To reconstruct historical social-economic series is one of the main difficulties in the research of socio-economic impacts of historical climate change. Based on the characteristics that a lot of social and economic information was scattered in the rich, long-term, and continuous historical literatures of China, a methodology for quantifying social and economic series on the base of semantic differential is proposed in this study. Semantic differential method originated from psychology. It usually adopts bipolar (opposite) adjectives to describe the properties of the objects, and quantitatively distinguishes the different levels according to the differential of semantics expressed by the additional words. The two key technologies of quantifying social and economic series based on semantic differential are gradation of literal description and uniformity of time resolution. The core of the semantic differential method is to establish the standards of gradation,which needs to consider both the differentiation degrees of the standard vocabulary and the physical meaning of the grades. The uniformity of time resolution focuses on the problems of the unequal time resolutions or discontinuity of the original grade series. Aimed at the 10 years as the basic time unit, up scaling method is used to convert the annual data, and down scaling method is used to convert the multi-decadal data.