Measures of eating-related characteristics for the general population: relationships with temperament

A questionnaire that assessed a broad range of eating-related characteristics for unselected, normal subjects was factor analyzed in a two-step process proposed by Comrey (1984). Twelve "factored homogeneous item dimensions" were identified first and yielded three primary-level factors in a second factor analysis: Predisposition to Obesity (including Dieting and Preoccupation with, and Fear Of, Gaining Weight), Uncontrollable Urges to Eat (including Eating Momentum Beyond Control, Food a Panacea and Constant Temptation, and Secret Binging), and Predisposition to Anorexia (including Insufficient Eating Obvious to Others, Food Phobia, Inability to Eat, and Vomiting After Meals). The three primary-level factors were positively intercorrelated and exhibited significant positive, though weak, correlations with a measure of trait arousability. Also, weak results tentatively indicated that individuals with more pleasant and/or more arousable temperaments were less likely to be overweight. Subjects reported sharply higher levels of food consumption when feeling "depressed" (i.e., bored, lonely, sad) than when feeling "distressed" (i.e., uncomfortable, anxious, in pain). Uncontrollable Urges to Eat correlated positively and significantly with self-reports of food consumption while depressed, showing that those lacking control over eating ate especially more while feeling bored, lonely, or sad. Predisposition to Obesity correlated negatively and significantly with self-reports of food consumption while distressed, showing that those tending more toward obesity ate less while upset or anxious.