- Email: [email protected]
Anger rumination: an antecedent of athlete aggression?
Psychology of Sport and Exercise 5 (2004) 279–289 www.elsevier.com/locate/psychsport
Anger rumination: an antecedent of athlete aggression? J.P. Maxwell ∗ Institute of Human Performance, University of Hong Kong, 111-113 Pokfulam Road, Hong Kong Received 14 May 2002; received in revised form 18 September 2002; accepted 6 January 2003
Abstract Objectives. The aim of this study was to examine the relationship between anger rumination (the propensity to think almost obsessively over past experiences that have provoked negative affect in the form of anger) and athlete aggression. It was predicted that high levels of anger rumination would be associated with an increased propensity to aggress. Method. A questionnaire comprising the Anger Rumination Scale (Sukhodolsky, Golub, & Cromwell, 2001), aggression and demographic questions was distributed to 305 male and female competitive athletes of varying ability who represented several team and individual sports. Results. Principal component factor analysis revealed a single rumination factor rather than the fourfactor solution previously described. No differences in Anger Rumination Scale score were found between males and females, team and individual sport players or competitive level. Provocation and anger rumination were significantly correlated with athletes’ reported aggressive behaviour. Aggression was higher in males compared to females. Type of sport was also related to incidence of aggression; athletes who participated in individual sports reported lower levels of aggression than athletes who played team sports. Conclusions. It was concluded that provocation and anger rumination were significant predictors of subsequent aggression and suggestions for preventing rumination, such as thought stopping and thought switching, were made. 2003 Elsevier Ltd. All rights reserved. Keywords: Rumination; Aggression; Sport; Gender; Provocation
Introduction Sport brings individuals and groups together in direct competition. Direct competition can often lead to conflict and, as with many conflicts between competing individuals or groups, attempts ∗
Tel.: +856-2817-9576. E-mail address: [email protected] (J.P. Maxwell).
1469-0292/$ - see front matter 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1469-0292(03)00007-4
280
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
at hierarchical resolution may involve the use of aggression (Leith, 1982). Baron and Richardson (1994) define aggression as ‘…any form of behaviour directed toward the goal of harming or injuring another living being who is motivated to avoid such treatment (p. 7)’. Behaviour may be verbal or physical and must be directed at another person rather than an inanimate object. Throwing one’s racket to the floor or cursing one’s poor play would not be categorised as aggressive behaviour; rather, they would be signs of frustration or anger. When assessing aggressive behaviour in sport, however, the distinction between sanctioned and unsanctioned aggression must be recognised (Kerr, 1999). Kerr argues that only unsanctioned aggression is cause for concern and points out that many sports have incidences of aggression that are tolerated or informally accepted. Such behaviours, once accepted, become sanctioned even though they do not comply with the official rule structure. In soccer, for example, it is common practice for players to argue with officials. Tolerance of aggressive behaviour simply because it is common does not justify its use nor does it alter the fact that the recipient is often motivated to avoid such behaviour. Therefore, the definition of aggression adopted in this report will follow that suggested by Baron and Richardson with the addition of official endorsement. That is, aggression in sport is any behaviour, not recognised as legal within the official rules of conduct, directed towards an opponent, official, team-mate or spectator who is motivated to avoid such behaviour. This definition assumes, naturally, that behaviour is intentional and, potentially, reflects both hostile and instrumental aggression (Husman & Silva, 1984). Instrumental aggression is included within this definition because the intent to cause injury, which the recipient is unlikely to welcome, is present; however, not all instrumental aggression need fall within the definition. In the pugilistic sport of boxing, for example, where attempts to harm the opponent by punching are integral (instrumental) to the combatants’ success, biting, head butting or kicking an opponent would be considered aggressive acts. Additionally, informally sanctioned behaviour such as arguing with officials would also be considered aggressive if the official rules of the game identify it as unacceptable. Research examining the antecedents of aggression has led to the development of a number of theories. By far the most popular models used to examine aggression in sport are FrustrationAggression (Dollard, Doob, Miller, Mowrer, & Sears, 1939), Frustration-Aggression revised (Berkowitz, 1965, 1969, 1989; Baron & Richardson, 1994) and Social Learning Theory (Bandura, 1973). The Frustration-Aggression theory typifies early attempts to explain aggressive behaviour. As with many theories of behaviour its roots were firmly embedded in drive theory (Hull, 1951; Spence & Spence, 1966; Spence, 1956). Dollard et al., proposed that frustration resulted from the blocking of goals or desires and that the build up of frustration inevitably led to the behavioural expression of frustration through aggressive actions. It was the insistence of inevitability that led to the revision of this theory by, amongst others, Berkowitz (1965). Berkowitz pointed out that not all people respond to frustration with overt aggression; rather, situational cues and learned responses contribute to the probability of aggressive behaviour. We would not expect a Bishop to attack a parishioner with his crosier, for example, regardless of how frustrated he became with his flock. Berkowitz (1983, 1989) subsequently added cognitive factors to his model allowing emotional responses and personal motivation to contribute to the propensity for aggressive behaviour. The proposition that aggressive behaviour can be learned was championed by Albert Bandura’s (1973, 1983) Social Learning Theory. Whilst acknowledging the role of physiological, genetic and motivational factors (Baron & Richardson, 1994), Bandura stressed the importance of learned
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
281
behaviour, acquired through social interaction, to the expression of aggression. Bandura proposed that aggression is learned via observation or direct experience of aggressive acts, together with perceived or actual approval for acting aggressively. In a famous and well-reported experiment, Bandura demonstrated that children replicate the aggressive behaviour of adults who they have observed behaving aggressively towards a Bobo doll (Bandura, Ross, & Ross, 1963); supporting the notion that aggressive behaviour is mimicked. Social learning theory has received some support from research examining aggression in sport. Celozzi, Kazelskis and Gutsch (1981) found that watching a violent ice hockey match increased aggression in persons with high trait aggression scores but talking about violent hockey matches did not. This suggests, indirectly, that aggressive behaviour is learned via observation of others accomplishing their goals through the use of violence. Direct support was provided by Smith (1988) who observed young ice hockey players imitating the aggressive actions committed by professional players. Aggression in ice hockey is actively supported and encouraged; therefore, young children can quickly learn from ‘expert’ role models that aggression is an acceptable and often desired behaviour (Weinberg & Gould, 1999). The majority of aggressive acts, however, are not simple regurgitations of learned behaviour. The aggressor normally acts in response to perceived threat, and is influenced by the situation and various personal factors such as trait aggression, cognition and affect. Research examining the antecedents of athlete aggression in sport has, almost exclusively, focused on situational factors and performance outcome. Aggression has been associated with situational factors such as game location (Keltikangas-Jarvinen & Kelnonen, 1988; Lefebvre & Passer, 1974), home team advantage (McGuire, Courneya, & Widmeyer, 1992; Varca, 1980), competition level (Butt & Cox, 1992; Coulomb & Pfister, 1998), frequency of competition (Widmeyer & McGuire, 1997) and opposition aggression (Harrell, 1980; Russell, 1974). Due to the high frequency of aggressive acts, the bulk of research examining the relationship between aggression and performance has focused on the sport of ice hockey (e.g. McCarthy & Kelly, 1978; Russell, 1974). These studies have found positive relationships between the use of aggressive behaviour and success; however, others have insisted that aggression can only decrease individual performance (Gill, 1986; Silva, 1980; Wann, 1997). No attempt, however, has been made to identify cognitive processes that precede the expression of aggressive acts in sport. Berkowitz (1989) claimed that the experience of provocation, frustration or aversive stimuli lead to aggression through the generation of negative affect that is interpreted by the individual as anger (Berkowitz, 1988). If negative affect is interpreted as fear, the individual is more likely to demonstrate avoidance or escape behaviours. Within the general aggression literature, anger has been associated with increased incidence of aggressive forms of expression (Berkowitz, 1983) and recently the concept of ruminating over experiences that cause anger and the emotions they arouse has been implicated in the development of aggression (Averill, 1983; Spielberger, 1988). Rumination is described as a repetitive and often unavoidable process of dwelling on past experiences (Sukhodolsky, Golub & Cromwell, 2001) and may be associated with counterfactual thinking, the practice of inserting ‘what if?’ statements into appraisals of past events (Spellman & Mandel, 1999). However, counterfactual thinking need not involve repetition, a fundamental characteristic of rumination (Sukhodolsky, Golub & Cromwell, 2001). Counterfactual thinking involves imagining different outcomes to events that have occurred in the past. A football defender who allows an attacker, who subsequently scores, to pass him might imagine
