Abstract: Investigating the difference between the calculated and measured values of the collapsible settlement is useful and necessary to accurately evaluate the loess collapsibility. Although the difference has been noted by many researches, it has not yet been examined deeply and systematically. Based on the results of two field loess immersion tests in Xi’an loess tableland area, the numerical method is used to compare the collapsible deformations of loess foundation with different proportions and arrangements of collapsible and non-collapsible soils as well as with different collapsible degrees of collapsible soils. The results reveal that the inhibition effect on collapsible deformation, which is caused by the discontinuous distribution of collapsible loess, is one of the main reasons for the discrepancy between the calculated and measured values of the collapsible settlement. In addition, this inhibition effect is more significant for the cases with more random distribution of collapsible and non-collapsible soils, or with higher proportion of non-collapsible soils. Through comparison, the impact on the inhibition effect caused by the collapsible degree of collapsible soils is not very strong. The inhibition effect should be taken into account in calculating the collapsible deformation, and the related data should be collected and summarized in future engineering practice.