Test‐retest reliability for the eating disorder inventory

The Eating Disorder Inventory (EDI) has been demonstrated to have high internal consistency in the form of split-half reliability. Demonstration of stability in the form of test-retest reliability has been absent. Test-retest reliability for the EDI was obtained with a sample of 70 university undergraduates. Test-retest reliability was also obtained with a restricted group drawn from the original sample who were considered to be at risk on the basis of their EDI scores. The interval between test administrations was 3 weeks. With both samples, test-retest reliability of the total EDI score was exceptionally high. Test-retest reliability for the eight subscales was, with one exception, within the usual range of acceptability.