Abstract This study shows the effectiveness of hierarchical agglomerative cluster analysis (HACA), discriminant analysis (DA), principal component analysis (PCA), and multiple linear regressions (MLR) for assessment of air quality data and recognition of air pollution sources. 12 months data (January-December 2007) consisting of 14 stations in Peninsular Malaysia with 14 parameters were applied. Three significant clusters - low pollution source (LPS), moderate pollution source (MPS), and slightly high pollution source (SHPS) were generated via HACA. Forward stepwise of DA managed to discriminate eight variables, whereas backward stepwise of DA managed to discriminate nine variables out of fourteen variables. The PCA and FA results show the main contributor of air pollution in Peninsular Malaysia is the combustion of fossil fuel from industrial activities, transportation and agriculture systems. Four MLR models show that PM10 account as the most and the highest pollution contributor to Malaysian air quality. From the study, it can be stipulated that the application of chemometrics techniques can disclose meaningful information on the spatial variability of a large and complex air quality data. A clearer review about the air quality and a novelty design of air quality monitoring network for better management of air pollution can be achieved via these methods.