To explore the relative contributions of self-esteem, body-esteem components and body mass index to disordered eating in aesthetic female athletes and non-athletic females and specifically to determine if the body-esteem components are risk factors for disordered eating.One hundred and fifty two participants, of which 61 rhythmic gymnasts, 42 synchronized swimmers and a non-athletic group of 49 female college students completed the self-esteem scale, the body-esteem scale (satisfaction with general appearance, weight satisfaction and others' evaluations of one's body and appearance), the eating attitudes test, and the body mass index was computed.Results showed a strong heteroscedasticity for EAT-26 and therefore scores were modeled separately for the three groups. For rhythmic gymnasts, the final regression model only emphasizes the role of body-esteem attribution. For the synchronized swimmers, the final regression model combines two body-esteem dimensions (body-esteem for weight and body-esteem attribution) and their interaction. For the non-athletic group, the final regression model only emphasizes body-esteem for weight. Body mass index and low self-esteem were not predictive of disordered eating and no significant relationship was found between body-esteem for appearance and eating attitudes scores. The use of exploratory graphs such as graphs of conditioning and level plots provided more detailed information on the relationship between body-esteem dimensions and eating attitude scores.Results contributed to the growing literature on disordered eating suggesting that attention must be paid to body-esteem for weight and attribution in the understanding of disordered eating and their interaction. Future research should take into consideration the complexity of these results and use a larger sample of aesthetic athletes to elaborate on the current findings.