Statistical modelling of key body dimensions in developing a size chart for the full-figured, pear-shaped South African women

Abstract

Anthropometric body measurement is a crucial process in the development of size charts for garment manufacturing. Body measurements differ between specific population groups, and garments catering for different populations must be manufactured based upon accurate size charts informed by accurate body measurements for that population. Amongst population groups, the full-figured, pear-shaped South African woman is a unique identifiable body type. This body type is not adequately catered for in garment manufacturing, as it requires a unique configuration of garments with different sizes for the upper and lower torso. The relative absence of well-fitting garments for this body type necessitates taking body measurements in order to develop a size chart to inform design and manufacturing of garments.

The purpose of this paper is to develop a statistical model of key body dimensions (bust, waist and hip) to populate a size chart for the manufacturing of ready-to-wear garments for the full-figured, pear-shaped South African woman. A correlational research method using body measurements from purposively selected women of ages 25 to 55 years was carried out. After categorizing the height measurements into three groups, the means of body measurements for the medium height group were used to develop a size chart for sizes 16 to size 24, using principal component analysis (PCA) and least squares regression.

Results showed that the bust, waist and hip values highly contribute towards body type sizing. The bust and hip were highly correlated at R2=0.996 (99.6%), and the model predicted the true value of the hip at R2=0.993 (99.3%). Bust measurement positively correlated to the waist at 93.8%, and the model predicted the true value of the waist at R2=0.880 (88%). Bust dimension was significant in predicting the hip and waist dimensions. Variances among hip dimensions in the current sizing were in 4 to 6 cm intervals. Full-figured, pear-shaped figures present 5.5 to 7.5 cm values across the hip measurements.

In conclusion, findings of the simulated values for hip and waist at different sizes based on bust measurements suggest that the values obtained respectively were almost the same with measurements in the customised size chart developed in the study. This makes the model dependable, reliable and valid for the size chart determination targeting the full-figured, pear-shaped South African woman. It is recommended that the model be used for determining size charts for other body shapes.