In defense of quantitative metrics in researcher assessments

Qualitative assessments of researchers are resource-intensive, untenable in nonmeritocratic settings, and error-prone.Although often derided, quantitative metrics could help improve research practices if they are rigorous, field-adjusted, and centralized.Scientists are continuously assessed for hiring, promotion, funding, and recognitions.A welcome movement for incentive reforms is aiming to align assessments with good scientific practice (e.g., open science, reproducibility, and diverse types of social impact).Measuring such progressive contributions and impact requires qualitative assessments that capture a wide range of achievements, beyond what traditional bibliometric quantitative indicators capture.Meanwhile, several influential international proposals for reforming research assessment have lambasted flawed quantitative metrics.For example, the San Francisco Declaration on Research Assessment (DORA) justifiably urges the abandoning of Journal Impact Factors (JIFs), and the thoughtful Leiden manifesto [1] offers skeptical advice against bibliometrics acquiring inordinate influence.Quantitative metrics undoubtedly have limitations; however, their uncritical dismissal may aggravate injustices and inequities, especially in nonmeritocratic environments.We would argue that centralized, quantitative resources can serve as a public good at little or no marginal cost in diverse situations and settings.Their judicious use may even reduce manipulative practices in scientific publishing and lead to fairer allocation of credit.In institutions with sufficient resources, the most important assessment for recruitment or for tenure is typically made by numerous highly respected assessors, including those external to an institution.Under ideal conditions, established, knowledgeable, responsible, and accountable peers will carefully review a candidate's research production and wider impact.Faculty and researchers selected through such a process may excel in the broadest possible sense.Yet even so, one cannot always be sure that those chosen are the very best among the dozens or even hundreds of competing applicants.No assessment can perfectly rank very strong alternative candidates.Local needs and preferences, and strong elements of subjectivity, may decide who is recruited among several outstanding options.The same applies to very selective, competitive awards and recognitions.