Past research evaluating the: continuity and discontinuity models of bulimi
a has produced inconclusive results. In the current study, we performed a t
axometric analysis of bulimia nervosa using means above minus below a slidi
ng cut and maximum covariance analysis with a sample of women diagnosed wit
h bulimia nervosa (n = 201) or women college students (n = 412). Indicators
were derived from the Bulimia Test-Revised and the Eating Attitudes Test-2
6, and both a mixed sample and the nonclinical sample were analyzed. With b
oth taxometric methods and both mixed and nonclinical samples, results were
consistently suggestive of a latent taxon for bulimia. These results chall
enge a dimensional model of bulimia nervosa.