A principal component analysis was performed on the cephalometric variables of 100 male obstructive sleep apnea (OSA) patients. Thirty cephalometric variables of cervicocraniofacial skeletal morphology were reduced to 8 principal components (PCs), which described 83.2% of the total variance. Sixteen cephalometric variables of hyoid bone position and head posture were reduced to 4 PCs, which described 85.5% of the total variance. Twenty cephalometric variables of upper airway soft tissue were reduced to 7 PCs, which described 83.7% of the total variance. These PCs described the actual characteristics of the OSA patients examined. For further analysis of PCs, stepwise multiple regression analysis was chosen. Two dependent variables of interest are the minimal distance of posterior pharyngeal airway space (PASmin) and the apnea-hypopnoea index (AHI). The PASmin was reduced to 7 PCs, which accounted for 79.4% of the variance and AHI was reduced to 3 PCs, which accounted for 37.6% of the variance. Both principal component analysis and multiple regression analysis provide multivariate data analysis that is very useful in sorting out and clarifying the complexity of the interrelated cervico-craniofacial skeletal morphology and upper airway soft tissue in OSA patients.