282
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
what might have happened if he had fouled the player. He might imagine that the subsequent penalty may not have resulted in a goal and, thus, may come to believe that fouling in similar circumstances might be preferable. In this way, the propensity to aggress might increase and may be magnified by repeated rumination of counterfactual thoughts. Sukhodolsky, Golub and Cromwell (2001) devised a scale based on the premise that individuals have a lesser or greater disposition towards rumination or thinking over past events and emotions. The Anger Rumination Scale (ARS) was devised ‘…to measure the tendency to focus attention on angry moods, recall past anger experiences, and think about the causes and consequences of anger episodes (p. 689)’. The 19-item ARS was constructed after exploratory factor analysis of a pool of 25 items revealed four subscales (Angry Afterthoughts, Thoughts of Revenge, Angry Memories and Understanding of Causes) accounting for 54% of the total variance. High internal reliability (Cronbach’s Alpha = 0.93) and good test-retest reliability (r = 0.77) over a one-month period were reported. Confirmatory factor analysis, however, only partially supported the initial factor analysis and it was suggested that a single factor solution might be more appropriate. Sukhodolsky et al. reported higher scores on the Thoughts of Revenge subscale for males (mean = 1.88, sd = 0.59) compared to females (mean = 1.57, sd = 0.52). All other scores were similar across gender. High scores on the scale are purported to indicate a greater propensity towards anger rumination. Incorporating rumination into Berkowitz’s revised frustration-aggression model, it could be predicted that anger induced by frustration, followed by rumination of the event, will lead to a greater probability of aggression than without rumination. Thoughts of Revenge in response to provocation from an opponent seem particularly relevant in this respect. A number of studies have identified provocation as the clearest predictor of later aggressive behaviour in sport (Harrell, 1980; Russell, 1974). Retaliation in sport is often penalised more severely than the infraction that preceded it; the notoriously innocuous retaliation leading to David Beckham’s dismissal during the 1998 World Cup, for example, was regarded by many as the reason for England’s failure to progress. Therefore, if persons who have continual thoughts of revenge are more likely to aggress, the consequences for team performance could be catastrophic. It must be noted, however, that frustration need not be caused by provocation from an opponent; memories of past failures, current failure, point deficits, bad officiating or a host of other factors can initiate the chain of events that lead to aggression (Cox, 1994). The research presented here attempts to confirm that anger rumination contributes to the behavioural expression of frustration by athletes in the form of unsanctioned aggression. In addition, the contribution of gender, sport type (individual and team), competitive level, experience and provocation to aggressive behaviour were also considered to allow comparison with previous research. It was predicted that high ARS score, perceived provocation, participation in team sport and male gender would be associated with higher levels of unsanctioned aggression than would low ARS score, lack of perceived provocation, female gender and participation in individual sport. Method Participants Competitive athletes (n = 305) recruited from the staff and student populations of two UK universities completed a questionnaire assessing anger rumination and aggression. Ages ranged
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
283
from 18 to 32 years (mean = 21.54, sd = 2.71); 154 were male, 151 female; 178 (105 male and 73 female) competed in team sports such as football, hockey, rugby and netball, the remainder (49 male and 78 female) in predominantly individual sports such as running, tennis, rowing and martial arts. Apparatus and procedure A single questionnaire constructed from the 19 items of the Anger Rumination Scale, four questions about the expression of aggressive acts in sport (designed to reflect aggressive acts that might appear in virtually any sport), five demographic questions (sex, age, sport, level of participation [advanced, intermediate and novice] and number of years’ participation) and one question assessing the experience of aggressive acts (or provocation) was completed by all participants. Participants responded on a four-point scale ranging from 1 (almost never) to 4 (almost always). A confirmatory Principal Component Factor Analysis was performed using the 19 ARS items to discover whether the original factor solution could be applied to a sample of athletes. The four questions relating to aggression were then principal component factor analysed. Finally, enter and stepwise multiple regressions were performed to discover any relationships between reported aggression and rumination, provocation, type of sport (1 = team sport and 2 = individual sport), level of participation (1 = advanced, 2 = intermediate and 3 = novice), number of years participation, age and gender (1 = male and 2 = female). For the purpose of comparison with previous and future research, norms are reported in Table 1 for all sports with a sample size greater than 15, however, further analysis was not carried out. Table 1 Aggression scale and anger rumination scale norms for selected groups
Male athletes Female athletes Individual sport Team sport Advanced Intermediate Novice Athletics Badminton Football Hockey Netball Rugby Squash Swimming Tennis Other individual sports Other team sports
Anger rumination scale mean
Anger rumination scale SD
Aggression scale mean
Aggression scale SD
1.79 1.71 1.77 1.73 1.77 1.70 1.83 1.78 1.71 1.77 1.74 1.66 1.84 1.74 1.71 1.68 1.95 1.56
0.53 0.45 0.47 0.51 0.54 0.44 0.52 0.55 0.38 0.54 0.54 0.48 0.55 0.48 0.38 0.31 0.57 0.31
1.76 1.36 1.36 1.71 1.67 1.52 1.45 1.24 1.20 1.86 1.69 1.29 2.00 1.46 1.28 1.40 1.49 1.45
0.71 0.47 0.52 0.67 0.67 0.59 0.66 0.41 0.32 0.64 0.72 0.35 0.81 0.68 0.39 0.37 0.73 0.44
284
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
Results Anger rumination scale: factor analysis A high Cronbach’s Alpha Coefficient (a = 0.91) revealed adequate internal reliability. A principal component factor analysis was carried out on the 19 items of the Anger Rumination Scale. Means, standard deviations and factor loadings for individual items ranged from 0.50 to 0.77. The Kaiser-Meyer-Olkin Measure of Sampling Adequacy (0.94) and Bartlett’s Test of Sphericity (2138.47, p ⬍ 0.001) demonstrated adequate multivariate normality, replicating the findings of Sukhodolsky et al. (2001). Unlike Sukhodolsky et al., however, a two-factor solution was generated. Inspection of the factor loadings revealed that the second factor consisted of item 12 (‘I think about the reasons people treat me badly’), which also loaded heavily on the first factor; in addition, the eigenvalue for the second factor was 1.26, only marginally higher than the traditional cut off point. Attempts to replicate Sukhodolsky et al.’s original findings using Principal axis analysis with oblimin rotation and Kaiser Normalization revealed an identical factor structure; therefore, a single factor solution was accepted consisting of all 19 items. Anger rumination scale norms Anger Rumination Scale means and standard deviations for selected groups are reported in Table 1. One-way Analyses of Variance (ANOVA) were performed on each of the grouping factors to identify differences in the dependent variable, Anger Rumination Scale score. There was no significant effect of Gender (F(1,304) = 2.24, p = 0.14), partially supporting the findings of Sukhodolsky et al., who found gender differences for the Thoughts of Revenge subscale only. The mean values reported by athletes, however, are slightly lower than those reported by Sukhodolsky et al. for university students (Mean = 1.99 and 1.91, for male and female college students, respectively). No significant differences were found for Type of Sport (F(1,304) = 0.46, p = 0.50) or Level of Competition (F(1,304) = 1.66, p = 0.19). Aggression scale factor analysis The four items that comprise the Aggression Scale demonstrated adequate internal reliability for a short scale (a = 0.79). A principal component factor analysis was carried out on the four aggression items to ensure they measured the same construct. Means, standard deviations and factor loadings for individual items are shown in Table 2. The Kaiser-Meyer-Olkin Measure of Table 2 Factor loading and item means for the aggression scale Aggression scale
Factor loading Mean
SD
I get involved in angry verbal exchanges with opponents I intentionally anger my opponents I get involved in physical fights with opponents I argue with officials Provocation I am the victim of verbal or physical abuse
0.80 0.66 0.70 0.66
1.66 1.74 1.30 1.57
0.84 0.89 0.64 0.84
1.70
0.89
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
285
