The application of Chelladurai’s (2001) Multidimensional Model of Leadership
The Multi-dimensional Model of Leadership (MML) (Chelladurai 1978,1990,1993,2001 cited in Riemer, 2007) is a framework which builds on research from non-sporting settings in order to analyse effective leadership in sport. MML demonstrates how success in leadership is influenced by many different factors (Riemer, 2007).
The main objective of the leader/coach is to bring about high levels of performance and satisfaction in the athlete (OU 16-1, 2013). To achieve this, three ‘antecedents’ must be taken into account: situational characteristics (the environment in which the leader is performing); leader characteristics (eg personality, experience, education); and athlete characteristics (e.g. age, gender, skill level, background) (Riemer, 2007). These three antecedents produce three types of leader behaviour: ‘required’, ‘actual’ and ‘preferred’ behaviour (OU 16-1.3, 2013).
MML shows that a leader/coach’s `actual` behaviour is affected both by his awareness of the athlete’s `preferred` behaviour and by the behaviour `required` of him by the coaching context. For example, an elite tennis player may question coaching methodology and seek more input into training. The coach may modify his/her behaviour accordingly. At the same time the coach will be limited by the constraints of the regional governing body regulations when selecting a county team.
The coach’s actual behaviour will also be influenced by his/her character traits and skill/knowledge base (Riemer, 2007). Thus, a mature golf professional, who has recently increased her knowledge base by attending a video analysis seminar, may modify her leadership behaviour to involve the use of smart phone apps in her efforts to improve feedback to the athlete.
Chelladurai (1978, 1990, 1993, 2001 cited in Riemer, 2007) states that when the three types of leader behaviour (`required`, `actual`, and `preferred`) complement and sustain one and other, athlete performance and satisfaction levels will increase.
Athlete preferred behaviour is influenced by a combination of athlete characteristics and situational characteristics (Riemer, 2007). Most studies into preferred coaching behaviour have used the Leadership Scale for Sport (LSS), which was developed by Chelladurai and Saleh (1980, cited in OU 17.1, 2013). The LSS measures five leadership dimensions: training and instruction; social support; positive feedback; autocratic behaviour; and democratic behaviour. Results show that the two most valued leadership styles are positive feedback and training/instruction, with the autocratic approach being least appreciated (OU 17.2, 2013).
Research into the effect of gender on athletes’ leadership preference has produced mixed results (Riemer, 2007). The OU (17.2, 2013) suggests that this may be due to the false equation of gender with biological sex, and that the differences in leadership preferences may be more related to gender role (masculinity/femininity) than to biological sex (male/female) (Riemer, 2007). Findings from Riemer and Toon, 2001, cited in Riemer 2007 suggest that some of the variance in results might be connected to the coach’s gender, rather than the athlete’s, as they showed that female athletes expressed different leadership preference in female coaches compared with male.
Similarly, research has produced no significant relationship between age and preferred leadership styles (Riemer, 2007), although the author does suggest that an overview of the results indicates a shift from a preference for task behaviours to a preference for relationship behaviours as the athlete matures (Riemer, 2007).
Research into how situational characteristics influence leadership preferences has also been mixed (Riemer, 2007). Chelladurai (1978, cited in Riemer, 2007) hypothesised that team/open sport athletes would prefer more training/instruction and positive feedback than those involved in individual/closed sports. Research by Terry and Howe (1984), Terry (1984) and Kang (2003), all cited in Riemer 2007) supported this hypothesis, but Chelladurai and Saleh (1978, cited in Riemer 2007) found the polar opposite. Riemer and Chelladurai (1995, cited in Riemer 2007) found that, within American football, athletes whose tasks differed from their teammates (e.g. offensive or defensive) preferred different coaching behaviours. The defensive squad operating in a more open environment preferred more democratic, autocratic and socially supportive coaching behaviours.
Given the varied results of studies into athlete and situational characteristics on leadership preferences, it is unclear how coaches should adapt their coaching strategies based on member characteristics.
The role of transformational leadership in coaching
Chelladurai (1978, 1990, 1993, 2001 cited in Riemer, 2007) adapted his model of the MML to include transformational leadership (Open University, 17, 2013). In the previous versions of the model, leadership was thought to be transactional in its nature (Riemer, 2007). Transactional leaders have a reciprocal relationship with their athletes whereby they trade something the athlete wants (e.g. knowledge), with something they want (e.g. respect) (Riemer, 2007 cited in Open University, 16.5, 2013). Transactional leadership is built on mutual trust in which the coach must manage the changing needs of the athlete in order to be effective (Riemer, 2007).
