Abstract: The shear stiffness of saturated sand deposit will drop significantly under the disturbance of strong earthquake shaking (e.g., liquefaction) and then recover gradually with time. The difference between the post-earthquake field testing index and the pre-earthquake value will cause systematic error in the simplified method of liquefaction evaluation based on the field case histories. In order to evaluate this difference and propose the correction approach, the HVSR method is used to analyze the acceleration records at REHS strong motion station in Christchurch from 2010 to 2011, and to observe the time variation of the small-strain shear stiffness of the liquefiable sandy soil deposit after each strong earthquake event. It is found that the average shear stiffness of the deposit drops suddenly after earthquake and then increases logarithmically, and it will take one to two weeks to approach a relatively stable state but cannot totally recover the pre-earthquake value. By considering the combined effects of the primary consolidation and the secondary consolidation, a computational model for post-earthquake small-strain shear modulus of saturated sandy soils is proposed. The model predicts the general trend of the time-dependent development of site stiffness after the occurrence of earthquake, and can be regarded as a feasible way to correct the post-earthquake field testing index to the corresponding pre-earthquake value and help to improve the reliability of the existing simplified methods for liquefaction evaluation based on the field case histories.