Anger rumination: an antecedent of athlete aggression?

https://doi.org/10.1016/S1469-0292(03)00007-4Get rights and content

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 four-factor 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.

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 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 Frustration-Aggression (Dollard, Doob, Miller, Mowrer, & Sears, 1939), Frustration-Aggression revised (Berkowitz, 1965, Berkowitz, 1969, Berkowitz, 1989, Baron and 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 and 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, Berkowitz, 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 Bandura, 1973, Bandura, 1983 Social Learning Theory. Whilst acknowledging the role of physiological, genetic and motivational factors (Baron & Richardson, 1994), Bandura stressed the importance of learned 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 and Kelnonen, 1988, Lefebvre and Passer, 1974), home team advantage (McGuire, Courneya and Widmeyer, 1992, Varca, 1980), competition level (Butt and Cox, 1992, Coulomb and 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 and 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 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.

Section snippets

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 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

Anger rumination scale: factor analysis

A high Cronbach’s Alpha Coefficient (α=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

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

References (46)

  • L Berkowitz

    The frustration-aggression hypothesis revisited

  • L Berkowitz

    The experience of anger as a parallel process in the display of impulsive ‘angry’ aggression

  • L Berkowitz

    Frustrations, appraisals and aversively stimulated aggression

    Aggressive Behavior

    (1988)
  • L Berkowitz

    Frustration-aggression hypothesis: Examination and reformulation

    Psychological Bulletin

    (1989)
  • D.S Butt et al.

    Motivational patterns in Davis Cup, University and recreational tennis players

    International Journal of Sport Psychology

    (1992)
  • M.J Celozzi et al.

    The relationship between viewing televised violence in ice hockey and subsequent levels of personal aggression

    Journal of Sport Behavior

    (1981)
  • G Coulomb et al.

    Aggressive behaviors in soccer as a function of competition level and time: A field study

    Journal of Sport Behavior

    (1998)
  • R.H Cox

    Sport psychology: Concepts and applications

    (1994)
  • J Dollard et al.

    Frustration and aggression

    (1939)
  • A.H Eagly et al.

    Gender and aggressive behaviour: A meta-analytic review of the social psychological literature

    Psychological Bulletin

    (1986)
  • R.G Geen

    Human aggression

    (2001)
  • D.L Gill

    Psychological dynamics of sport

    (1986)
  • W.A Harrell

    Aggression by high school basketball players: An observational study of the effects of opponent aggression and frustration-inducing factors

    International Journal of Sport Psychology

    (1980)
  • Cited by (105)

    • Child images affect sports fans' prosociality and aggression

      2024, Journal of Behavioral and Experimental Economics
    • The long-term effect of media violence exposure on aggression of youngsters

      2020, Computers in Human Behavior
      Citation Excerpt :

      Rumination, especially anger rumination, is considered to be one of the most important antecedents of aggression. A large number of studies have proven that anger rumination can affect aggression, and it can even increase the duration and intensity of anger, thus potentially increasing aggressive behavior (i.e., Bushman, 2002; Denson, 2012; Denson, Grisham, & Moulds, 2011aMaxwell, 2004). Pedersen et al. (2011) found that anger rumination after provocation is more likely to increase aggressiveness than distracted tasks, even aggression toward those who were not initially related to the provocation.

    View all citing articles on Scopus
    View full text