Purpose: Designed as a self-report measure of "psychological and behavioral traits common in anorexia nervosa and bulimia."
Population: Ages 12 and over.
Scores: Eight subscale scores.
Time: (15-25) minutes.
Authors: David M. Garner, Marion P. Olmsted, and Janet Polivy.
Publisher: Psychological Assessment Resources, Inc.
Description: The Eating Disorder Inventory (EDI) is a 64-item, 6-point forced-choice inventory assessing several behavioral and psychological traits common in two eating disorders, bulimia and anorexia nervosa. The EDI, a self-report measure, may be utilized as a screening device, outcome measure, or part of typological research. It is not purported to be a diagnostic test for anorexia nervosa or bulimia.
Scoring: The EDI consists of 8 subscale scores: drive for thinness, bulimia, body dissatisfaction, ineffectiveness, perfectionism, interpersonal distrust, interoceptive awareness, and maturity fears.
Reliability: The average item total correlation of the eight subscales of the Eating Disorder Inventory was .63 (SD = .13). Reliability information was based on 271 college women on whom completed information on all subscales was obtained. Reliability coefficients (standardized Chronbach's alphas) for the anorexia nervosa group ranged from .83 (Interoceptive Awareness) to .93 (Ineffectiveness). Reliability coefficients for the female college students ranged from .72 (Maturity Fears) to .92 (Body Dissatisfaction).
Validity: Criterion-related validity studies were performed by comparing the EDI patient profiles with the judgments of clinicians familiar with the patient's psychological presentation. A subgroup of 49 of the anorexia nervosa patients who had completed the EDI was assigned two raters: a psychologist and psychiatrist who were familiar with the patients, being their primary therapist or consultant. The raters were provided with the description of the subscale content and with the patients' total score percentile rank within the entire anoretic sample. All interrater correlations were significant at the p < .001 level and ranged from .43 (Maturity Fears) to .68 (Ineffectiveness).
Norms: Two groups of respondents participated in the validation of the EDI. The criterion group (n = 129) was composed of three subsamples of women, mostly anoretics. These women averaged 20% below the expected weight for their height and age. In this sample, 56 were classified as "restricters" and 73 were diagnosed as "bulimic." The comparison group (n = 770) consisted of three samples of female university students who were enrolled in introductory and upper-level psychology classes. These volunteers were administered the EDI in their classes.
Suggested Uses: The EDI is recommended to delineate subtypes
of anorexia nervosa in clinical or research settings.