The present investigation evaluated the psychometric integrity of the Attitudes to Chocolate Questionnaire, an instrument recently developed to assess various aspects of this specific craving experience. The scale consists of three subscales: Cravings (preoccupation with chocolate); Guilt (negative affect following chocolate consumption); and Function (practical uses for chocolate). A confirmatory factor analysis (n=701) did not support the 3-factor solution, but an exploratory factor analysis identified only 2 stable factors: Craving and Guilt. Internal consistency estimates were high for both subscales, and females scored significantly higher than males on both. In the test of subscale validity and temporal reliability (n=96), the retest correlation after 3 months for both subscales was high (above 0.68; n=47). High Craving scores were associated with high depression and obsession; but low self-esteem and social desirability. High Guilt scores were associated with high anxiety, depression, disordered eating patterns, and obsession; but with low self-esteem. Implications for the diagnosis of eating disorders and directions for future research are discussed.