The data documentation initiative: a preservation standard for research

When the answer is the number “42” (or actually “forty-two”) some know instantly that the question is “The Great Question” concerning “Life, the Universe and Everything ...” (Adams 1986, p. 128). This demonstrates that even when we know the answer to a question its meaning and usefulness are not always obvious. Context is required. This is especially true for quantitative data. More information is needed in order to understand the numbers and transform data into useful knowledge. This further information is itself data, and thus metadata or “data about data”. The importance of context and metadata are widely recognized in the social science community. This paper discusses a project to provide standardized metadata to document social science datasets: the Data Documentation Initiative (DDI). The two authors worked together on the DDI committee and are authors of a previous article on the DDI (Blank and Rasmussen 2004).

We begin by discussing the fundamental problems of social science dataset documentation. We focus on the value that standardized documentation in the form of the DDI creates for the social science community. Footnote1 To understand the virtues of the new, we look briefly at some historical documentation standards. We describe the fundamental features of the DDI standard by describing a current application. The DDI is a standard still being developed and we close with a discussion of future development plans and a summary of the value of the DDI for research.