Abstract: Most historical data have a complex generation process and data may vary widely due to different historical sources, measurement standards, statistical methods and estimates. Therefore, it is necessary to analyze the basic data in quantitative history meticulously. However, former studies of famine history neither paid enough attention to the regulation of famines, characteristics of the documentation and the system of famine relief, nor did these studies use sophisticated methods of statistics and calculation. So both the disaster records and famine relief records are inaccurate. In the process of obtaining historical data, not only do we have to consider fully the reality of academic research to use the data scientifically, but we also must grasp the basic characteristics of the object of study and the inner logic of documentary records in order to use the data historically. It is important to pay attention to data matching and its scope of application when dealing with data.

摘要： 有关历史时期的数据大都有着复杂的生成过程，根据不同史料来源、计量标准、统计方法、 估算依据所得出的结果千差万别。因此，进行量化历史研究时，有必要对基础数据本身进行一番细致的考 察。已有的灾荒史研究中，由于对灾荒规律、文献特征、荒政制度等方面欠缺考虑，以及统计和估算方法 上的粗糙与失误，相关的灾情、灾赈数据还有待进一步完善。在量化数据的获取过程中，不仅要充分考虑 学术研究的现实需要，实现方法的科学化，更要注意把握研究对象的基本特征和文献记录的内在逻辑，实 现方法的史学化。在数据的运用过程中，要特别注意数据之间的匹配关系及其适用范围。