Why wasn't impact factor designed to measure quality?
Because it was built as a library subscription tool, not a research-assessment metric — and applied to a single paper it misuses a journal-level average.
Eugene Garfield created the impact factor in the 1960s to help librarians decide which journals to stock and to select titles for his citation index — a purchasing and indexing aid, never a measure of individual research (Garfield 2006). Using it to judge a paper commits a category error: it applies a journal-wide average to one article, when citations inside a journal are extremely uneven. A small share of papers earns most of the citations (articles in the most-cited half are cited about ten times as often as the least-cited half) so the average describes almost no actual paper (Larivière et al. 2016). The metric is also shaped by technicalities like which item types fall in the denominator, which makes it gameable (Garfield 2006; Larivière & Sugimoto 2019). None of this makes impact factor useless for its original job: comparing journals as publications. It makes it the wrong instrument for the question people actually ask it: is this paper any good?
Sources
Frequently asked
Who invented the impact factor?
Eugene Garfield, in the 1960s, as a tool for librarians and for building his citation index (Garfield 2006).
Is impact factor useful for anything?
Yes, comparing journals as publications. The error is applying it to individual papers or people.
Related
Last updated