Sampling Adequacy (0.79) and Bartlett’s Test of Sphericity (354.02, p ⬍ 0.001) verified satisfactory multivariate normality. A single factor solution was generated with factor loadings ranging from 0.66 to 0.80. The four questions appear to measure the same construct and their face validity suggests that construct is aggressive behaviour; therefore, their mean was taken as a measure of aggressive behaviour (Aggression Level). Aggression level norms Aggression Level means and standard deviations for selected groups are reported in Table 1. One-way ANOVAs were carried out with Aggression level as the dependent measure and Gender, Type of Sport and Level of Competition as between group factors. A significant effect of Gender was found (F(1,304) = 33.83, p ⬍ 0.001) indicating a higher level of reported aggression for male athletes compared with female athletes. A significant difference was also found for Type of Sport with Team sport athletes reporting higher levels of aggression than individual sport athletes (F(1,304) = 25.52, p ⬍ 0.001). The effect of competition level, however, failed to reach significance, although it approached the 0.05 level (F(1,304) = 2.91, p = 0.056). The trend suggested a general increase in aggression level with increasing level of competition (Table 1). Regression Seven variables (Table 3) were entered into standard then stepwise multiple regressions together with Aggression Level as the dependent variable. The results of the stepwise regression are reported in Table 3. The eight variables accounted for 47.6% of the variance in Aggression Level. The stepwise regression identified four variables, Provocation, Anger Rumination, Gender and Type of Sport with significant unique contributions to Aggression Level variability; total model R2 = 0.47, (F(4,303) = 65.88, p ⬍ 0.001). The analysis revealed that Aggression Level is positively related to Provocation; specifically, persons who report high levels of aggressive behaviour report high levels of provocation, in the form of verbal or physical assaults. Aggression Level was also found to correlate positively with Score on the ARS, supporting the contention that rumination may increase the likelihood of aggressive behaviour. Aggression level correlated negatively with type of sport played, specifically, athletes who participated in team sport report higher levels of Table 3 Results of a stepwise multiple aggression examining antecedents of aggressive behaviour Variable
B
SE
t
P
Provocation Anger rumination scale Gender Type of sport Agea Competitive levela Length of participationa
0.34 0.31 ⫺0.20 ⫺0.13 0.02 ⫺0.08 0.07
0.03 0.06 0.06 0.06
9.91 5.28 ⫺3.54 ⫺2.22 0.52 ⫺1.84 1.56
⬍0.001 ⬍0.001 ⬍0.001 0.03 0.60 0.07 0.12
a
Variables not entered into stepwise multiple regression, values shown are for Enter method
286
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
aggression than athletes from individual sports. Males tend to report higher levels of aggression than females. Discussion The predictive validity of the Anger Rumination Scale was tested within the context of sport. It was predicted that high score on the ARS would be associated with high levels of reported aggression. In addition, the relationship between various demographic and situational factors and sports aggression were also examined. The results are consistent with the idea that ruminating on provocations and associated anger may lead to a greater risk of retaliation than when rumination is not present. The four-factor construction of the Anger Rumination Scale proposed by Sukhodolsky et al. (2001) was not replicated with the sample of British athletes presented here. Rather, a single factor solution consisting of all 19 items was discovered and is consistent with Sukhodolsky et al.’s confirmatory factor analysis, which also suggested a possible single factor solution. ARS score was unaffected by gender, sport or competitive level, suggesting that it is more likely to be a personality disposition than a learned response. Athletes reported slightly lower ARS scores than college students. This may indicate a lowering of rumination through sports participation, but this suggestion can only be inferred with the current data. The four questions designed to measure aggression demonstrated good internal consistency and loaded on the same factor (despite some questions appearing to be milder than others), which was utilised as a measure of aggressive behaviour. Males and team players reported higher levels of aggression than females and individual players, respectively. These findings are intuitively unsurprising; there is a large body of research that highlights the greater aggressive tendencies of males compared to females (Baron & Richardson, 1994; Eagly & Steffen, 1986). Players of team sport have far more opportunity to aggress against opponents, thus, their higher level of reported aggression is, again, unsurprising. The effect of competition level just failed to reach significance, although, there was a trend towards increasing aggression with higher competitive level, which supports previous findings that high level performers are more aggressive (Butt & Cox, 1992). A stepwise multiple regression was performed to establish which factors best predict aggression. It was found that provocation accounted for the greatest proportion of unique variance, followed by ARS score, gender and type of sport. The finding that provocation is a good predictor of aggression supports findings reported by Harrell (1980). Harrell found that the home team matched levels of aggression displayed by visiting basketball teams in a reciprocal manner. The finding that provocation increases retaliatory aggression is also well reported in the general aggression literature (e.g. Richardson, Vandenberg, & Humphries, 1986; Taylor, 1967). It is possible, however, that respondents in this study reported high levels of provocation simply in an attempt to justify their own high levels of aggression. A study of the temporal relationship between actual incidences of aggressive behaviour is needed to address this possibility. The data reported here suffer from a number of limitations. Firstly, the measure of aggression is rather crude and provides no insight into the relationship between rumination and the type of aggressive behaviour expressed by the athlete. Two types of aggression have been identified within
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
287
sport, namely, hostile (reactive) aggression and instrumental aggression (Geen, 2001; Husman & Silva, 1984). The questions used here to identify aggression are broad enough to be classed as measures of both hostile and instrumental aggression. It is possible that rumination will affect the expression of these two behaviours differently. Instrumental aggression, for example, need not require anger for its expression; therefore, one would expect anger rumination to play a minor role in its antecedence. Hostile aggression, conversely, is likely to be a highly emotive behaviour that may follow long periods of anger rumination. Indeed, so lengthy may be the period of rumination that it is sometimes impossible to ascertain the cause of the aggressive act unless the player admits, perhaps in a biography, intentionally harming the opponent. Anger rumination as a general construct may be predictive of other forms of behaviour that are expressed as a result of frustration or anger. Not all instances of frustration or anger result in aggression. Often, anger is expressed through cathartic acts that are not directed towards another person. In tennis, for example, rumination over perceived injustices or one’s poor performance might lead to expressions of anger by the non-aggressive act of breaking one’s racket on the ground (Hanegby & Tenenbaum, 2001). There has been relatively little research examining the cognitive factors that contribute to the occurrence of aggression in sport. In this respect the data reported here are a step forward. Whilst directions of causality cannot be confidently detailed with the design used here, it may be useful to speculate. Anger is seen to arise from frustration or provocation (the latter of which is deemed to require intention on the part of an opponent or aggressor, whereas, the former does not). This may lead directly or eventually, through the process of rumination, to the intention to aggress. Finally, when available coping skills are unavailable or insufficient and an opportunity arises, intention to aggress may be expressed as aggression. Situational cues, such as others acting aggressively or disapproving peers may intensify or moderate any of the stages. This conception is intuitively appealing and owes much to the work of Berkowitz (1965) and Sukhodolsky et al. (2001). However, it is possible that persons who are aggressive tend to ruminate more about anger provoking experiences, that is, rumination may follow rather than precede aggression. If this were the case, one might expect the content of ruminative thoughts to differ depending on the temporal relationship to the aggressive act. Thoughts of revenge might typify rumination prior to aggression, whereas, angry afterthoughts and understanding of causes may well predominate post-aggression ruminations. Despite the general lack of research relating the experience of anger or aggression to performance (see Widmeyer, (1984) for an example of the performance enhancing effects of aggressive behaviour), there is general agreement that the use of unsanctioned aggression in sport is deplorable (Weinberg & Gould, 1999; Widmeyer, 1984). For this reason, suggestions have been proposed to reduce the level of aggression in sport (Tenenbaum, Stewart, Singer, & Duda, 1997). These suggestions typically focus on punishment, education and rule change or stringent enforcement, although they have received criticism for unrealistic expectations and misdirected motivations (Kerr, 1999). The idea that rumination is a cognitive phenomenon typified by negative attributional patterns provides encouragement for teaching more appropriate cognitions in response to provocation. The ruminative modus operandi can be disrupted through the use of several techniques such as thoughtstopping and thought-switching. Thought-stopping involves recognising an inappropriate thought and (silently) yelling ‘STOP’ then breathing deeply and exhaling slowly whilst counting back-
288
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
wards or focusing on a non-aggressive/neutral image. This technique has proved successful in the treatment of various forms of rumination (Martin, 1982; Parenteau & Lamontagne, 1981) and may also be appropriate for combating anger rumination in aggressive athletes. Thought-switching involves replacing negative thoughts with positive ones, such as increasing the desire to win rather than exact revenge. Either method should prove capable of reducing rumination and, as a consequence, help combat the materialization of unwanted aggression in sport.
