•A reliable method was developed for the analysis of four bisphenols in foodstuffs.•Plackett-Burman design and response surface methodology were used to determine the optimal conditions.•Dietary exposure assessment was performed combining the contamination monitoring data and daily food consumption. An effective method has been developed for the simultaneous determination of four bisphenols (bisphenol A, S, F and B) in various foodstuffs. The contaminants were extracted by QuEChERS-based strategy and subjected to ion-exchange solid-phase extraction for further clean-up. The critical variables were screened by Plackett-Burman design and then optimized by central composite design. Under the optimized conditions, satisfactory accuracy (recoveries 76%–116%) and precision (RSDs < 12%) were achieved. The established method was then used to assess the contamination status of 379 real samples. Bisphenol A was demonstrated to be the predominant bisphenol with highest incidence (79.7%) and average concentration (14.3 μg/kg). The positive rates (mean concentration) of bisphenol S, F and B were 37.7% (1.6 μg/kg), 26.9% (1.4 μg/kg) and 0.0% (not detected), respectively. Finally, the daily dietary intakes of ∑4bisphenolsfor adult residents were estimated (55.9–76.1 ng/kg bw/day) according to the contamination concentrations and the daily recommended intakes.