Athletics, perfectionism, and disordered eating.

OBJECTIVES: To determine the effects of sex, perfectionism, level of athletic participation (varsity vs. recreational), and type of sport (swimming, running, or soccer) on disordered eating.
METHOD: Male and female swimmers, runners, and soccer players at Stanford University were recruited at both the recreational and varsity level. Athletes were given the Multidimensional Perfectionism Scale (MPS), the Eating Attitudes Test-26 (EAT-26), the Eating Disorders Examination Questionnaire (EDE-Q), and a survey on demographics and level of athletic competition. Of 257 athletes identified, 250 (97.3%) returned the questionnaires.
RESULTS: A Multivariate Analysis of Variance (MANOVA) was used to assess the variables' effects on disordered eating attitudes. Significant main effects were found for Sex (F5,199=25.7, p<0.001), Level (F5,199=2.5, p<0.05), and Perfectionism (F10,400=3.0, p<0.001). Only two of the ten interaction terms were significant: Sex by Level of athletic participation (F5,199=3.2, p<0.01) and Sex by Perfectionism (F10,400=2.1, p<0.05). Females displayed higher disordered eating than males. Disordered eating attitudes increased with perfectionism and were greater for recreational athletes than varsity, but this trend was significant for females only. Type of sport showed no significant effects. When examining physiological data a higher percentage of varsity females had irregular periods (42.9%) or amenorrhea (14.3%) than recreational females (13.4% and 2.9%, respectively). DISCUSSION: Males showed little variability in their data due to low disordered eating scores overall, thus demonstrated few significant results. The greatest risk factor for disordered eating attitudes for females was perfectionism, which crossed all athletic divisions. Although recreational athletes seemed more at risk psychologically, the varsity athletes had more physiological risk.