Partial least squares (PLS) multivariate statistical models were developed to predict the surface composition and chemistry of a set of model homopolymers based on their static SIMS fragmentation patterns. In the calibration or model-building step, the positive and negative ion static SIMS spectra of different classes of model homopolymers were related to specific chemical attributes of the polymers. The models were then used to examine the surface chemistry of oxygen-containing plasma-deposited films prepared from a variety of precursors. PLS models were developed to predict the surface oxygen concentration and H/C ratios. The results obtained from the PLS models were compared with experimental results.