Sport participation is continually rising nationally and internationally emphasizing the need for effective coaches. Literature regarding coaching in sport is also increasing yearly leading to an ample amount of studies available. Despite the available results, there is still a lot more to understand concerning the complex nature of coaching. Although the majority of the studies focused on coaching behaviours, only limited results exist measuring the perceptions of coaches and players to determine whether the behaviours are effective. More scientific research in the area of coaching effectiveness is therefore needed, especially in South Africa. The objectives of this study were firstly to determine the difference between players‟ perceptions of coaching effectiveness between larger and smaller secondary schools and secondly, to determine the difference between the players and the coaches‟ perceptions of coaching effectiveness at university/club level. Twenty high schools (n = 20) and four hundred and seventy-six (n = 476) male rugby union players (15-19 years) participated in the first part of the study. Players were asked to fill in the adapted version of the Coaching Efficacy Scale (CES). To determine the validity of the sample size a power calculation was done followed by Chronbach alpha values to determine the reliability of the questionnaire on the specific group. Discriptive statistics, followed by independant t-test were done to measure statistically differences between large and small schools with regard to the 4 constructs measured. Cohen‟s effect sizes were then used to determine practical significantly differences. The sample size was found to be valid (> 0.9) together with high Cronbach alpha values (> 0.7) indicating that the questionnaire is reliable for the specific sample. It was found that no practically significant differences exist between coaches from large schools compared to coaches from smaller schools. It was therefore concluded that the players from high schools that participated in the study perceive their coaches‟ effectiveness with regard to the constructs measured similarly, irrespective of the size (number of learners) of the school. One hundred and forty-two (n =142) male rugby union players (age: 18-15) and thirteen (n = 13) coaches (age: 23-55) from the Puk Rugby Institute (PRI) participated in the second part of this study. All the players completed the adapted version of the CES and the coaches completed the CES. Confirmatory factor analysis (CFA) was done to determine the factor structure of the adapted version of the CES. The CFA of the CES was not done due to the small number (13) of coaches that participated in the study. Regression estimates were set at p<0.05. Cronbach alpha values were measured to determine the reliability of the questionnaires. To adapt for inter-dependency, mixed models (set at 5%) were used to measure perception differences between coaches and players. All the regression estimates were found to be significant (p < 0.05). High Cronbach alpha values were found (>0.70). Statistically significant differences between coaches and players‟ perceptions were found with regard to the four constructs measured where coaches‟ perception values were higher than those of the players. Although it is not important to practice, the findings should not be ignored. For total coaching effectiveness values, the majority of the coaches (8 out of 13) rated their own coaching effectiveness more positive than did their respective players. It was therefore concluded that researchers should be aware of the perception differences that exist when measuring coaching effectiveness. These results complement the existing literature with regard to sport coaching and the complexity thereof. / MA (Sport Science), North-West University, Potchefstroom Campus, 2014

Sport participation is continually rising nationally and internationally emphasizing the need for effective coaches. Literature regarding coaching in sport is also increasing yearly leading to an ample amount of studies available. Despite the available results, there is still a lot more to understand concerning the complex nature of coaching. Although the majority of the studies focused on coaching behaviours, only limited results exist measuring the perceptions of coaches and players to determine whether the behaviours are effective. More scientific research in the area of coaching effectiveness is therefore needed, especially in South Africa. The objectives of this study were firstly to determine the difference between players‟ perceptions of coaching effectiveness between larger and smaller secondary schools and secondly, to determine the difference between the players and the coaches‟ perceptions of coaching effectiveness at university/club level. Twenty high schools (n = 20) and four hundred and seventy-six (n = 476) male rugby union players (15-19 years) participated in the first part of the study. Players were asked to fill in the adapted version of the Coaching Efficacy Scale (CES). To determine the validity of the sample size a power calculation was done followed by Chronbach alpha values to determine the reliability of the questionnaire on the specific group. Discriptive statistics, followed by independant t-test were done to measure statistically differences between large and small schools with regard to the 4 constructs measured. Cohen‟s effect sizes were then used to determine practical significantly differences. The sample size was found to be valid (> 0.9) together with high Cronbach alpha values (> 0.7) indicating that the questionnaire is reliable for the specific sample. It was found that no practically significant differences exist between coaches from large schools compared to coaches from smaller schools. It was therefore concluded that the players from high schools that participated in the study perceive their coaches‟ effectiveness with regard to the constructs measured similarly, irrespective of the size (number of learners) of the school. One hundred and forty-two (n =142) male rugby union players (age: 18-15) and thirteen (n = 13) coaches (age: 23-55) from the Puk Rugby Institute (PRI) participated in the second part of this study. All the players completed the adapted version of the CES and the coaches completed the CES. Confirmatory factor analysis (CFA) was done to determine the factor structure of the adapted version of the CES. The CFA of the CES was not done due to the small number (13) of coaches that participated in the study. Regression estimates were set at p<0.05. Cronbach alpha values were measured to determine the reliability of the questionnaires. To adapt for inter-dependency, mixed models (set at 5%) were used to measure perception differences between coaches and players. All the regression estimates were found to be significant (p < 0.05). High Cronbach alpha values were found (>0.70). Statistically significant differences between coaches and players‟ perceptions were found with regard to the four constructs measured where coaches‟ perception values were higher than those of the players. Although it is not important to practice, the findings should not be ignored. For total coaching effectiveness values, the majority of the coaches (8 out of 13) rated their own coaching effectiveness more positive than did their respective players. It was therefore concluded that researchers should be aware of the perception differences that exist when measuring coaching effectiveness. These results complement the existing literature with regard to sport coaching and the complexity thereof. / MA (Sport Science), North-West University, Potchefstroom Campus, 2014