On Measuring Scientific Influence

Bibliometric measures based on citations are widely used in assessing the scientific publication records of authors, institutions and journals. Yet currently favored measures lack a clear conceptual foundation and are known to have counter-intuitive properties. The authors propose a new approach that is grounded on a theoretical"influence function,"representing explicit prior beliefs about how citations reflect influence. They provide conditions for robust qualitative comparisons of influence -- conditions that can be implemented using readily-available data. An example is provided using the economics publication records of selected universities and the World Bank.