Transformational leadership on the other hand, is a charismatic type of leadership in which followers associate themselves with leaders, giving rise to greatly increased levels of performance (Yammarino, Dubinsky, Comer and Jolson, 1997 cited in Riemer, 2007). A transformational leader will exhibit five traits: charisma, an assertive and self confident quality that results in athletes connecting emotionally to their leaders, generating high levels of trust; idealised influence, which enables leaders to motivate athletes to follow their beliefs and visions for the future; inspiration, which enables leaders to instil confidence in athletes so that they are able to meet the challenging aspirations s/he has set for them; intellectual stimulation, through which leaders provoke original and imaginative thought processes; and individual consideration, by which leaders pay attention to each member of their team and are able to set specific goals for each dependant on their skill level (Bass, 1985; Bass and Aviolo, 1990; Doherty and Danylchuk, 1996; Weese, 1994 cited in Open University, 17.1, 2013).
Transformational leaders affect both the situational and athlete characteristics of the MML (Riemer, 2007). Firstly, situational characteristics are affected by the coach introducing new ambitions for the organisation / team they are involved with. These ambitions are based on the coach’s beliefs and visions for the future. Secondly, athlete characteristics are changed by the coach transmitting his/her vision so that athletes embrace this as their own, and by motivating athletes to believe they will achieve this vision (Open University, 17.1, 2013).
Bass (1985, cited in Riemer, 2007) suggests that not all coaches are transformational, rather that leaders will display a spectrum of transformational behaviours. Although there is minimal literature into the role of transformational leadership in sport (Riemer, 2007), the MML suggests that a transformational leader may be a more effective coach (Chelladurai, 2001 cited in Riemer, 2007). This is backed up by findings in Arthur et al’s (2011) study into athlete narcissism, coach’s transformational behaviours and athlete motivation. The study found that the dimensions of individual consideration and intellectual stimulation were linked with higher levels of athlete effort.
Creating an effective coaching environment
In order to further understand the complicated, intertwined relationship between coach, athlete and the conditions they work under, Smith and Smoll (1977, 1989, cited in OU 18.1, 2013) proposed the meditational model of sport leadership. This model was formulated to improve coach understanding around how to create a positive environment for children in sport (Smith and Smoll, 2007) and focuses on three variables: coach behaviour (what the coach actually does); athlete perceptions (how behaviour is understood by athletes); and athlete reactions (subsequent responses to the coach) (Smith and Smoll, 2007, cited in OU, 18.1, 2013).
With a view to measuring the first of these variables, Smoll, Smith and Hunt (1977, cited in OU, 18.1, 2013) designed the coaching behaviour assessment system (CBAS). Leadership behaviour was observed and coded in coaches working with children across a collection of different sports (Smith and Smoll, 2007). Although widely used to research leadership behaviours (OU, 18.1, 2013) the statistical methodology used in the CBAS has been considered too narrow an approach to develop an understanding of the complex nature of effective coaching practice (Cushion, 2007 cited in OU, 18.1, 2013). To gain deeper insight into coaching behaviours, the origins of these behaviours need to be examined, along with how these behaviours affect perceptions and finally how these perceptions shape actions (Chelladurai, 1993; Cumming, Smith and Smoll, 2006 cited in OU, 18.1, 2013).
Research by Stebbings et al (2011) explored the antecedents of autonomy supportive and controlling behaviours in coaches in regards to the coaches’ own psychological needs. Autonomy supportive environments are created by a coach offering athletes’ choice in making decisions, the opportunity for experimentation, meaningful justification for activities and a recognition of an athlete’s emotional needs (Mageau and Vallerand, 2003, cited in Stebbings et al, 2013). A controlling environment is created by a coach with an autocratic, dictatorial demeanour, giving athletes no choice in the decision making process. Punishment, criticism and tangible rewards are used to control athletes (Bartholemew, Ntoumanis and Thogersen-Ntoumani, 2009 cited in Stebbings et al, 2011). Autonomy supportive environments are thought to elicit positive responses from athletes such as increased intrinsic motivation and higher levels of performance (Amorose, 2007; Gillett, Vallerand Amoura and Baldes, 2010; Mageau and Vallerand, 2003, cited in Stebbings et al, 2011). Conversely, a controlling environment is associated with decreased motivation and increased levels of sport drop out (Pelletier et al, 2009 cited in Stebbings et al, 2011). This research hypothesised that coaches whose psychological needs were met would work in an autonomy supportive manner, whilst those whose needs were not met would exhibit more controlling behaviour (Stebbings et al, 2011). The results proved to be in line with the initial hypothesis with the pivotal conclusion being that the ‘satisfaction of coaches psychological needs can allow coaches to thrive, and to create an adaptive interpersonal coach environment for athletes.’ (Stebbings et al, 2011, p269). Lack of governing body support in terms of training, feedback and guidance were cited as possible reasons for coaches’ psychological needs not being met (Allen, Shaw, 2009 cited in Stebbings et al, 2011). However it is also noted that further research into the effect of environmental factors on coach psychological need satisfaction would aid understanding (Stebbings et al, 2011).
