Three studies examined psychological, physiological, and behavioral characteristics related to disordered-eating patterns. In the first study, psychological characteristics were investigated. Subjects were 265 students and 173 eating-disordered patients diagnosed as anorexic, bulimic or bulimarexic. The Millon Clinical Multiaxial Inventory (MCMI) was administered, and item differences between the groups investigated by constructing three dummy variables contrasting the groups and correlating the dummy variables with item responses. None of the items significantly differentiated between the bulimics and bulimarexics. Only two items significantly (p <.005) differentiated the anorexics from the bulimics, and anorexics from the bulimarexics. Of the items, 64% significantly (p <.005) differentiated the eating-disordered patients as a group, from the students. In order to determine which items were most discriminating the range of Z scores between the groups were examined and those which differed by at least one standard deviation identified. The content of the items revealed that the eating-disordered females were socially withdrawn and depressed. In the second study, psychological and behavioral differences between weight-classified eating-disordered patients were investigated. The Eating Disorders Instrument (EDI) was administered. Subjects were 196 patients, classified as either overweight or normal weight bulimic, overweight or normal weight bulimarexic, or underweight anorexic. Differences between the groups were investigated by constructing ten dummy variables contrasting the groups, and correlating the scale scores with the dummy variables. Three of the scales significantly (p <.005) differentiated between the contrasted groups: Desire for Thinness, Body Dissatisfaction, and Bulimia. In the third study, physiological, psychological and behavioral characteristics were investigated. Subjects were 99 eating-disordered females diagnosed as anorexic, bulimic, or bulimarexic; 27 psychiatric controls; and 24 medical controls. The Disordered Eating and Weight Control Instrument (DEWCI) was administered. Differences between the groups were investigated by constructing ten dummy variables contrasting the groups, and correlating the scales with the dummy variables. All of the DEWCI scales significantly (p <.005) differentiated between the eating-disordered, psychiatric, and medical patients.