On measuring scholarly influence by citations

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 theoretical foundation and are known to have counter-intuitive properties. The paper proposes a new approach that is grounded on a theoretical “influence function,” representing explicit prior beliefs about how citations reflect influence. Conditions are derived for robust qualitative comparisons of influence—conditions that can be implemented using readily-available data. Two examples are provided, one using the world’s top-10 economics department, the other using the top-10 economics journals.