Page 78 - Anthropometry, Apparel Sizing and Design

Developing apparel sizing system using anthropometric data 107

After analysis of factor loading, the current findings yield two significant compo-
nents, which consist of variables of the highest factor loading. Thus two components
are retained, namely, the principal component 1 (PC1) and principal component 2
(PC2) for each sample groups. For sample group male and female (LaBat, 1987;
Otieno and Fairhurst, 2000; Ashdown, 2014; Manuel et al., 2010; Kaiser, 1997),
44 variables and 42 variables, respectively, are grouped into two components. PCA
technique has successfully achieved the goal of getting a parsimonious group of vari-
ables, which extracted almost all of the variables in two components.
The two retained components gave cumulative percentage as follows: female
(LaBat, 1987; Otieno and Fairhurst, 2000; Ashdown, 2014; Manuel et al., 2010;
Kaiser, 1997) with 60.0%. The present findings seem to be consistent with other
research, which commonly retained two final components. However, the cumulative
percentage for these two components was found differently in different studies rang-
ing from 65% to 81% (Xia and Istook, 2017; Xinzhou et al., 2018).
Consequently, it was found that the first component (PC1) consists of all girth
dimensions including bust girth, chest girth, upper arm girth, hip girth, and waist girth
and few width dimensions like back width and shoulder width. The second component
(PC2) consists of all length dimensions such as height, cervical height, upper arm
length, and arm length. This finding is found similar to many previous sizing studies
where two components represent the girth and length factors (Hsu, 2008, 2009;
Beazley, 1998b).

4.4.4 The results of factor loadings analysis

In this section, all the body dimensions with high factor loading ( 0.75) are listed in
Table 4.5 for one sample female group. The aim of this section is to select the key
dimensions. It is apparent from this table that most of the variables are highly corre-
lated to the individual components, which can be seen from the high factor loadings.
Overall, almost all variables are reduced into two factors.
As can be seen from Table 4.5, upper arm girth and waist girth are distinguished as
the strongest variables correlated to girth factor for male samples (age 13–17). In con-
trast, bust girth has the highest factor loading for female samples age 13–17. Under
arm length and inside leg length are noted to have the highest factor loading correlated
to length in males aged 13–17 as compared with arm length and hip height in females
aged 13–17. Hence PCA analysis for each sample group is shown later.

4.4.4.1 Female samples (age 13–17)

For female samples, as can be seen from Table 4.5, 28 variables were found correlated
to the girth and length component as compared with 26 variables in female samples.
On the other hand, 16 and 15 variables are loaded on length girths for males and
females, respectively.