中文摘要：

针对SP数据存在的行为结果与意向结果不一致的问题，介绍了利用RP数据修正SP数据的方法．以北京为例，建立了RP模型、SP模型和融合数据模型，并对模型的灵敏度进行了分析，结果表明，RP模型中费用变量的参数估计值符号与实际不符；SP数据模型精度较高，但某些参数的影响不是很显著；融合数据模型的参数检验较显著，模型的弹性小，敏感性被钝化．

英文摘要：

The inconsistence of SP data between behavior and behavior intention is a serious problem. The article introduces the method of revises the SP data using the RP data. Then take Beijing as the example, establishes the RP data models, the SP data model and the fusion data model, it carries on the analysis of model sensitivity. The results indicate： the mark of expense parameter is not logical in the RP data mode; The precision of SP data model is high, but the effects of some parameters are not remarkable; The fusion data model parameters＇ test are remarkable, the model elasticity is small, and its＇ sensitivity is deactivated.