In this study, an experimental design was used to optimize dispersive liquid-liquid microextraction for the extraction/preconcentration of micropollutants that have endocrine disruption potential, for the determination by gas chromatography-mass spectrometry. The analytes were selected from endocrine disrupting chemicals, such as pesticides, phthalates, hormones, alkyl phenols and pharmaceuticals. The most influential parameters of the extraction method determined in a pre-optimization step were included in the design model to determine their optimum amounts and to evaluate their interactions in a series of combinations. The design model predictor was used to obtain optimum experimental values, with which analytical performance of each analyte was determined with calibration plots. Deuterated bisphenol A (BPA D16) was added as an internal standard to all samples and calibration standards to improve the precision of the results. The limits of detection for analytes ranged between 0.10 and 16 mu g/L. The developed method was applied to synthetic and municipal wastewater samples, and the results obtained from spiked recovery tests ranged between 79 and 113%. These results confirmed the applicability of the method for the selected sample matrices. The complex nature of the samples tested in this study affected the recovery of analytes. However, matrix matching calibration was employed to overcome the matrix effects and this improved the recovery results.