PCA outputs for the data matrix are shown in Fig. 1. Dimension
reduction by PCA separated the observed factors into two distinct
directions: kitchen related and non-kitchen related statements.
PC1 (kitchen related component) has high positive loadings (>0.5)
on hygiene, cooking, quality, approval, storage and temperature
with no negative loadings. PC2 (non-kitchen related component)
has high positive loadings (>0.5) on employees, inspections and
negative loading on serving of food.