As coach behaviours are thought to influence athlete perceptions, and these perceptions will subsequently shape actions (Smith and Smoll, 2007), coaches must be careful not to make immediate judgements towards their athletes based on first impressions (OU, 18.5, 2013). This is to minimise the risk of creating a self-fulfilling prophecy (OU, 18.5, 2013) whereby a coach will make initial predictions of the athlete’s behaviour and performance. These predictions will affect the way the coach communicates with the athlete, which in turn will affect the athlete’s perception of themselves and their performance levels (Horn, 2008 cited in OU, 2013).
Coaching behaviours and team performance
In contrast to those working with individuals, coaches working with teams have to be much more concerned with team cohesion. Team cohesion is associated with the amount of unity, closeness and camaraderie displayed by a team (Carron, Eys and Burke, 2007) and is defined as a dynamic process, reflected in the tendency of a group to remain united in pursuit of its goals and objectives (Carron, Brawley and Widmeyer, 1998 cited in Carron, Eys and Burke, 2007). Cohesion has been described as the most important group property (Bollen and Hoyle, 1990; Golembiewski, 1962; Lott and Lott, 1965, cited in Carron, Eys and Burke, 2007) and involves two dimensions: task cohesion – the ability of a team to work towards shared goals; and social cohesion – the degree to which team mates get along with each other (Carron, Ely and Burke, 2007). These dimensions are not mutually exclusive, and change in one dimension will effect change in the other. When formulating a coach agenda a coach should aim to achieve a balance of both appropriate for a given context (OU, 19.2, 2013).
The formulation of a cooperative training environment has also been cited as benefitting team performance (Collins and Collins, 2011 cited in OU, 19.2, 2013). Coaches need athletes to work in collaboration with each other (e.g. on tactics, formations and patterns of play) and against each other (e.g. attack versus defence) in order to drive overall and individual performance levels up (Collins and Collins, 2011, cited in OU, 19.2, 2013). As part of this cooperative training environment, coaches and athletes need to be able to understand each other’s strengths and weaknesses, both as individuals and in their team responsibilities (Collins and Collins, 2011, cited in OU, 19.2, 2013). Effective teams are considered to have a high degree of shared knowledge built up through communal experience of intense training and performance situations (Bourbouson et al, 2011 cited in OU, 19.2, 2013). For example, a team may change seamlessly between a number of formations and tactical variations during a match at their coach’s request. Shared knowledge can also extend to off field activities. When coaches are creating and developing their coaching agenda they should examine their players in many different domains. Information should be garnered from multiple sources, such as discussions with colleagues, personal observation, formal assessments and past experiences, in order to surmount current obstacles and to plan for the future (OU, 19.2, 2013). Systems of work and team ethics need to be established so that existing and new team mates can carve an individual niche for themselves but also to see what is expected of them as part of the team (Martens, 2012 cited in OU, 19.2, 2013).
When considering how a coach develops his/her team, research has suggested that team cohesion and performance are mutually beneficial, with increased task and social cohesion leading to increase performance, and vice versa (Carron et al, 2002; Cox, 2012 cited in OU, 19.3, 2013). Therefore in order to raise performance levels, coaches must attempt to create cohesion, achieved by developing motivation and a sense of personal reward. They can identify leaders to help them facilitate change but also must preserve and unite all existing members of the team (including the assistants, parents and others in the support network). Individual and group accomplishments need to be recognised, changes need to be discussed and performance regularly assessed through feedback in order to facilitate a supportive culture (Martens, 2012 cited in OU, 19.3, 2013).
Allowing athletes more influence in team goal setting can also lead to increased task and social cohesion (Carron et al, 2005 cited in OU, 19.3, 2013). This increased influence also leads to greater understanding of the tasks they need to undertake, and stronger beliefs that the goals will influence team behaviour positively (Collins and Collins cited in OU, 19.3, 2013). The coach should develop individual and communal understanding of roles, and encourage team members to have confidence in each other. Through regular meetings and discussions with various team groupings, teams and coaches can build agreement, understanding and commitment (Collins and Collins, 2011, cited in OU, 19.3, 2013). This is as necessary within coaching teams as between the coach and athletes. Experimentation, questioning and argument are far more acceptable in a sharing and cooperative environment which reinforces team cohesion (Piggott, 2012, cited in OU, 19.3, 2013).