Network analysis has recently been introduced as a clinically relevant methodology for understanding the structure of mental disorders and for evaluating cognitive behavioral models of psychopathology. The current study uses network analysis to validate the transdiagnostic model of eating disorders by examining the association between overvaluation of shape and weight and eating disorder symptoms. Eating disorder symptoms were measured among a sample of 194 treatment-seeking children, adolescents, and adults presenting to an outpatient eating disorder clinic. We created transdiagnostic and disorder-specific symptom networks and assessed symptom strength and connectivity. Congruent with the transdiagnostic model, overvaluation of weight and shape emerged among the strongest symptoms in the network, and global network connectivity was higher among individuals with high overvaluation when compared to individuals with low overvaluation. An exploratory analysis revealed that overvaluation of weight and shape was central to anorexia nervosa, bulimia nervosa, and binge eating disorder. Results highlight the associative strength of overvaluation of shape and weight with eating disorder symptoms, regardless of the specific eating disorder diagnosis. Our findings corroborate overvaluation of weight and shape as a transdiagnostic treatment target and potentially useful severity specifier for binge eating disorder.