Date of Award


Document Type


Degree Name

Doctor of Philosophy


College of Education and International Services


Counseling Psychology, Ph.D.

First Advisor

Jimmy Kijai

Second Advisor

Nancy J. Carbonell

Third Advisor

Frederick A. Kosinski, Jr.



Eating disorders, in the form of anorexia nervosa and bulimia nervosa, have been recognized as significant mental-health issues for the last three decades, and the incidence is rising as we approach the new millennium. Currently, many women who come into a mental-health setting due to depression, anxiety, low self-esteem, relationship issues, sexual issues, etc., are also struggling with eating-disordered behaviors, thoughts, and feelings. These behaviors, thoughts, and feelings may remain well hidden from the counselor throughout the course of therapy or until they become severe and more difficult to treat. Because eating disorders are very complex involving psychological, physical, and mental functioning, and because the symptoms become progressively more severe, early detection and intervention are essential for optimal outcome. Numerous assessment instruments exist but are not employed until obvious signs of eating disorders are exhibited. This study was designed to develop a subscale from the MMPI-2 items which will screen for eating disorders. Since the MMPI-2 is widely used early in the process of psychological evaluation, it was deemed the desirable instrument to use.


The methodology for this study involved scale development and included four phases. Phase 1 was the initial study in which 354 MMPI answer sheets from eating-disordered individuals were compared to 238 MMPI answer sheets from non-eating-disordered individuals in order to determine MMPI items which differentiate between the two groups. Phase 2 utilized expert judges to evaluate the pertinence of each item on an eating-disorder questionnaire and to assign directionality. Phase 3 involved administering the items remaining after the first two phases to a new research sample comprised of eating-disordered and non-eating-disordered subjects. Phase 4 entailed eliminating the items which did not meet the total correlation criterion, and computing internal consistency using Cronbach’s alpha coefficient for the remaining items.


This research resulted in the development of a 68-item proposed MMPI-2 subscale to screen for anorexia nervosa and bulimia nervosa. Each item met the differential criterion at the .01 level and the correlation coefficient at .33. The proposed MMPI-2 subscale has a reliability of .971 and its composition is unifactorial.


This research establishes the efficacy of utilizing the MMPI-2 to screen for eating disorders. Additional administration of the instrument is needed before it should move from research into practice.

Subject Area

Eating disorders--Diagnosis: Minnesota Multiphasic Personality Inventory


Files over 3MB may be slow to open. For best results, right-click and select "save as..."