References Averill, J. R. (1983). Studies on anger and aggression: Implications for theories of emotion. American Psychologist, 38, 1145–1160. Bandura, A. (1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice-Hall. Bandura, A. (1983). Psychological mechanisms of aggression. In R. G. Geen, & E. I. Donnerstein (Eds.), (pp. 1–40). Aggression: Theoretical and empirical reviews, 1. New York: Academic Press. Bandura, A., Ross, D., & Ross, S. A. (1963). Imitation of film-mediated aggressive models. Journal of Abnormal and Social Psychology, 67, 601–607. Baron, R. A., & Richardson, D. R. (1994). Human aggression (2nd ed.). New York: Plenum Press. Berkowitz, L. (1965). The concept of aggressive drive: Some additional considerations. In L. Berkowitz (Ed.), (pp. 301–329). Advances in experimental psychology, 2. New York: Academic Press. Berkowitz, L. (1969). The frustration-aggression hypothesis revisited. In L. Berkowitz (Ed.), Roots of aggression (pp. 1–28). New York: Atherton Press. Berkowitz, L. (1983). The experience of anger as a parallel process in the display of impulsive ‘angry’ aggression. In R. G. Geen, & E. I. Donnerstein (Eds.), (pp. 103–133). Theoretical and empirical reviews, 1. New York: Academic Press. Berkowitz, L. (1988). Frustrations, appraisals and aversively stimulated aggression. Aggressive Behavior, 14, 3–11. Berkowitz, L. (1989). Frustration-aggression hypothesis: Examination and reformulation. Psychological Bulletin, 106, 59–73. Butt, D. S., & Cox, D. N. (1992). Motivational patterns in Davis Cup, University and recreational tennis players. International Journal of Sport Psychology, 23, 1–13. Celozzi, M. J., Kazelskis, R., & Gutsch, K. U. (1981). The relationship between viewing televised violence in ice hockey and subsequent levels of personal aggression. Journal of Sport Behavior, 4, 157–162. Coulomb, G., & Pfister, R. (1998). Aggressive behaviors in soccer as a function of competition level and time: A field study. Journal of Sport Behavior, 21, 222–231. Cox, R. H. (1994). Sport psychology: Concepts and applications. Dubuque, IA: Wm. C. Brown & Benchmark. Dollard, J., Doob, L., Miller, N., Mowrer, O. H., & Sears, R. R. (1939). Frustration and aggression. New Haven, CT: Yale University Press. Eagly, A. H., & Steffen, V. J. (1986). Gender and aggressive behaviour: A meta-analytic review of the social psychological literature. Psychological Bulletin, 100, 309–330. Geen, R. G. (2001). Human aggression (2nd ed.). Buckingham, UK: Open University Press. Gill, D. L. (1986). Psychological dynamics of sport. Champaign, IL: Human Kinetics. Hanegby, R., & Tenenbaum, G. (2001). Blame it on the racket: Norm-breaking behaviours among junior tennis players. Psychology of Sport and Exercise, 2, 117–134. Harrell, W. A. (1980). Aggression by high school basketball players: An observational study of the effects of opponent aggression and frustration-inducing factors. International Journal of Sport Psychology, 11, 290–298. Hull, C. L. (1951). Essentials of behavior. New Haven, CT: Yale University Press. Husman, B. F., & Silva, J. M. (1984). Aggression in sport: Definitional and theoretical considerations. In J. M. Silva, & R. S. Weinberg (Eds.), Psychological foundations of sport (pp. 246–260). Champaign, IL: Human Kinetics. Keltikangas-Jarvinen, L., & Kelnonen, M. (1988). Aggression, self-confidence and cardiovascular reactions in competitive performance in adolescent boys. Aggressive Behavior, 14, 245–254.
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
289
Kerr, J. H. (1999). The role of aggression and violence in sport: A rejoinder to the ISSP position stand. The Sport Psychologist, 13, 83–88. Lefebvre, L. M., & Passer, M. W. (1974). The effects of game location and importance on aggression in team sport. International Journal of Sport Psychology, 5, 102–110. Leith, L. M. (1982). The role of competition in the elicitation of aggression in sport. Journal of Sport Behavior, 5, 168–174. Martin, G. L. (1982). Thought-stopping and stimulus control to decrease persistent disturbing thoughts. Journal of Behavioural Therapy and Experimental Psychiatry, 13, 215–220. McCarthy, J. F., & Kelly, B. R. (1978). Aggressive behavior and its effect on performance over time in ice hockey athletes: An archival study. International Journal of Sport Psychology, 9, 90–96. McGuire, E. J., Courneya, K. S., & Widmeyer, W. N. (1992). Aggression as a potential mediator of the home advantage in professional ice hockey. Journal of Sport and Exercise Psychology, 14, 148–158. Parenteau, P., & Lamontagne, Y. (1981). The thought-stopping technique: A treatment for different types of ruminations? Canadian Journal of Psychiatry, 26, 192–195. Richardson, D. R., Vandenberg, R. J., & Humphries, S. A. (1986). Effect of power to harm on retaliative aggression among males and females. Journal of Research in Personality, 20, 402–419. Russell, G. W. (1974). Machiavellianism, locus of control, aggression, performance and precautionary behavior in ice hockey. Human Relations, 27, 825–837. Silva, J. M. (1980). Understanding aggressive behavior and its effects upon athletic performance. In W. F. Straub (Ed.), Sport psychology: An analysis of athlete behavior (pp. 177–186). Ithaca, NY: Mouvement. Smith, M. D. (1988). Interpersonal sources of violence in hockey: The influence of parents, coaches and teammates. In F. L. Smoll, R. A. Magill, & M. J. Ash (Eds.), Children in sport (3rd ed.) (pp. 301–313). Champaign, IL: Human Kinetics. Spellman, B. A., & Mandel, D. R. (1999). When possibility informs reality: Counterfactual thinking as a cue to causality. Current Directions in Psychological Science, 8, 120–123. Spence, J. T., & Spence, K. W. (1966). The motivational components of manifest anxiety: Drive and drive stimuli. In C. D. Spielberger (Ed.), Anxiety and behavior. New York: Academic Press. Spence, K. W. (1956). Behavior theory and conditioning. New Haven, CT: Yale University Press. Spielberger, C. D. (1988). Manual for the state-trait anger expression inventory (STAXI). Odessa, FL: Psychological Assessment Resources. Sukhodolsky, D. G., Golub, A., & Cromwell, E. N. (2001). Development and validation of the anger rumination scale. Personality and Individual Differences, 31, 689–700. Taylor, S. P. (1967). Aggressive behavior and physiological arousal as a function of provocation and the tendency to inhibit aggression. Journal of Personality, 35, 197–310. Tenenbaum, G., Stewart, E., Singer, R. N., & Duda, J. (1997). Aggression and violence in sport: An ISSP position stand. The Sport Psychologist, 11, 1–7. Varca, P. E. (1980). An analysis of home and away game performance of male college basketball teams. Journal of Sport Psychology, 2, 245–257. Wann, D. L. (1997). Sport psychology. Upper Saddle River, NJ: Prentice Hall, Inc. Weinberg, R. S., & Gould, D. (1999). Foundations of sport and exercise psychology. Champaign, IL: Human Kinetics. Widmeyer, W. N. (1984). Aggression-performance relationships in sport. In J. M. Silva, & R. S. Weinberg (Eds.), Psychological foundations of sport (pp. 274–286). Champaign, IL: Human Kinetics. Widmeyer, W. N., & McGuire, E. J. (1997). Frequency of competition and aggression in professional ice hockey. International Journal of Sport Psychology, 28, 57–66.
