Discrimination of Performance Art Groups Using Anthropometric Variables

Background/Purpose: Physique is a significant factor to success in dance and fashion model performances and is particularly important from an aesthetic point of view, where the most successful dancers could be expected to acquire morphology suitable for their performance task (Carter, 1985). Much time and effort has been spent trying to identify the particular physical characteristics that contribute to the selection and development of talents in dance and fashion model performances. The purpose of this study was using discriminant function analysis to predict performance art group membership for female college students majoring in dance and fashion modeling based on the selected variables, including height, weight, body girth, length, and breadth variables.

Method: Participants were 284 female elite dance and fashion model students (95 Korean dance; 99 Dance sport; and 90 Fashion model). Predictor variables were height, weight, body lengths (sitting height, arm and leg lengths, and foot length), body girths (head, neck, arm-relaxed, chest, waist, hips, thigh, and calf), and body breadths (humerus and femur). Participants were measured by experienced anthropometrists using an internationally-recognized protocol proposed by International Society for the Advancement of Kinanthropometry (ISAK) to describe physique. Evaluations of the assumptions of linearity, normality, and multicollinearity were satisfactory. Stepwise discriminant function analysis by SPSS DISCRIMINANT was used to predict membership in three groups.

Analysis/Results: Two discriminant functions were calculated, with a combined x²(20) = 530.82, p < .001. There was strong association between groups and predictors, x²(9) = 168.85, p < .001, even after removal of the first function. The two discriminant functions accounted for 76.3%, and 23.7%, respectively, of the between-group variability. The first discriminant function maximally separates fashion models from Korean dance and dance sport groups. The second discriminant function discriminates dance sport group from Korean dance group, with fashion models falling between these two groups. The best predictors for distinguishing between fashion models and the other two groups are height and waist girth. Fashion models have taller height and smaller waist girth than Korean dance and dance sport groups. The classification results showed that 88.0% (n=250) of the samples were correctly classified into their groups while 25.0% (n=71) were correctly classified based on chance.

Conclusions: Body girths, lengths, and breadths are significant predictors of performance art group membership. Further studies and analyses are warranted to develop models to predict success in different performance art groups.

Keywords: anthropometry, dance, performance art, discriminant function analysis