In this study, a chemometrics approach has been used to optimize the particle size of quercetin-containing nanoemulsions prepared with spontaneous emulsification method. The experiments were performed according to the Box – Behnken experimental design, one of the most suitable experimental designs for modeling studies. The effect of three experimental parameters on the droplet size was studied using multivariate analysis. The factors studied (and their variation levels) were the concentration of lecithin in aqueous phase (0.7 – 2% w/w), the concentration of tween-80 in aqueous phase (2 – 8% w/w), and sonication time (10 – 60 minutes). In each step of experimental design, the aqueous phase was added to the organic phase including lecithin, tween-80, and quercetin in ethanol solvent. Then, the mixture was treated in ultrasonic bath for 15 min and organic solvent was removed by rotary evaporator. The droplet size was measured by a Zetasizer instrument. After determining the average particle size for each experimental run according to experimental design, modeling of this parameter was conducted in terms of experimental factors by using multiple linear regressions with SPSS software. The obtained regression model was characterized by both its descriptive and predictive ability (R 2 =0.999, standard error SE =4.205, and F =245.698) and allowed the preparation of quercetin nanoemulsions in a desired range.