Fourier transform infrared spectroscopy (FTIR) coupled to multivariate analysis was used to detect and quantify the adulteration of avocado oil with different edible oils (sunflower, canola and soybean). A SIMCA (Soft Independent Modeling Class Analogy) method was also developed to discriminate between adulterated and unadulterated samples. Avocado oil and adulterants (sunflower, canola and soybean oils) were characterized based upon their chemical analysis (peroxide and iodine value). In order to build the quantitative calibration model for each adulterant, mixtures of avocado oil and each adulterant were prepared in a range of 2-50% (vol/vol). Partial Least Square (PLS) algorithm was tested to model each system (avocado oil+adulterant) and the chemical analysis of the mixture. The SIMCA model developed showed 100% correct classification rate of adulterated samples from unadulterated ones. The PLS model shows values of R2 greater than 0.98, standard errors of calibration (SEC) in the range of 0.04-1.47 and standard errors of prediction (SEP estimated) between 0.09 and 2.81 for both the adulterant and chemical analyses. Chemometric models represent a rapid and attractive option for avocado oil quality screening without sample pretreatments. © 2012 Elsevier Ltd.