early signs For this investigation, a 21-item inventory was constructed to m e a s u r e potential early signs of b u r n o u t (see Table 1). The i n s t r u m e n t contained five subscales designed to measure; (a) competence; (b) rugby hassles; (c) perceived control; (d) social support; and (e) money hassles. Items are responded to on a five-point Likert scale anchored by descriptors of "strongly disagree" (1), "disagree" (2), "neither agree nor disagree" (3), "agree" (4) and "strongly agree" (5}. Pilot work suggested that items had face validity among rugby players. Very low response variability for some competence and control subscale items was observed. These items were subsequently removed from further analyses. Principal-axis exploratory factor analyses (EFA) with direct oblim rotation were conducted specifying five factors. Cross loading items were removed to get a clean solution for s u b s e q u e n t analysis. A total of 16 items were retained in the final solution: Social Support (five items), Rugby Hassles (five items), Money
482
The athlete burnout syndrome...
M
DescriptiveD a t a SD Range
Money HasslesSubscale
2.74
1.06
1.0.5.0.
• Having enough money for basic needs (e.g., living expenses) was a hassle in the last month ° Financial security was a hassle in the last month
2.73
1.21
1.0- 5.0
2.75
1.11
1.0- 5.0
Rugby HasslesSubscale
2.72
0.63
1.0- 4.4.
, Not getting enough sleep/rest because of rugby commitments was a hassle in the last month • Too many responsibilities for rugby (eg, sponsorship duties) was a hassle in the last month , Too many extra rugby tasks (eg, watching video analysis) was a hassle in the last month ° Having too many rugby training sessions was a hassle in the last month ° Too many unexpected (extra) things to do for rugby was a hassle in the last month
2.73
1.00
1.0-5.0
2.67
0.89
1.0- 5.0
2.52
0.88
1.0- 5.0
2.80
0.82
1.0- 5.0
2,86
0.87
1.0- 5.0
Social Support Subscale
3.86
0.62
1.6- 5.0
, I am satisfied with the support people in my life give me when I am performing well (eg, winning, gaining selection in teams) ° I am satisfied with the support people in my life give me to console me when I am very upset or frustrated about rugby , I am satisfied with the support people in my life give me when I am under stress from rugby demands • I am satisfied with the support people in my life give me when my rugby performance is not going well • I am satisfied with the way support people in my life accept my highs (eg, wins) and lows (eg, losses, mistakes) in rugby
4.12
0.70
1.0- 5.0
3.84
0.84
1.0- 5.0
3.70
0.81
1.0- 5.0
3.75
0.84
1.0- 5.0
3.8
80.6
Correlations RA Exh Dev 21"*
.04
17"
.46** .26"*
-.42** -.17"
.04
.,30"*
1.0- 5.0
Competence Items ° I am able to make a contribution on the rugby field* • I have limitations in the rugby skills required for my position (reverse scored item) , In my position I can play well against anyone in my opposite number ^ ° [ am happy with my level of rugby skills
4.29 3.27
0.55 0.95
1.0- 5.0 1.0-5.0
-.10 +
.05 + -.10 +
4.13
0.65
1.0-5.0
-.26*
.02
-.18'
3.67
0.77
1.0- 5.0
-,23**
-.00
-.02
3.82 3.92
0.79 0,82
1.0- 5.0 2.0 - 5.0
-.42**
-.18'
-.29**
4.14 3.35
0.68 0.87
2.0-5.0 1.0- 5.0
4.30
0,62
2.0- 5.0
Control Items • I feel I have control of reaching my career goals in rugby ^ • I believe being successful at rugby is to do with luck (reverse scored item)* , My success on the rugby field will be determined by me • Despite my best efforts, my value as a rugby player will not be recognised (reverse scored item) ^ • I believe being successful at rugby is to do with hard work
-.11 -.17" -.16' ..28 +** -.30+**-.33 +** -.34**
-.13
-.28**
Note: RA = Reduced Accomplishment, Exh = Exhaustion, Dev = Devaluation. *items removed from subsequent analysis due to extremely low variability, ^ items removed following Exploratory Factor Analysis, +Correlations calculated after items were reversed, ** = significant correlation (p<.01), *=significant correlation (p<.05)
Table 1: Descriptive statistics and correlations for the proposed early signs of burnout.
483
The athlete burnout syndrome...
Hassles {two items), Competence (two items) and Control (two items). The r u g b y hassle, social s u p p o r t and m o n e y hassle subscales all displayed acceptable reliability (Cronbach's a l p h a coefficients between.77 and .85). The competence and control s u b s c a l e s did not display acceptable reliability (Cronbach's a l p h a coefficients of .48 and .50 respectively). Due to the low alphas correlations were calculated between b u r n o u t s u b s c a l e s and all of individual item in those subscales. The highest loading c o m p e t e n c e and control item on the respective factors was employed in the canonical analysis.
