Objective
This project was designed to develop and test the psychometric properties and factor structure of a revision of the Mizes Anorectic Cognitions questionnaire (MAC). The goals of the revision were to improve the reliability and discriminant validity of the Weight and Approval subscale and to equalize the length of the three subscales. Also, the study compared the original MAC and the MAC‐R in terms of their psychometric properties.
Method
Twenty‐four new items were developed for potential inclusion in the MAC‐R, in addition to the original 33 items of the MAC. These items were administered to 205 eating disorder patients from five eating disorder clinics or programs, including inpatient, outpatient, and residential treatment settings that served diverse patient populations. Additionally, other measures of eating disorder constructs were administered to assess construct validity.
Results
Factor analysis of the large pool of items and item reduction resulted in the final 24‐item MAC‐R, each subscale being eight items in length. Results showed that the MAC‐R highly correlated with the MAC and other eating disorder questionnaires. Reliability of the MAC‐R was improved over that of the MAC. Two subscales of the MAC‐R discriminated among diagnostic groups, whereas the original MAC did not, indicating improved sensitivity of the revised scale.
Discussion
The MAC‐R appears to be an improvement over the original MAC. It provides useful information on the cognitions of eating‐disordered patients and merits further investigation into its psychometric properties. © 2000 by John Wiley & Sons, Inc. Int J Eat Disord 28: 415–421, 2000.