As a combination of statistical and mathematical techniques, response surface methodology gives explicit functions to express the relationship between the inputs and outputs of a physical system. This methodology has been widely applied to design optimization, response prediction and model validation but so far little literature related to its application in structural damage identification has been found. Therefore this paper presents a systematic damage identification procedure consisting of four steps of feature selection, parameter screening, primary response surface modeling and updating, reference-state response surface modeling with damage identification realization. 2k factorial design and central composite design are adopted to construct response surface models for parameter screening and model updating purposes, respectively. The proposed method is verified against an experimental reinforced concrete frame and it is found that the proposed method works well in damage prediction.