Procedures Permission was secured t h r o u g h telephone contact from t e a m coaches to a p p r o a c h players at t e a m training sessions. Players were a p p r o a c h e d on site immediately prior to or following t e a m training a n d invited to participate. Standardised introduction information and instructions were given to all participants, a n d written informed consent was obtained prior to administration of the questionnaire.
Statistical analysis These d a t a were analysed using bivariate correlations and a canonical correlation. A canonical correlation w a s performed between a key characteristic of b u r n o u t variate and an early signs variate using SPSS (version 9.0). Interpretation of significant canonical functions is best accomplished by examination of the canonical loadings-correlations between the variables a n d the canonical variates. Loadings above .30 are typically interpreted as meaningful contributors to the multivariate relationships 9.
Results Descriptive statistics and bivariate associations Descriptive statistics for ABQ s u b s c a l e s a n d proposed early signs m e a s u r e s as well as relevant bivariate correlations between these m e a s u r e s are p r e s e n t e d in Table 1. Observed m e a n s for the ABQ included Reduced Accomplishment 2.43 (SD=0.68), E x h a u s t i o n 2.32 (SD=0.72) and Devaluation 2.16 (SD=0.76). C o m p a r i s o n s with US collegiate athletes aged 17 to 23 from a range of different sports 6 reveal t h a t the college athletes h a d slightly higher levels of e x h a u s t i o n (M=2.62, SD-=0.86) and similar levels of reduced a c c o m p l i s h m e n t (M=2.37, SD=0.76) and sport devaluation (M=2.02, SD=-0.88). Significant correlations were present a m o n g proposed early signs and virtually all the key characteristics of b u r n o u t (see Table 1).
Canonical correlation A canonical correlation was p e r f o r m e d between the key characteristics of b u r n o u t and the proposed early signs. The canonical loadings and standardised canonical coefficients for the canonical variates are presented in Table 2. Of the three possible canonical functions, only two produced significant relationships. Dimension reduction analysis produced a significant Wilks L a m b d a = .55, F(15, 508.34)= 8.12, p< .001 with all three canonical correlations included and w h e n the first canonical correlation was removed, Wilks L a m b d a = .59, F(3, 184.00)= 42.18, 12< .001. S u b s e q u e n t steps in the dimension reductions analysis were not significant. The first canonical correlation
484
The athlete burnout syndrome...
Canonical Loadings
Standardised coefficient
Burnout
Reduced Accomplishment Emotional Exhaustion Sport Devaluation Percent of Variance Redundancy
-.69 -.86 -.72 .58 .18
-.38 -.64 -.26
Proposed Early Signs
Competence Rugby Hassles Control Social Support Money Hassles Percent of Variance Redundancy Correlation
.17 -.74 .32 .61 -.20 .29 .07 .56
.11 -.78 .20 .63 .18
Note: Control item number 3 (ie, "My success on the rugby field will be determined by me") and Competence item 4 (ie, '1 am happy with my level of rugby skills") were used in the canonical analyses as representative of the relevant constructs because of the low alpha coefficients observed in the subscales. Table 2: Canonical Ioadings and standardised canonical coefficients for burnout and early signs variates
was .58 (30% overlapping variance). Although significant, the second canonical correlation showed insignificant r e d u n d a n c y to be worthy of interpretation. The remaining canonical correlation was effectively zero. The canonical loadings for each variate are also presented in Table 2. Perceived control (.32) and satisfaction with social support (.61) were associated with a low frequency of b u r n o u t characteristics. Rugby hassles (.74) was associated with a high frequency of b u r n o u t characteristics. The competence and money hassles did not load above .3, b u t were associated with the early signs variate in the expected direction (positive and negative respectively).
Discussion T h e p u r p o s e of t h e c u r r e n t r e s e a r c h w a s to a s s e s s t h e a s s o c i a t i o n a m o n g
proposed early signs and the key characteristics of burnout. Results provide preliminary evidence of the relationship between some proposed early signs of b u r n o u t and the b u r n o u t syndrome itself. As predicted a high level of perceived rugby hassles was positively associated with key characteristics of burnout. Previous research within h u m a n care settings highlighted that work-related stressors predicted the experience of b u r n o u t in males a. Players' perceptions of money hassles were associated with a high level of reduced accomplishment. This finding supports previous research that has linked the perception of financial stress with indicators of illbeing (eg, anxiety and depression) 4. The m o n e y hassles subscale, however,
485
The athlete burnout syndrome...
