"Let’s make science metrics more scientific"
by Prof. Julia Lane
The increased push for accountability in science has led to increasing demand for science metrics. Yet the most common metrics, like the "Journal Impact Factor" or the more recent "Hirsch Index", are deeply flawed. What, however, are the alternatives in a world of science, in which quantitative measures are necessary to assess performance? Finding good answers is critical, since the metrics will profoundly influence the conduct of science.
Julia Lane will discuss the requirements of an internationally consistent rating system for scientific research. She proposes an international strategy for a model that profits scientists, publishers and research funding bodies alike. The Open Researcher and Contributor ID (ORCID) project, in addition to building on new projects, like STAR METRICS http://sites.nationalacademies.org/PGA/fdp/PGA_057189, could serve as starting points.
One core issue to developing successful metrics is that they must capture the ways in which scientific knowledge is created, disseminated and adopted, ranging beyond publications and citations to include policy advising and teaching, or even the production of blogs and Youtube videos. But probably the most important issue is to develop scientific incentive structures that engage scientists in the exercise of measurement.
Julia Lane is a former Professor of Economics at the American University, an American Statistical Association Fellow, a research Associate of IZA, the Institute for the Study of Labor, and a Program Director at the National Science Foundation.
Let's make science metrics more scientific by Julia Lane
Nature 464, 488-489 (25 March 2010)