Invariance, Convergent and Discriminant Validity Across Occasions for the Male Body Image Disturbance Scale

Timothy Freson

The Male Body Image Disturbance Scale (MBIDS) is a multidimensional questionnaire to assess the symptoms associated with body image issues in men (Freson, 2011). The initial research developed a 25-item, five-factor questionnaire, where the five factors were Body Dissatisfaction (BDIS), Drive for Muscularity (DMUS), Drive for Leanness (DL), Obsessive Exercise Pattern (OEP), and Interference with Life Functions and Responsibilities (ILFR). The purpose of this study was to evaluate the invariance of the five-factor model along with the convergent and discriminant validity of the five-factors across the one-month interval. A random sample of men (n = 626 at time one and 308 at time 2) from the general population was used for these analyses. The sample was recruited with the assistance of Qualtrics, a company that specializes in online panel aggregation. The first two analyses applied an a priori five-factor exploratory factor analysis to the time one and time two data (two separate analyses) for identification of any weak items (e.g., items with high cross-loadings). A total of 22 of the 25 items showed good convergent and discriminant validity. A series of confirmatory factor analyses were then used to evaluate the invariance of the scores on the 22-item MBIDS across the one-month interval (i.e., (1) invariance of form with correlated residuals between the same items across time; (2) invariance of like-item loadings; (3) invariance of like-item intercepts; (4) invariance of like-item residuals; (5) invariance of like-factor variances; (6) invariance of like-factor covariances; and (7) and invariance of like-factor means) as well as the convergent and discriminant validity of the five-factors across the interval. The results provided strong support for the invariance of like item-loadings, like-item intercepts, like-item residuals, like-factor variances, like-factor covariances, and like-factor means. In the addition, the five-factors showed convergent validity (like factor correlations across the interval were significant and substantial) as well as discriminant validity (like factors correlated significantly stronger across the interval than different factors across the interval). The resulting MBIDS is a brief 22-item questionnaire with good construct validity for the assessment of male body image disturbance.

Invariance, Convergent and Discriminant Validity Across Occasions for the Male Body Image Disturbance Scale

Files and links (1)

Abstract

Metrics

Details