This paper aims at examining the extent of attrition biases specifically included in Stated Preference data and also demonstrates the effectiveness of a correcting method which exclude them in model building. SP panel data for the New Light Rail Transit System in Hiroshima, measured at two different points in time, was used for this purpose. As a result, the effect of attrition bias on mode choice model parameters based on SP data was empirically clarified. Attrition models which determine whether the respondents of the first wave participate in the second survey or not were developed using first wave data. In addition, a mode choice model was constructed based on the stayers (i.e. participants in the first wave who participated in the second wave) at the second wave. The attrition bias of this model was corrected by sequential steps on the assumption that the error in the mode choice model correlates with that of the attrition model. It was found that this type of correcting could effectively cancel out the biased share of each travel mode.