Explaining the differential outcomes of research topics: Productivity and institutional outcomes in four sciences.

摘要

The ultimate goal of this project is to generate a theory of why scientific knowledge areas expand, contract, or stay stagnant. Science is comprised of the actual research practices involved with producing knowledge and publishing articles, and it is also comprised of organizations designed to support research, such as university departments and professional societies. Every research project is directed toward a specific set of goals and utilizes a specific set of theoretical and methodological concepts. Similarly, the social organizations that support research are also directed toward a specific set of goals and concepts. These goals and concepts can be called "research topics". A research topic is any theory, method, or object of study that scientists make use of in their research. Some research topics get a lot of attention from scientists, and have a large literature associated with them. Other research topics receive hardly any attention and have very little literature associated with them. Research topics also vary in the extent to which they have social organizations devoted them. Some research topics are highly institutionalized, having university departments devoted to them and large professional societies. Other research topics are weakly institutionalized, having very little formal social organization associated with them. Variations in productivity and institutionalization among research topics are explained. The methodological approach includes case studies of four research topics--anaerobic bacteriology, aeronautics, forensic psychology, and clinical biochemistry--each with a different set of values on productivity and institutionalization, which I conceptualize as high and low. Data come from bibliometric indicators derived from over 8,500 scientific articles and qualitative interviews with 52 scientists active in the four research topics (14 in-depth interviews and 38 email exchanges).