Anger rumination: an antecedent of athlete aggression? J.P. Maxwell ∗ Institute of Human Performance, University of Hong Kong, 111-113 Pokfulam Road, Hong Kong Received 14 May 2002; received in revised form 18 September 2002; accepted 6 January 2003
Abstract Objectives. The aim of this study was to examine the relationship between anger rumination (the propensity to think almost obsessively over past experiences that have provoked negative affect in the form of anger) and athlete aggression. It was predicted that high levels of anger rumination would be associated with an increased propensity to aggress. Method. A questionnaire comprising the Anger Rumination Scale (Sukhodolsky, Golub, & Cromwell, 2001), aggression and demographic questions was distributed to 305 male and female competitive athletes of varying ability who represented several team and individual sports. Results. Principal component factor analysis revealed a single rumination factor rather than the fourfactor solution previously described. No differences in Anger Rumination Scale score were found between males and females, team and individual sport players or competitive level. Provocation and anger rumination were significantly correlated with athletes’ reported aggressive behaviour. Aggression was higher in males compared to females. Type of sport was also related to incidence of aggression; athletes who participated in individual sports reported lower levels of aggression than athletes who played team sports. Conclusions. It was concluded that provocation and anger rumination were significant predictors of subsequent aggression and suggestions for preventing rumination, such as thought stopping and thought switching, were made. 2003 Elsevier Ltd. All rights reserved. Keywords: Rumination; Aggression; Sport; Gender; Provocation
Introduction Sport brings individuals and groups together in direct competition. Direct competition can often lead to conflict and, as with many conflicts between competing individuals or groups, attempts ∗
Tel.: +856-2817-9576. E-mail address: [email protected] (J.P. Maxwell).
1469-0292/$ - see front matter 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1469-0292(03)00007-4
280
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
at hierarchical resolution may involve the use of aggression (Leith, 1982). Baron and Richardson (1994) define aggression as ‘…any form of behaviour directed toward the goal of harming or injuring another living being who is motivated to avoid such treatment (p. 7)’. Behaviour may be verbal or physical and must be directed at another person rather than an inanimate object. Throwing one’s racket to the floor or cursing one’s poor play would not be categorised as aggressive behaviour; rather, they would be signs of frustration or anger. When assessing aggressive behaviour in sport, however, the distinction between sanctioned and unsanctioned aggression must be recognised (Kerr, 1999). Kerr argues that only unsanctioned aggression is cause for concern and points out that many sports have incidences of aggression that are tolerated or informally accepted. Such behaviours, once accepted, become sanctioned even though they do not comply with the official rule structure. In soccer, for example, it is common practice for players to argue with officials. Tolerance of aggressive behaviour simply because it is common does not justify its use nor does it alter the fact that the recipient is often motivated to avoid such behaviour. Therefore, the definition of aggression adopted in this report will follow that suggested by Baron and Richardson with the addition of official endorsement. That is, aggression in sport is any behaviour, not recognised as legal within the official rules of conduct, directed towards an opponent, official, team-mate or spectator who is motivated to avoid such behaviour. This definition assumes, naturally, that behaviour is intentional and, potentially, reflects both hostile and instrumental aggression (Husman & Silva, 1984). Instrumental aggression is included within this definition because the intent to cause injury, which the recipient is unlikely to welcome, is present; however, not all instrumental aggression need fall within the definition. In the pugilistic sport of boxing, for example, where attempts to harm the opponent by punching are integral (instrumental) to the combatants’ success, biting, head butting or kicking an opponent would be considered aggressive acts. Additionally, informally sanctioned behaviour such as arguing with officials would also be considered aggressive if the official rules of the game identify it as unacceptable. Research examining the antecedents of aggression has led to the development of a number of theories. By far the most popular models used to examine aggression in sport are FrustrationAggression (Dollard, Doob, Miller, Mowrer, & Sears, 1939), Frustration-Aggression revised (Berkowitz, 1965, 1969, 1989; Baron & Richardson, 1994) and Social Learning Theory (Bandura, 1973). The Frustration-Aggression theory typifies early attempts to explain aggressive behaviour. As with many theories of behaviour its roots were firmly embedded in drive theory (Hull, 1951; Spence & Spence, 1966; Spence, 1956). Dollard et al., proposed that frustration resulted from the blocking of goals or desires and that the build up of frustration inevitably led to the behavioural expression of frustration through aggressive actions. It was the insistence of inevitability that led to the revision of this theory by, amongst others, Berkowitz (1965). Berkowitz pointed out that not all people respond to frustration with overt aggression; rather, situational cues and learned responses contribute to the probability of aggressive behaviour. We would not expect a Bishop to attack a parishioner with his crosier, for example, regardless of how frustrated he became with his flock. Berkowitz (1983, 1989) subsequently added cognitive factors to his model allowing emotional responses and personal motivation to contribute to the propensity for aggressive behaviour. The proposition that aggressive behaviour can be learned was championed by Albert Bandura’s (1973, 1983) Social Learning Theory. Whilst acknowledging the role of physiological, genetic and motivational factors (Baron & Richardson, 1994), Bandura stressed the importance of learned
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
281
behaviour, acquired through social interaction, to the expression of aggression. Bandura proposed that aggression is learned via observation or direct experience of aggressive acts, together with perceived or actual approval for acting aggressively. In a famous and well-reported experiment, Bandura demonstrated that children replicate the aggressive behaviour of adults who they have observed behaving aggressively towards a Bobo doll (Bandura, Ross, & Ross, 1963); supporting the notion that aggressive behaviour is mimicked. Social learning theory has received some support from research examining aggression in sport. Celozzi, Kazelskis and Gutsch (1981) found that watching a violent ice hockey match increased aggression in persons with high trait aggression scores but talking about violent hockey matches did not. This suggests, indirectly, that aggressive behaviour is learned via observation of others accomplishing their goals through the use of violence. Direct support was provided by Smith (1988) who observed young ice hockey players imitating the aggressive actions committed by professional players. Aggression in ice hockey is actively supported and encouraged; therefore, young children can quickly learn from ‘expert’ role models that aggression is an acceptable and often desired behaviour (Weinberg & Gould, 1999). The majority of aggressive acts, however, are not simple regurgitations of learned behaviour. The aggressor normally acts in response to perceived threat, and is influenced by the situation and various personal factors such as trait aggression, cognition and affect. Research examining the antecedents of athlete aggression in sport has, almost exclusively, focused on situational factors and performance outcome. Aggression has been associated with situational factors such as game location (Keltikangas-Jarvinen & Kelnonen, 1988; Lefebvre & Passer, 1974), home team advantage (McGuire, Courneya, & Widmeyer, 1992; Varca, 1980), competition level (Butt & Cox, 1992; Coulomb & Pfister, 1998), frequency of competition (Widmeyer & McGuire, 1997) and opposition aggression (Harrell, 1980; Russell, 1974). Due to the high frequency of aggressive acts, the bulk of research examining the relationship between aggression and performance has focused on the sport of ice hockey (e.g. McCarthy & Kelly, 1978; Russell, 1974). These studies have found positive relationships between the use of aggressive behaviour and success; however, others have insisted that aggression can only decrease individual performance (Gill, 1986; Silva, 1980; Wann, 1997). No attempt, however, has been made to identify cognitive processes that precede the expression of aggressive acts in sport. Berkowitz (1989) claimed that the experience of provocation, frustration or aversive stimuli lead to aggression through the generation of negative affect that is interpreted by the individual as anger (Berkowitz, 1988). If negative affect is interpreted as fear, the individual is more likely to demonstrate avoidance or escape behaviours. Within the general aggression literature, anger has been associated with increased incidence of aggressive forms of expression (Berkowitz, 1983) and recently the concept of ruminating over experiences that cause anger and the emotions they arouse has been implicated in the development of aggression (Averill, 1983; Spielberger, 1988). Rumination is described as a repetitive and often unavoidable process of dwelling on past experiences (Sukhodolsky, Golub & Cromwell, 2001) and may be associated with counterfactual thinking, the practice of inserting ‘what if?’ statements into appraisals of past events (Spellman & Mandel, 1999). However, counterfactual thinking need not involve repetition, a fundamental characteristic of rumination (Sukhodolsky, Golub & Cromwell, 2001). Counterfactual thinking involves imagining different outcomes to events that have occurred in the past. A football defender who allows an attacker, who subsequently scores, to pass him might imagine
