The Historian and Social Science Data Archives in the United States

DURING THE PAST thirty years social scientists have learned that quantitative data, when converted to machine-readable form in the course of even relatively modest research projects, may also be useful to scholars interested in other research problems as well as to those who wish to use it for replicative research.When a source is so rich that it provides data for many scholars working on different kinds of research in a variety of disciplines, the utility of placing it in standard machine-readable format ready for easy distribution becomes readily apparent.T h e onerous tasks of conversion would then need be undertaken only once; subsequent investigators can begin their research at the stage of analysis, and an almost incalculable amount of time and labor can be saved within the research community.Such is the rationale that underlies the development of machine-readable data archives.Does it also apply to historical data as well as to data generated by survey analysis and the contemporary activity of governments, business firms and other societal institutions?Certainly there are important historical sources that are quantitative in nature, or that can be converted to quantitative form.Historians active in research are comparable in number to researchers in the more populous social science disciplines and some of them have created or helped to create machine-readable data files.T h e analogy can be pushed too far, however, and in this article I will discuss the degree to which historians have been involved in the data archiving movement and the challenges it presents to their discipline.