Analysis of the EAT-26 in a non-clinical sample

The Eating Attitudes Test (EAT) is one of the most commonly used measures of abnormal eating habits; however, it has been criticized for its unstable factorial structure. Different studies proposed different solutions, but as yet there is no consensus whether the three-, five- or seven-factor solution is appropriate. We examined the cues provided by previous studies that investigated the EAT structure, and tested our expectations on a non-clinical group of 617 women. Based on bi-factor confirmatory factor analysis, we demonstrated that the EAT measures general eating pathology as well as more specific factors, such as ‘Social pressure’, ‘Food awareness’, ‘Food preoccupation’ and ‘Purging behaviors’. Among distinguished specific factors, ‘Social pressure’ seems to be a promising scale to be used for screening purposes. The ‘Food awareness’ factor could be interpreted in terms of orthorexia nervosa; however, further evidence is required to support this conclusion. We also provided evidence that ‘Bulimia’ and ‘Food preoccupation’ factors should be maintained within the EAT-26 as important behavioral aspects of eating pathology. These results shed new light on this measure and pro-vide the basis for a discussion of its psychometric issues.