Multivariate statistical techniques, hierarchical cluster analysis (HCA), and principal component analysis (PCA) integrating graphical method (Piper trilinear graphical diagram) were applied to the factor identification of ground water quality in a coastal aquifer, Fujian province, South China. Ground water samples were collected at 12 sites in January (dry season) and July 2011 (wet season). Eleven ground water quality parameters (pH, TH, TDS, Ca²⁺, Mg²⁺, Na⁺, Cl^¯, SO₄ ^2-, HCO₃^¯, NO₃^¯, Mn) were selected in order to perform multivariate statistical analysis. During both the past-monsoon and the summer seasons, PCA results revealed the existence of three significant principal components revealing how processes like salinization, water-rock interaction, and anthropogenic pollution influence ground water quality. Three factors which together explain 90.3% and 83.3% of the total variance in the summer and post-monsoon dataset were retained and interpreted. Cluster analysis using the Ward method with squared Euclidean distance measure was performed, which indicated the distribution of the studied wells according to their water quality. Water samples from 12 wells were clustered into three distinct groups to depict different hydrochemical facies. The results proved that multivariate analysis methods like HCA and PCA could be useful for evaluating ground water pollution and identifying ground water hydrochemistry.