282
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
what might have happened if he had fouled the player. He might imagine that the subsequent penalty may not have resulted in a goal and, thus, may come to believe that fouling in similar circumstances might be preferable. In this way, the propensity to aggress might increase and may be magnified by repeated rumination of counterfactual thoughts. Sukhodolsky, Golub and Cromwell (2001) devised a scale based on the premise that individuals have a lesser or greater disposition towards rumination or thinking over past events and emotions. The Anger Rumination Scale (ARS) was devised ‘…to measure the tendency to focus attention on angry moods, recall past anger experiences, and think about the causes and consequences of anger episodes (p. 689)’. The 19-item ARS was constructed after exploratory factor analysis of a pool of 25 items revealed four subscales (Angry Afterthoughts, Thoughts of Revenge, Angry Memories and Understanding of Causes) accounting for 54% of the total variance. High internal reliability (Cronbach’s Alpha = 0.93) and good test-retest reliability (r = 0.77) over a one-month period were reported. Confirmatory factor analysis, however, only partially supported the initial factor analysis and it was suggested that a single factor solution might be more appropriate. Sukhodolsky et al. reported higher scores on the Thoughts of Revenge subscale for males (mean = 1.88, sd = 0.59) compared to females (mean = 1.57, sd = 0.52). All other scores were similar across gender. High scores on the scale are purported to indicate a greater propensity towards anger rumination. Incorporating rumination into Berkowitz’s revised frustration-aggression model, it could be predicted that anger induced by frustration, followed by rumination of the event, will lead to a greater probability of aggression than without rumination. Thoughts of Revenge in response to provocation from an opponent seem particularly relevant in this respect. A number of studies have identified provocation as the clearest predictor of later aggressive behaviour in sport (Harrell, 1980; Russell, 1974). Retaliation in sport is often penalised more severely than the infraction that preceded it; the notoriously innocuous retaliation leading to David Beckham’s dismissal during the 1998 World Cup, for example, was regarded by many as the reason for England’s failure to progress. Therefore, if persons who have continual thoughts of revenge are more likely to aggress, the consequences for team performance could be catastrophic. It must be noted, however, that frustration need not be caused by provocation from an opponent; memories of past failures, current failure, point deficits, bad officiating or a host of other factors can initiate the chain of events that lead to aggression (Cox, 1994). The research presented here attempts to confirm that anger rumination contributes to the behavioural expression of frustration by athletes in the form of unsanctioned aggression. In addition, the contribution of gender, sport type (individual and team), competitive level, experience and provocation to aggressive behaviour were also considered to allow comparison with previous research. It was predicted that high ARS score, perceived provocation, participation in team sport and male gender would be associated with higher levels of unsanctioned aggression than would low ARS score, lack of perceived provocation, female gender and participation in individual sport. Method Participants Competitive athletes (n = 305) recruited from the staff and student populations of two UK universities completed a questionnaire assessing anger rumination and aggression. Ages ranged
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
283
from 18 to 32 years (mean = 21.54, sd = 2.71); 154 were male, 151 female; 178 (105 male and 73 female) competed in team sports such as football, hockey, rugby and netball, the remainder (49 male and 78 female) in predominantly individual sports such as running, tennis, rowing and martial arts. Apparatus and procedure A single questionnaire constructed from the 19 items of the Anger Rumination Scale, four questions about the expression of aggressive acts in sport (designed to reflect aggressive acts that might appear in virtually any sport), five demographic questions (sex, age, sport, level of participation [advanced, intermediate and novice] and number of years’ participation) and one question assessing the experience of aggressive acts (or provocation) was completed by all participants. Participants responded on a four-point scale ranging from 1 (almost never) to 4 (almost always). A confirmatory Principal Component Factor Analysis was performed using the 19 ARS items to discover whether the original factor solution could be applied to a sample of athletes. The four questions relating to aggression were then principal component factor analysed. Finally, enter and stepwise multiple regressions were performed to discover any relationships between reported aggression and rumination, provocation, type of sport (1 = team sport and 2 = individual sport), level of participation (1 = advanced, 2 = intermediate and 3 = novice), number of years participation, age and gender (1 = male and 2 = female). For the purpose of comparison with previous and future research, norms are reported in Table 1 for all sports with a sample size greater than 15, however, further analysis was not carried out. Table 1 Aggression scale and anger rumination scale norms for selected groups
Male athletes Female athletes Individual sport Team sport Advanced Intermediate Novice Athletics Badminton Football Hockey Netball Rugby Squash Swimming Tennis Other individual sports Other team sports
Anger rumination scale mean
Anger rumination scale SD
Aggression scale mean
Aggression scale SD
1.79 1.71 1.77 1.73 1.77 1.70 1.83 1.78 1.71 1.77 1.74 1.66 1.84 1.74 1.71 1.68 1.95 1.56
0.53 0.45 0.47 0.51 0.54 0.44 0.52 0.55 0.38 0.54 0.54 0.48 0.55 0.48 0.38 0.31 0.57 0.31
1.76 1.36 1.36 1.71 1.67 1.52 1.45 1.24 1.20 1.86 1.69 1.29 2.00 1.46 1.28 1.40 1.49 1.45
0.71 0.47 0.52 0.67 0.67 0.59 0.66 0.41 0.32 0.64 0.72 0.35 0.81 0.68 0.39 0.37 0.73 0.44
284
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
Results Anger rumination scale: factor analysis A high Cronbach’s Alpha Coefficient (a = 0.91) revealed adequate internal reliability. A principal component factor analysis was carried out on the 19 items of the Anger Rumination Scale. Means, standard deviations and factor loadings for individual items ranged from 0.50 to 0.77. The Kaiser-Meyer-Olkin Measure of Sampling Adequacy (0.94) and Bartlett’s Test of Sphericity (2138.47, p ⬍ 0.001) demonstrated adequate multivariate normality, replicating the findings of Sukhodolsky et al. (2001). Unlike Sukhodolsky et al., however, a two-factor solution was generated. Inspection of the factor loadings revealed that the second factor consisted of item 12 (‘I think about the reasons people treat me badly’), which also loaded heavily on the first factor; in addition, the eigenvalue for the second factor was 1.26, only marginally higher than the traditional cut off point. Attempts to replicate Sukhodolsky et al.’s original findings using Principal axis analysis with oblimin rotation and Kaiser Normalization revealed an identical factor structure; therefore, a single factor solution was accepted consisting of all 19 items. Anger rumination scale norms Anger Rumination Scale means and standard deviations for selected groups are reported in Table 1. One-way Analyses of Variance (ANOVA) were performed on each of the grouping factors to identify differences in the dependent variable, Anger Rumination Scale score. There was no significant effect of Gender (F(1,304) = 2.24, p = 0.14), partially supporting the findings of Sukhodolsky et al., who found gender differences for the Thoughts of Revenge subscale only. The mean values reported by athletes, however, are slightly lower than those reported by Sukhodolsky et al. for university students (Mean = 1.99 and 1.91, for male and female college students, respectively). No significant differences were found for Type of Sport (F(1,304) = 0.46, p = 0.50) or Level of Competition (F(1,304) = 1.66, p = 0.19). Aggression scale factor analysis The four items that comprise the Aggression Scale demonstrated adequate internal reliability for a short scale (a = 0.79). A principal component factor analysis was carried out on the four aggression items to ensure they measured the same construct. Means, standard deviations and factor loadings for individual items are shown in Table 2. The Kaiser-Meyer-Olkin Measure of Table 2 Factor loading and item means for the aggression scale Aggression scale
Factor loading Mean
SD
I get involved in angry verbal exchanges with opponents I intentionally anger my opponents I get involved in physical fights with opponents I argue with officials Provocation I am the victim of verbal or physical abuse
0.80 0.66 0.70 0.66
1.66 1.74 1.30 1.57
0.84 0.89 0.64 0.84
1.70
0.89
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
285
