The relationship between exercise addiction, eating attitudes, anxiety, and self-esteem was examined in a gym population and a comparison group. A questionnaire based social survey, consisting of the Exercise Dependence Scale-21, Eating Attitudes Test-26, Beck Anxiety Inventory, and Rosenberg Self-Esteem scale was distributed to (n=85) respondents (40 men and 45 women). 57 respondents were actively recruited from three separate gyms, and 28 respondents from a household drop-off administration, in Co. Tipperary. The dependent variables were the respondents' sex, type of exercise, and number of hours spent exercising a week. The independent variables were the responses to the four scales. A one-way ANOV A was used to test the first hypothesis, that there would be evidence of primary exercise dependence, by looking at the seven EDI subscales in relation to EAT scores. The results indicated that five subscales had a significant main effect on eating attitudes at the dependence level as follows: Withdrawal (f(2,82)=5.107, p<0.05), Continuance (f(2,82)=9.972, p<0.05), Tolerance (f(2,82)=7.143, p<0.05), Lack of Control (f(2,82)=8.913, p<0.05), and Time(f(2,82)=5.383, p<0.05). These findings therefore do not support the first hypothesis but instead indicate that there is evidence for secondary exercise dependence. A Univariate Analysis of Variance was used to test the second hypothesis that there would be a difference between male and female scores on the EAT and EDS scales. The results partially supported the hypothesis by indicating that the interaction between gender and EAT scores had a higher significant main effect for females (F(l, 78)=6.440, p<0.05) compared to males. However, when the above test was used to look at the interaction between both gender and exercise dependence level in relation to EAT scores the results were not statistically significant. Pearsons r Correlations were used to investigate the third and fourth hypotheses. The findings supported the third hypothesis that low self-esteem would correlate with higher EAT scores. (r=0.211, p<0.05, 1 tailed) The findings also supported the fourth hypothesis that there would be a correlation between BAI and EAT scores. (r=0.371, p<0.01, 2 tailed) These results highlight the importance of awareness and support surrounding gym members who may be struggling with exercise dependence and a related eating disorder. The findings also indicate that a lower self-esteem and anxiety may be related to symptoms and characteristics of an eating disorder amongst this population. The findings highlight the importance of conducting further exercise dependence research, both in the gym population and outside of it, which is directed towards eating attitudes, gender, self-esteem and anxiety in particular.