Multidimensional Model of Sport Leadership

An  established  model  of  leadership  in  sports  is Packianathan   Chelladurai’s   multidimensional model  of  leadership  (MML).  This  model  was  the substance of a doctoral dissertation in management science.  It  represented  a  synthesis  and  reconciliation of the models of leadership found in the mainstream  management  literature.  These  preexisting models tended to focus more on either the leader, or the member, or the situation. However, as leadership  is  a  concept  that  encompasses  all  three factors—the leader; the members; and the organizational  context  including  goals,  structures,  and processes—it was reasonable to propose the model illustrated in Figure 1.

A unique feature of the model is that it includes three states of leader behaviors. Required behavior (Box  4)  is  the  set  of  prescriptions  and  proscriptions  of  the  situation  in  which  leadership  occurs. Required  behavior  is  mostly  defined  by  the  situational  characteristics  (Box  1)  that  include  the goals of the group, the type of task (e.g., individual vs. team, closed vs. open tasks), and the social and cultural  context  of  the  group.  The  nature  of  the group defined by gender, age, skill level, and such other  factors  would  also  partly  define  required behavior. Preferred behavior (Box 6) refers to the preferences  of  the  followers  for  specific  forms  of behavior  (such  as  training,  social  support,  and feedback)  from  the  leader.  Members’  preferences are a function of their individual difference characteristics (Box 3) such as personality (e.g., need for affiliation, tolerance for ambiguity, attitude toward authority)  and  their  ability  relative  to  the  task  at hand.  Members  are  also  aware  of  the  situational requirements; thus, their preferences are influenced by  those  requirements.  The  actual  behavior  (Box 5, i.e., how the leader actually behaves) is largely based  on  leader  characteristics  (Box  2)  in  terms of personality, expertise, and experience. However, the  leader  would  also  be  constrained  to  abide  by the  requirements  of  the  situation  (Box  4)  and  to accommodate member preferences (Box 6) as well.

Another  significant  feature  of  the  MML  is  its congruence  hypothesis.  That  is,  the  model  specifies  that  the  desired  outcomes  of  individual  and team  performance,  and  member  satisfaction  will be  realized  if  the  three  states  of  leader  behaviorare congruent with each other. Any misalignmentamong  the  three  states  of  leader  behavior  would diminish performance and/or satisfaction. Further, if  there  is  continued  discrepancy  between  actual leader behavior and the other two states of leader behavior,  the  leader’s  position  within  the  group would become untenable. The dynamic nature of leadership  is  highlighted  in  the  model  with  backward  arrows  indicating  feedback  from  attained performance  and/or  member  satisfaction.  That is,  the  leader  may  begin  to  exhibit  more  of  task-oriented  behaviors  if  she  or  he  feels  that  the  performance  was  below  expectations.  On  the  other hand,  leader  behaviors  may  begin  to  be  more interpersonally  oriented  if  it  is  felt  that  members were low on morale and/or satisfaction.

leadership-in-sport-multidimensional-model-sports-psychologyFigure 1    Multidimensional Model of Leadership

In 2007, Chelladurai made one significant modification to the model. It was the incorporation of the concept of transformational leadership into the multidimensional  model.  In  his  view,  leadership exhibited by coaches is largely concerned with pursuit of excellence. In the process of pursuing excellence, the person is transformed from a relatively unaccomplished  novice  into  an  expert  performer. Thus  successful  coaches  do  exhibit  transformational  leadership  and,  as  such,  incorporating  the concept  into  the  model  was  necessary  as  well  as easy. In essence, the coach as the leader transforms member characteristics in terms of aspirations and attitudes and changes the situational requirements by articulating a new mission and convincing the members of the viability of the mission and their capacity to achieve that mission.

The  idea  (as  shown  in  Figure  1)  that  transformational  leadership  influences  leader  characteristics  (Box  2)  would  be  most  relevant  where  a coach has one or more assistant coaches. That is, the  chief  coach  would  attempt  to  transform  the assistant coaches in the same way he or she would transform  player  characteristics.  If  there  is  only one  coach,  it  would  mean  that  the  coach  would change  his  or  her  own  characteristics  to  fit  the transformational mold and attempt to change the situational characteristics as well as the characteristics  of  the  members  as  indicated  by  the  dotted arrows  flowing  from  actual  behavior  (Box  5)  to situational characteristics (Box 1) and to member characteristics (Box 3).

A  theory  is  useful  only  to  the  extent  that  the variables of the study can be measured and the relationships among the variables can be verified. With this in mind, Chelladurai developed the Leadership Scale for Sports (LSS) to measure the three forms of  leader  behavior  contained  in  the  model.  It  is composed of 40 items to measure these five dimensions  of  leader  behavior:  training  and  instruction (13  items),  democratic  behavior  (9  items),  autocratic behavior (5 items), social support (8 items), and  positive  feedback  and  rewarding  behavior (5  items).  The  response  format  is  a  5-point  scale ranging from (1) always; (2) often, about 75% of the time; (3) occasionally, about 50% of the time; (4)  seldom,  about  25%  of  the  time;  to  (5)  never. The scale has been used to measure athletes’ preferences, their perceptions of their coaches’ behavior, and  coaches’  perceptions  of  their  own  behavior pertaining to those five dimensions of behavior.