Sampling Adequacy (0.79) and Bartlett’s Test of Sphericity (354.02, p ⬍ 0.001) verified satisfactory multivariate normality. A single factor solution was generated with factor loadings ranging from 0.66 to 0.80. The four questions appear to measure the same construct and their face validity suggests that construct is aggressive behaviour; therefore, their mean was taken as a measure of aggressive behaviour (Aggression Level). Aggression level norms Aggression Level means and standard deviations for selected groups are reported in Table 1. One-way ANOVAs were carried out with Aggression level as the dependent measure and Gender, Type of Sport and Level of Competition as between group factors. A significant effect of Gender was found (F(1,304) = 33.83, p ⬍ 0.001) indicating a higher level of reported aggression for male athletes compared with female athletes. A significant difference was also found for Type of Sport with Team sport athletes reporting higher levels of aggression than individual sport athletes (F(1,304) = 25.52, p ⬍ 0.001). The effect of competition level, however, failed to reach significance, although it approached the 0.05 level (F(1,304) = 2.91, p = 0.056). The trend suggested a general increase in aggression level with increasing level of competition (Table 1). Regression Seven variables (Table 3) were entered into standard then stepwise multiple regressions together with Aggression Level as the dependent variable. The results of the stepwise regression are reported in Table 3. The eight variables accounted for 47.6% of the variance in Aggression Level. The stepwise regression identified four variables, Provocation, Anger Rumination, Gender and Type of Sport with significant unique contributions to Aggression Level variability; total model R2 = 0.47, (F(4,303) = 65.88, p ⬍ 0.001). The analysis revealed that Aggression Level is positively related to Provocation; specifically, persons who report high levels of aggressive behaviour report high levels of provocation, in the form of verbal or physical assaults. Aggression Level was also found to correlate positively with Score on the ARS, supporting the contention that rumination may increase the likelihood of aggressive behaviour. Aggression level correlated negatively with type of sport played, specifically, athletes who participated in team sport report higher levels of Table 3 Results of a stepwise multiple aggression examining antecedents of aggressive behaviour Variable
B
SE
t
P
Provocation Anger rumination scale Gender Type of sport Agea Competitive levela Length of participationa
0.34 0.31 ⫺0.20 ⫺0.13 0.02 ⫺0.08 0.07
0.03 0.06 0.06 0.06
9.91 5.28 ⫺3.54 ⫺2.22 0.52 ⫺1.84 1.56
⬍0.001 ⬍0.001 ⬍0.001 0.03 0.60 0.07 0.12
a
Variables not entered into stepwise multiple regression, values shown are for Enter method
286
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
aggression than athletes from individual sports. Males tend to report higher levels of aggression than females. Discussion The predictive validity of the Anger Rumination Scale was tested within the context of sport. It was predicted that high score on the ARS would be associated with high levels of reported aggression. In addition, the relationship between various demographic and situational factors and sports aggression were also examined. The results are consistent with the idea that ruminating on provocations and associated anger may lead to a greater risk of retaliation than when rumination is not present. The four-factor construction of the Anger Rumination Scale proposed by Sukhodolsky et al. (2001) was not replicated with the sample of British athletes presented here. Rather, a single factor solution consisting of all 19 items was discovered and is consistent with Sukhodolsky et al.’s confirmatory factor analysis, which also suggested a possible single factor solution. ARS score was unaffected by gender, sport or competitive level, suggesting that it is more likely to be a personality disposition than a learned response. Athletes reported slightly lower ARS scores than college students. This may indicate a lowering of rumination through sports participation, but this suggestion can only be inferred with the current data. The four questions designed to measure aggression demonstrated good internal consistency and loaded on the same factor (despite some questions appearing to be milder than others), which was utilised as a measure of aggressive behaviour. Males and team players reported higher levels of aggression than females and individual players, respectively. These findings are intuitively unsurprising; there is a large body of research that highlights the greater aggressive tendencies of males compared to females (Baron & Richardson, 1994; Eagly & Steffen, 1986). Players of team sport have far more opportunity to aggress against opponents, thus, their higher level of reported aggression is, again, unsurprising. The effect of competition level just failed to reach significance, although, there was a trend towards increasing aggression with higher competitive level, which supports previous findings that high level performers are more aggressive (Butt & Cox, 1992). A stepwise multiple regression was performed to establish which factors best predict aggression. It was found that provocation accounted for the greatest proportion of unique variance, followed by ARS score, gender and type of sport. The finding that provocation is a good predictor of aggression supports findings reported by Harrell (1980). Harrell found that the home team matched levels of aggression displayed by visiting basketball teams in a reciprocal manner. The finding that provocation increases retaliatory aggression is also well reported in the general aggression literature (e.g. Richardson, Vandenberg, & Humphries, 1986; Taylor, 1967). It is possible, however, that respondents in this study reported high levels of provocation simply in an attempt to justify their own high levels of aggression. A study of the temporal relationship between actual incidences of aggressive behaviour is needed to address this possibility. The data reported here suffer from a number of limitations. Firstly, the measure of aggression is rather crude and provides no insight into the relationship between rumination and the type of aggressive behaviour expressed by the athlete. Two types of aggression have been identified within
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
287
sport, namely, hostile (reactive) aggression and instrumental aggression (Geen, 2001; Husman & Silva, 1984). The questions used here to identify aggression are broad enough to be classed as measures of both hostile and instrumental aggression. It is possible that rumination will affect the expression of these two behaviours differently. Instrumental aggression, for example, need not require anger for its expression; therefore, one would expect anger rumination to play a minor role in its antecedence. Hostile aggression, conversely, is likely to be a highly emotive behaviour that may follow long periods of anger rumination. Indeed, so lengthy may be the period of rumination that it is sometimes impossible to ascertain the cause of the aggressive act unless the player admits, perhaps in a biography, intentionally harming the opponent. Anger rumination as a general construct may be predictive of other forms of behaviour that are expressed as a result of frustration or anger. Not all instances of frustration or anger result in aggression. Often, anger is expressed through cathartic acts that are not directed towards another person. In tennis, for example, rumination over perceived injustices or one’s poor performance might lead to expressions of anger by the non-aggressive act of breaking one’s racket on the ground (Hanegby & Tenenbaum, 2001). There has been relatively little research examining the cognitive factors that contribute to the occurrence of aggression in sport. In this respect the data reported here are a step forward. Whilst directions of causality cannot be confidently detailed with the design used here, it may be useful to speculate. Anger is seen to arise from frustration or provocation (the latter of which is deemed to require intention on the part of an opponent or aggressor, whereas, the former does not). This may lead directly or eventually, through the process of rumination, to the intention to aggress. Finally, when available coping skills are unavailable or insufficient and an opportunity arises, intention to aggress may be expressed as aggression. Situational cues, such as others acting aggressively or disapproving peers may intensify or moderate any of the stages. This conception is intuitively appealing and owes much to the work of Berkowitz (1965) and Sukhodolsky et al. (2001). However, it is possible that persons who are aggressive tend to ruminate more about anger provoking experiences, that is, rumination may follow rather than precede aggression. If this were the case, one might expect the content of ruminative thoughts to differ depending on the temporal relationship to the aggressive act. Thoughts of revenge might typify rumination prior to aggression, whereas, angry afterthoughts and understanding of causes may well predominate post-aggression ruminations. Despite the general lack of research relating the experience of anger or aggression to performance (see Widmeyer, (1984) for an example of the performance enhancing effects of aggressive behaviour), there is general agreement that the use of unsanctioned aggression in sport is deplorable (Weinberg & Gould, 1999; Widmeyer, 1984). For this reason, suggestions have been proposed to reduce the level of aggression in sport (Tenenbaum, Stewart, Singer, & Duda, 1997). These suggestions typically focus on punishment, education and rule change or stringent enforcement, although they have received criticism for unrealistic expectations and misdirected motivations (Kerr, 1999). The idea that rumination is a cognitive phenomenon typified by negative attributional patterns provides encouragement for teaching more appropriate cognitions in response to provocation. The ruminative modus operandi can be disrupted through the use of several techniques such as thoughtstopping and thought-switching. Thought-stopping involves recognising an inappropriate thought and (silently) yelling ‘STOP’ then breathing deeply and exhaling slowly whilst counting back-
288
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
wards or focusing on a non-aggressive/neutral image. This technique has proved successful in the treatment of various forms of rumination (Martin, 1982; Parenteau & Lamontagne, 1981) and may also be appropriate for combating anger rumination in aggressive athletes. Thought-switching involves replacing negative thoughts with positive ones, such as increasing the desire to win rather than exact revenge. Either method should prove capable of reducing rumination and, as a consequence, help combat the materialization of unwanted aggression in sport.
