Psychometric Contributions

Questionnaire Development

The Food Approach and Avoidance Questionnaire (FAAQ) captures approach food craving (i.e., desire to consume) and avoidance food craving (i.e., desire to not consume). This 12-item measure can be shorted to as few as 8 items (4 items per subscale) and demonstrates diagnostic efficiency for use as an eating disorder screening measure. Specifically, the FAAQ adequately predicts sub- and full threshold anorexia nervosa (2.6 on the Avoidance subscale), sub- and full threshold bulimia nervosa (2.6 on the Avoidance subscale; 3.1 on the Approach subscale) and both sub- and full threshold binge eating disorder (3.1 Approach subscale). 

We have replicated the FAAQ factor structure in an undergraduate sample (Ahlich et al., 2020) and have demonstrated that approach and avoidance food craving are unique from approach and avoidance personality traits (Cunning et al., in preparation). Interestingly, in non-clinical samples, there seems to be an additive, rather than interactive, effect of approach and avoidance food craving on disordered eating behaviors (Rancourt et al., 2024).

The FAAQ is available for use by researchers here (coming soon). We also are in the process of translating this measure into Spanish.

Psychometric Research

CEBRACS

The Compensatory Eating Behaviors Related to Alcohol Consumption (CEBRACS) is a popular measure of FAD that was originally developed at USF. We recently re-evaluated the factor structure based on the original scoring methods and proposed an alternate scoring approach based on when individuals engage in FAD behaviors (i.e., before, during, or after alcohol use).

We have collected data to expand the CEBRACS to include additional FAD behaviors not currently included in the questionnaire (data analysis in progress).

EDI-Drive for Thinness

The EDI-Drive for Thinness subscale has demonstrated strong psychometric properties; however, until recently it has only been investigated in terms of categorical moderators (e.g., biological sex). We demonstrate that while the EDI-Drive for Thinness subscale was not invariant across undergraduate and community samples, and potential item bias was associated with sample type, age, and BMI for six of the seven items, this item bias appeared to be associated with minimal clinical impact (Rancourt et al., 2022).