This study proposes a Bayesian modal identification approach for assessing the contributions of supplemental damping and stiffness provided by dampers to building structures. The proposed approach can be utilized to quantitatively estimate the effect of such contributions on modal parameters，such as damping ratios，frequencies，and mode shapes，as well as their associated estimation uncertainty. The proposed approach adopts a new probabilistic modelling strategy，in which a direct model updating technique is utilized to establish the accurate relationship between the modal parameters and mass，stiffness and damping matrices of the structure. Given certain values of modal parameters，these matrices are uniquely determined. Then，a time-domain Bayesian model updating method is utilized to identify the modal parameters and quantify their estimation uncertainty resulting from unavoidable modelling and measurement errors. The effectiveness of the proposed approach is demonstrated with a real-world passively controlled building，in which the modal parameters of the structure system and of the main frame are separately identified by using its acceleration measurements during a selected large earthquake. By comparing those identification results，the modal contributions，the change between the identified modal parameters，provided by the added oil dampers are discussed.