An examination of the internal consistency and factor structure of the eating disorder inventory-2 in a clinical sample

Int J Eat Disord. 1994 Dec;16(4):371-9. doi: 10.1002/1098-108x(199412)16:4<371::aid-eat2260160406>;2-w.


We examined the internal consistency and factor structure of the newly revised version of the Eating Disorder Inventory (EDI), with particular emphasis on the examination of the three new provisional scales: Asceticism, Impulse Regulation, and Social Insecurity. Subjects were 300 women who were seeking treatment at a residential treatment facility for women with eating disorders. Chronbach's alphas ranged from .80 to .91 for the eight original scales whereas coefficients for the three new scales were all less than .80, the criterion stated as being necessary for inclusion in the test. Item-total correlations were also substantially lower for the three new scales. The results of principal components analyses with orthogonal rotation supported the factor structure of the original 64-item EDI, but not the three additional scales. These data generally supported the reliability and factor structure of the original eight scales but did not support the reliability or validity of the three provisional scales.

MeSH terms

  • Adolescent
  • Adult
  • Feeding and Eating Disorders / psychology*
  • Female
  • Humans
  • Middle Aged
  • Psychiatric Status Rating Scales*
  • Psychometrics
  • Reproducibility of Results