How prominent is a researcher? How influential is their body of work? Evaluating a researcher’s cumulative impact is an especially challenging task, and it can be tempting to turn instead to author-level metrics such as h-index: the largest number h for which the author has published h articles that have each been cited h or more times.

In more concrete terms, an author has an h-index of 3 if they’ve published 3 articles that have each been cited 3 or more times, regardless of how many articles they’ve published overall. And an author has an h-index of 35 if they’ve published 35 articles that have each been cited 35 or more times, again regardless of how many articles they’ve published overall.

Like other citation-based metrics, h-index is reductive and fails to account for the ways in which disciplinary differences (e.g., co-authorship norms and citational practices) can drastically affect an author’s total number of publications and the number of citations those publications attract. Also, an author can manipulate their own h-index via gratuitous self-citation. Further, the tools that track citations don’t have extensive data regarding citations to and from books, so h-index is ill-suited to authors who primarily write books, or whose work is primarily cited in books.

Additionally, with each successive increment in h-index, achieving the next increment becomes harder and harder to do. Think about what it takes to move from an h-index of 3 (i.e., 3 publications that each have 3+ citations) to an h-index of 4 (i.e., 4 publications that each have 4+ citations), and then think about what it takes to step up to 5 and 6 and 7 and 8 and so on. This ever-increasing bar eventually makes consistent, strong performance (which is typically considered a good thing!) look like stagnation.

Image credit: H-index by souvenirsofcanada is used under a Creative Commons Attribution-NonCommercial-ShareAlike 2.0 Generic license (CC BY-NC-SA 2.0)