A proposed framework to address metric inflation in research publications

Since the advent of online research metrics, which began with Web of Science in 1997, these metrics have been increasingly used to rank researchers and universities. Over the last two decades, the easy access to research metrics has greatly benefitted the academic community and beyond by providing quantitative measures for ranking researchers, universities and departments. However, this accessibility, accompanied by a tendency to quantitatively evaluate research quality and impact, has also shifted the focus toward practices aimed at enhancing research metrics rather than pursuing high-quality, potentially path-breaking research. This trend threatens to degrade global research advancement and invalidate rankings.

We perform an analysis of statistics from the Stanford's top 2% list and Nobel Laureates.

We demonstrate that an accelerating number of researchers - on the order of 10% or 20,000 researchers on Stanford's Top 2% researchers - are achieving implausibly high-publication and new coauthor rates, with many producing tens to hundreds of papers per year, and gaining hundreds to thousands of new coauthors annually.

We propose a method to renormalize research metrics. Our renormalized metrics aim to remove the incentive for researchers to prioritize quantity or resort to unethical practices to boost their metrics.