References Averill, J. R. (1983). Studies on anger and aggression: Implications for theories of emotion. American Psychologist, 38, 1145–1160. Bandura, A. (1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice-Hall. Bandura, A. (1983). Psychological mechanisms of aggression. In R. G. Geen, & E. I. Donnerstein (Eds.), (pp. 1–40). Aggression: Theoretical and empirical reviews, 1. New York: Academic Press. Bandura, A., Ross, D., & Ross, S. A. (1963). Imitation of film-mediated aggressive models. Journal of Abnormal and Social Psychology, 67, 601–607. Baron, R. A., & Richardson, D. R. (1994). Human aggression (2nd ed.). New York: Plenum Press. Berkowitz, L. (1965). The concept of aggressive drive: Some additional considerations. In L. Berkowitz (Ed.), (pp. 301–329). Advances in experimental psychology, 2. New York: Academic Press. Berkowitz, L. (1969). The frustration-aggression hypothesis revisited. In L. Berkowitz (Ed.), Roots of aggression (pp. 1–28). New York: Atherton Press. Berkowitz, L. (1983). The experience of anger as a parallel process in the display of impulsive ‘angry’ aggression. In R. G. Geen, & E. I. Donnerstein (Eds.), (pp. 103–133). Theoretical and empirical reviews, 1. New York: Academic Press. Berkowitz, L. (1988). Frustrations, appraisals and aversively stimulated aggression. Aggressive Behavior, 14, 3–11. Berkowitz, L. (1989). Frustration-aggression hypothesis: Examination and reformulation. Psychological Bulletin, 106, 59–73. Butt, D. S., & Cox, D. N. (1992). Motivational patterns in Davis Cup, University and recreational tennis players. International Journal of Sport Psychology, 23, 1–13. Celozzi, M. J., Kazelskis, R., & Gutsch, K. U. (1981). The relationship between viewing televised violence in ice hockey and subsequent levels of personal aggression. Journal of Sport Behavior, 4, 157–162. Coulomb, G., & Pfister, R. (1998). Aggressive behaviors in soccer as a function of competition level and time: A field study. Journal of Sport Behavior, 21, 222–231. Cox, R. H. (1994). Sport psychology: Concepts and applications. Dubuque, IA: Wm. C. Brown & Benchmark. Dollard, J., Doob, L., Miller, N., Mowrer, O. H., & Sears, R. R. (1939). Frustration and aggression. New Haven, CT: Yale University Press. Eagly, A. H., & Steffen, V. J. (1986). Gender and aggressive behaviour: A meta-analytic review of the social psychological literature. Psychological Bulletin, 100, 309–330. Geen, R. G. (2001). Human aggression (2nd ed.). Buckingham, UK: Open University Press. Gill, D. L. (1986). Psychological dynamics of sport. Champaign, IL: Human Kinetics. Hanegby, R., & Tenenbaum, G. (2001). Blame it on the racket: Norm-breaking behaviours among junior tennis players. Psychology of Sport and Exercise, 2, 117–134. Harrell, W. A. (1980). Aggression by high school basketball players: An observational study of the effects of opponent aggression and frustration-inducing factors. International Journal of Sport Psychology, 11, 290–298. Hull, C. L. (1951). Essentials of behavior. New Haven, CT: Yale University Press. Husman, B. F., & Silva, J. M. (1984). Aggression in sport: Definitional and theoretical considerations. In J. M. Silva, & R. S. Weinberg (Eds.), Psychological foundations of sport (pp. 246–260). Champaign, IL: Human Kinetics. Keltikangas-Jarvinen, L., & Kelnonen, M. (1988). Aggression, self-confidence and cardiovascular reactions in competitive performance in adolescent boys. Aggressive Behavior, 14, 245–254.
J.P. Maxwell / Psychology of Sport and Exercise 5 (2004) 279–289
289
Kerr, J. H. (1999). The role of aggression and violence in sport: A rejoinder to the ISSP position stand. The Sport Psychologist, 13, 83–88. Lefebvre, L. M., & Passer, M. W. (1974). The effects of game location and importance on aggression in team sport. International Journal of Sport Psychology, 5, 102–110. Leith, L. M. (1982). The role of competition in the elicitation of aggression in sport. Journal of Sport Behavior, 5, 168–174. Martin, G. L. (1982). Thought-stopping and stimulus control to decrease persistent disturbing thoughts. Journal of Behavioural Therapy and Experimental Psychiatry, 13, 215–220. McCarthy, J. F., & Kelly, B. R. (1978). Aggressive behavior and its effect on performance over time in ice hockey athletes: An archival study. International Journal of Sport Psychology, 9, 90–96. McGuire, E. J., Courneya, K. S., & Widmeyer, W. N. (1992). Aggression as a potential mediator of the home advantage in professional ice hockey. Journal of Sport and Exercise Psychology, 14, 148–158. Parenteau, P., & Lamontagne, Y. (1981). The thought-stopping technique: A treatment for different types of ruminations? Canadian Journal of Psychiatry, 26, 192–195. Richardson, D. R., Vandenberg, R. J., & Humphries, S. A. (1986). Effect of power to harm on retaliative aggression among males and females. Journal of Research in Personality, 20, 402–419. Russell, G. W. (1974). Machiavellianism, locus of control, aggression, performance and precautionary behavior in ice hockey. Human Relations, 27, 825–837. Silva, J. M. (1980). Understanding aggressive behavior and its effects upon athletic performance. In W. F. Straub (Ed.), Sport psychology: An analysis of athlete behavior (pp. 177–186). Ithaca, NY: Mouvement. Smith, M. D. (1988). Interpersonal sources of violence in hockey: The influence of parents, coaches and teammates. In F. L. Smoll, R. A. Magill, & M. J. Ash (Eds.), Children in sport (3rd ed.) (pp. 301–313). Champaign, IL: Human Kinetics. Spellman, B. A., & Mandel, D. R. (1999). When possibility informs reality: Counterfactual thinking as a cue to causality. Current Directions in Psychological Science, 8, 120–123. Spence, J. T., & Spence, K. W. (1966). The motivational components of manifest anxiety: Drive and drive stimuli. In C. D. Spielberger (Ed.), Anxiety and behavior. New York: Academic Press. Spence, K. W. (1956). Behavior theory and conditioning. New Haven, CT: Yale University Press. Spielberger, C. D. (1988). Manual for the state-trait anger expression inventory (STAXI). Odessa, FL: Psychological Assessment Resources. Sukhodolsky, D. G., Golub, A., & Cromwell, E. N. (2001). Development and validation of the anger rumination scale. Personality and Individual Differences, 31, 689–700. Taylor, S. P. (1967). Aggressive behavior and physiological arousal as a function of provocation and the tendency to inhibit aggression. Journal of Personality, 35, 197–310. Tenenbaum, G., Stewart, E., Singer, R. N., & Duda, J. (1997). Aggression and violence in sport: An ISSP position stand. The Sport Psychologist, 11, 1–7. Varca, P. E. (1980). An analysis of home and away game performance of male college basketball teams. Journal of Sport Psychology, 2, 245–257. Wann, D. L. (1997). Sport psychology. Upper Saddle River, NJ: Prentice Hall, Inc. Weinberg, R. S., & Gould, D. (1999). Foundations of sport and exercise psychology. Champaign, IL: Human Kinetics. Widmeyer, W. N. (1984). Aggression-performance relationships in sport. In J. M. Silva, & R. S. Weinberg (Eds.), Psychological foundations of sport (pp. 274–286). Champaign, IL: Human Kinetics. Widmeyer, W. N., & McGuire, E. J. (1997). Frequency of competition and aggression in professional ice hockey. International Journal of Sport Psychology, 28, 57–66.