should be interpreted with caution as it contained a minimal n u m b e r of items (n=2). Overall these associations s u p p o r t psychosocial theories t h a t propose b u r n o u t is one potential outcome of the stress process 8. While the results highlight a n association a m o n g these s t r e s s o r s and the key characteristics of b u r n o u t , they do not imply prediction or causation. While perceiving these hassles m a y lead to b u r n o u t , it is also possible t h a t athletes experience these hassles as a result of b u r n o u t . For example, money-related h a s s l e s could result in players experiencing enduring feelings of sport devaluation. Alternatively, players m a y perceive m o n e y h a s s l e s as salient b e c a u s e they have started to devalue their sport participation, losing intrinsic motivation and developing a n external focus. In addition it is possible t h a t b o t h b u r n o u t and the hassle m a y be a result of a third variable. For example, it is conceivable t h a t b o t h sport devaluation and m o n e y hassles m a y be the result of non-selection or injury. T h u s the fact t h a t there are significant correlations does not imply t h a t the early signs are in fact the c a u s e of b u r n o u t . Using self-determination theory, r e s e a r c h e r s have postulated t h a t the failure to m e e t needs of autonomy, c o m p e t e n c e and social affiliation m a y be related to individuals experiencing negative emotional states 7. In the c u r r e n t study, variables relating to the needs of autonomy, competence a n d social affiliation were related to key characteristics of b u r n o u t . For example, level of social s u p p o r t was negatively correlated with all key characteristics of b u r n o u t . This association is consistent with p a s t r e s e a r c h t h a t h a s identified social s u p p o r t as a buffer to the experience of b u r n o u t for m e n in general working situations 2. Preliminary evidence was also found for associations between players' perceptions of control, competence a n d the key characteristics of b u r n o u t . Results from the control and competence subscales, however, should be regarded as preliminary a n d interpreted with caution b e c a u s e single representative items were u s e d in the canonical correlation due to poor internal reliability of these subscales. For instance in the case of control, item 3 w a s chosen for the canonical correlation b e c a u s e it loaded the highest in the factor analysis a n d this m a y have b e e n conservative. Examination of Table 1 indicates t h a t item 3 was not correlated with b u r n o u t to the s a m e extent as other items (eg, item 5). These findings s u p p o r t previous r e s e a r c h t h a t h a s linked perceptions of a lack of perceived control to higher levels of b u r n o u t 3. Further r e s e a r c h is needed, however, to establish if perceptions of control are related to SDT's need for a u t o n o m y or other theoretical explanations of the b u r n o u t syndrome, s u c h as s t r e s s - b a s e d approaches. Overall, these results s u p p o r t elements within several possible theoretical explanations of b u r n o u t 8 and ill-being 7. Further examination of these theoretical contentions and those forwarded in the b u r n o u t guide proposed by Cresswell and Eklund 1 is warranted. Further r e s e a r c h could examine the guide's efficacy to predict key signs of b u r n o u t with athletes active in a range of sports and at different levels. Given limited specific a n d comprehensive inventories t h a t m e a s u r e the early signs a n d s y m p t o m s exist qualitative m e t h odology m a y be advantageous.
Authors' Note Scott Cresswell is now at the School of Sport, UNITEC New Zealand. This r e s e a r c h was funded by the New Zealand Rugby Union. The c u r r e n t report is
486
The athlete burnout syndrome...
based u p o n a data collection that also involved motivational constructs d o c u m e n t e d in Cresswell SL and Eklund RC. Motivation and b u r n o u t in professional rugby players. {Submitted but as yet unpublished}.
References 1. Cresswell SL. and Eklund RC. The athlete b u r n o u t syndrome: A practitioner's guide. New
Zealand Journal of Sports Medicine 2003;31:4-9. 2. Etzion D. Moderating effect of social support on the s t r e s s - b u r n o u t relationship. J Appl
Psychol 1984;69:615-622. 3. Glass DC, McKnight JD and Valdimarsdottir H. Depression, burnout, and perceptions of control in hospital nurses. J Consult Clin Psychol 1993;61:147-155. 4. Lange C and Byrd M. The relationship between perceptions of financial distress and feelings of psychological well-being in New Zealand university students. International Journal of
Adolescence & Youth 1998;7:193-209. 5. Price L and Spence SH. Burnout symptoms amongst drug and alcohol service employees: Gender differences in the interaction between work and home stressors. Anxiety, Stress and
Coping 1994;7:67-84. 6. Raedeke TD and Smith AL. Development and preliminary validation of an athlete b u r n o u t measure. Journal of Sport and Exercise Psychology 2001 ;23:281-306. 7. Ryan RM. and Deci EL. The darker and brighter sides of h u m a n existence: Basic psychological needs as a unifying concept. Psychological Inquiry 2000; 11:319-338. 8. Smith RE. Toward a cognitive-affective model of athletic burnout. Journal of Sport
Psychology 1986;8:36-50. 9. Tabachnick BC and Fidell LS. Using multivariate statistics (4th Ed.). Boston, USA. Allyn & Bacon. 2001.
487