This project will explore the feasibility of a new approach to analyzing the content of scientific papers which may lead to new quantitative science indicators and to improved understanding and interpretation of existing indicators. The content of scientific articles will be broken into "textual fragments" (such as title, authors, institutional affiliations, acknowledgements, references, word choice, stylistic elements, and so on), and "laboratory inscriptions" such as photographs, charts, equations, and the like. For exploratory purposes analyses will be made of three contrasting groups of papers in physics and one other field of science -- co-citation clustered papers, other highly cited papers in the same fields but outside the clusters, and a sample of non-highly cited papers. If successful the research will suggest which texual elements may be useful for the development of policy-related indicators and analyses, and may then become the basis for displays and analyses in future volumes of Science & Engineering Indicators. The authors will disseminate their results through a seminar on the methods and the results to be held at NSF, and through papers to be submitted to an appropriate professional journal and to the 1988 Annual Meeting of the Society for Social Studies of Science.
