Introduction: The purpose of the study was to investigate in a sample of Scandinavian elite coaches, the relationship between perceptionsof their own leadership behavior and their perception of the coach-athlete relationship.Method149 elite coaches (e.g. national top and/or international level) from Denmark (n=50), Norway (n=50) and Sweden (n=49) participated (134male, mean age 38.3±9.8yrs; 15 female, mean age 39.1±8.4). 58 percent of the coaches were coaches of individual sports and 42percent were coaches of team sports. 59 percent of the participants had at some point, undertaken academic study related to sport atuniversity level. A paper-copy questionnaires were distributed by post to all coaches listed in each country’s athletic federation index.Leadership behaviour: Coaches self reported their perceptions of their own leadership behaviors using the Leadership Scale for Sport(Chelladurai & Saleh, 1980). The Cronbach’s alpha for self reported leadership behavior was: .79 training and instruction, .59 positivefeedback, .66 social support, .78 democratic behavior, and .46 autocratic.Coach-athlete relationship: The nature of the coach-athlete relationship was evaluated using the 13-item Nordic Coach–Athlete RelationshipQuestionnaire (NOR-CART-Q; Jowett & Ntoumanis, 2003). The Cronbach’s alpha for closeness, commitment, and complementaritywas .83, .72 and .67 respectively.Results: Positive feedback, training and instruction, and democratic behavior were the most frequent self-reported behavior subscalesamongst the coaches. Moreover, it is a positive relationship between commitment and training and instruction (r=.25, p < 0.01) positivefeedback (r= 22, p < 0.01) and social support behavior(r=.22 p < 0.01). Complementarity was positively related to training and instructionbehavior (r=.17, p < 0.01). Multilevel logistic regression analyses indicated a significant difference between coaches in Denmark andSweden on commitment (5.39 vs. 5.95) and complementarity (5.82 vs. 6.26) and coaches who are educated in sport use more positivefeedback (4.25 vs. 4.41). Moreover, coaches with more than 10 years experiences in coaching us significantly more training and instruction(3.65 vs. 3.82) and social support (3.02 vs. 3.23) than coaches with less experiences. Furthermore, coaches in individual sport reportedmore democratic behavior (3.80 vs. 3.23) and less autocratic behavior (2.56 vs. 2.78) then coaches in team sport.

Discussion: In general coaches with more experience and more sport education used more positive feedback and social support. Thestudy also revealed differences in leadership behavior between individual and team sport which is in line with former research.