This study examined the perception of athletes' about their coaches' behavior and skills in terms of knowledge and skills, fairness and coaches' characteristic features. The research was conducted by using relational survey method. The subjects of the study were 95 females and 180 males from diffrent sports. Both team sports athletes such as football, volleyball, basketball, handball and individual sports such as karate, taekwondo, wrestling, kickboxing voluntarily involved ın the study.Perceived Coach Attitudes and Behaviours Scale (PCABS) which was developed by Uzum et al. (2018) was used to asses percevied coaches' knowledge and skills, fairness and characteristic features by athletes. The scale was composed of 24 items and 3 sub-dimensions (Characteristic Features, Skills and Knowledge, Fairness). Uzum et al. (2018) reported the internal consistency for subscales of PCABS ranging from .56 and .88. For the purpose of this study the reliability of two sub-scales of the PCABS was examined using Cronbach's alpha coefficient (α=.83 for knowledge and skills; α=.81for characteristic features of coaches. The data was analyzed by one-way analysis of variance (ANOVA), Tukey's Post-Hoc Analysis and Pearson Correlation analysis. The level of significance for the study was set at p<.05.The results of the study showed that the sub-dimension of characteristic features of coaches had the highest mean avarage. There were statistically significant differences between male athletes and female athletes in both dimensions of characteristic features of coaches and knowledge and skills (p<.05). Further analysis indicated that females scored higher than males in both dimensions. With regard to the level of coaching either professional or amateur, perception of characteristic features of coaches had higer scores in professional level than amateur (p<.05). Moreover, correlational analysis revealed that there was negatively significant correlation between age of the coaches and knowledge and skills dimension (r= -0.13). On the other hand, the variables such as age of athletes, year of sports performance, level of education, type of sports and the coachs' gender, marital status, education level of coaches, amount of time spent with the athlete did not show any significant relationship (p>.05).