The  psychometric  properties  of  the  LSS  have been  verified  and  supported  in  several  studies. However,  the  subscale  of  autocratic  behavior  has been shown to be weak in almost all studies. One reason for such low internal consistency estimates is  that  the  items  in  the  subscale  relate  to  three different  forms  of  behavior—being  aloof,  being authoritative,  and  making  autocratic  decisions. Another conceptual issue that plagues the dimension of autocratic behavior is that whether a coach should be autocratic or democratic is dependent on the attributes of the problem in question. The items in this subscale do not capture the situational contingencies. The entry on “Decision-Making Styles in Coaching” (this volume) deals with contextual differences that indicate the degree of participation by team members in decision making.

There has been an attempt to improve the LSS by James J. Zhang and his colleagues. Their Revised LSS includes the five dimensions, the instructions, and the response format of the original LSS. It also includes a new dimension titled situational consideration  behaviors.  However,  the  Revised  LSS  has not  been  subjected  to  confirmatory  analyses  and the new dimension is subsumed by the original five dimensions.  Therefore,  parsimony  would  dictate the use of the original LSS.


Both  the  MML  and  the  LSS  have  been  used  and tested in several studies. Much of the research on the  notion  of  congruence  suggested  in  the  model has  been  restricted  to  only  two  states  of  leader behavior—preferred   behavior   and   perceived behavior used as a proxy for actual behavior. While the notion of congruence between these two states of leader behavior has been largely supported, it is disappointing that the congruence among all three states of leader behavior has not been tested adequately.  Researchers  could  have  omitted  required behavior  from  consideration  because  of  the  difficulty of measuring required behavior. In his original research, Chelladurai employed the average of coaches’ reporting of their own behavior as a surrogate  of  required  behavior.  Future  research  may consider  asking  expert  coaches  specifically  about what  should  be  the  required  behavior  in  a  given situation defined by age, gender, ability level, goals of  the  program,  and  so  on.  The  average  of  these responses may be used as required behavior in the relevant situation.

In  developing  the  LSS,  Chelladurai  resorted to   the   leadership   scales   then   popular   in   the mainstream  management  literature  such  as  the Leadership  Behavior  Description  Questionnaire (LBDQ).  He  collated  more  than  100  items  from these scales and reworded them to suite the coaching context. However, he was forced to reduce the scale to 99 items because the computers available at  his  university  at  that  time  could  not  handle more  than  that  number  of  items.  He  administered  the  initial  questionnaire  to  members  of  the university basketball teams in Canada and derived five dimensions of leader behavior through exploratory factor analysis. Subsequent research studies employing  the  more  sophisticated  confirmatory factor analysis has supported the robustness of the LSS. While these steps are acceptable, it is necessary  to  generate  sport  specific  items,  group  them into  meaningful  categories,  administer  the  questionnaire to different teams in different sports, and subject  the  data  to  confirmatory  factor  analyses. Furthermore, the LSS does not tap into the dimensions  of  transformational  leadership,  which  has been  recently  incorporated  into  the  MML.  It  is expected  that  future  research  will  focus  on  refining the existing subscales and developing new subscales for transformational leadership.

Finally,  it  must  be  noted  that  although  the MML  was  advanced  in  research  related  to  leadership  in  athletics,  the  model  itself  is  applicable to  any  context  (e.g.,  business,  industry,  military) where leadership is a critical process. The model, after all, is a synthesis of other models from business and industry. Thus, reversing the process and applying  the  model  to  other  contexts  including business  and  industry  is  feasible.  While  the  situational and member characteristics may vary from context to context, the concepts of required, preferred, and actual behavior are meaningful in any context. On a different note, Chelladurai included group  performance  and  member  satisfaction  as the  outcome  variables.  But  other  outcomes  such as  individual  performance,  individual  growth, group  cohesion,  group  solidarity,  commitment, identification,  and  organizational  citizenship  can easily  be  accommodated  in  the  model.  Further, business  outcomes  such  as  profitability,  market share,  and  return  on  investments  can  be  used  as outcome variables.


  1. Chelladurai, P. (1978). A contingency model of leadership in athletics. Unpublished doctoral dissertation, Department of Management Sciences, University of Waterloo, Waterloo, ON, Canada.
  2. Chelladurai, P. (1993). Leadership. In R. N. Singer,M. Murphy, & K. Tennant (Eds.), The handbook on research in sport psychology (pp. 647–671).New York: Macmillan.
  3. Chelladurai, P. (2007). Leadership in sports. In G.Tenenbaum & R. C. Eklund (Eds.). Handbook of sport psychology (3rd ed., pp. 113–135). Hoboken, NJ: Wiley.
  4. Chelladurai, P., & Carron, A. V. (1978). Ottawa: CAHPER, Sociology of Sport MonographSeries.
  5. Chelladurai, P., & Riemer, H. (1998). Measurement of leadership in sports. In J. L. Duda (Ed.), Advancements in sport and exercise psychology measurement(pp. 227–253). Morgantown, WV: Fitness InformationTechnology.
  6. Chelladurai, P., & Saleh, S. D. (1980). Dimensions of leader behavior in sports: Development of a leadership scale. Journal of Sport Psychology, 2(1), 34–45.
  7. Zhang, J., Jensen, B. E., & Mann, B. L. (1997).Modification and revision of the Leadership Scale forSport. Journal of Sport Behavior, 20(1), 105–121.

See also: