Body image measurement in the prediction of eating pathology

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

William Goggin

Advisor Department



The study of body image has expanded exponentially in the past few decades, in large part because of its implication in understanding eating disorders. However, it continues to be an extremely difficult construct to objectively measure in a personally meaningful manner. The purpose of the current study was to provide evidence for the reliability and validity of the Individualized Body Image Task (IBIT), which is a measure of body-image disturbance. Additionally, the construct validity of the IBIT in identifying individuals who had eating disorders or eating disorder symptoms, was assessed. Thirty-nine undergraduate women, and 34-hospitalized patients who had eating disorders, were recruited to participate if they were over the age of 18, not pregnant and for the normal controls, not dieting by physician order. The IBIT was compared with the Body Image Assessment (BIA) and the Body Shape Questionnaire (BSQ), two commonly used measures of body-image disturbance. A Multi-Trait, Multi Method Matrix was used to test the hypothesis that the IBIT has adequate test-retest reliability and construct validity (discriminant and convergent validity). The utility of the IBIT was assessed, using logistic regression analyses to predict the presence or absence of eating disturbance, and multiple regression to predict level of eating disturbance (drive for thinness, and bulimia). First, results indicated adequate reliabilities in all cases for the current body size, ideal body size, and the dissatisfaction subscales of the IBIT. Second, results provided partial support for the construct validity of the IBIT, because the current body-size and dissatisfaction subscales of the IBIT and BIA were moderately correlated, but the ideal body-size subscales of the IBIT and BIA were not correlated. Third, results suggested discriminant validity for the IBIT because the current body-size measure was not correlated with the ideal body size measure. Fourth, IBIT-DISS was found to be related to body-image concern (as measured by the BSQ), eating-disorder behaviors as measured by the Eating Disorder Inventory-(EDI), and body-image dissatisfaction (as measured by the EDI), which suggested criterion-related validity. Fifth, the BSQ was found to be the only measure, and not the IBIT and the BIA, which predicted group membership (presence or absence of an eating disorder). Sixth, the IBIT was a better predictor of drive for thinness and bulimia than was the BIA, but the BSQ was a better predictor than the IBIT. Several limitations of the current study, such as sample size, differences in the scale of measurement of body image, and multicollinearity were discussed. Finally, future research directions were suggested.