Impact, not impact factor.

When the English philosopher Herbert Spencer introduced the phrase “survival of the fittest” in 1864, he could not have imagined that it would summarize the plight of young scientists years later (1). As competition for coveted faculty appointments and research funding continues to intensify, today’s researchers face relentless pressure to publish in scientific journals with high impact factors. But only a few decades ago, when I began my scientific career as a virologist in the 1970s, the common outlets in my field were journals that specialized in virology. Work that straddled disciplines was often published in journals catering to a broad readership. Most researchers read and published articles with little regard to the purported impact of the journals themselves. Faculty appointments, promotions, and the award of research grants were often largely based on perceived future impact of the work, not on whether the research was published in so-called “high-impact” journals. Unfortunately, the tide has since turned. Eugene Garfield, founder of the Institute for Scientific Information, which later became part of Thomson Reuters, introduced the concept of journal impact factors. Originally conceived to help guide librarians’ decisions regarding journal subscriptions, this metric has long since been used to rank-order scientific journals. According to Garfield, “A journal’s impact factor is based on 2 elements: the numerator, which is the number of citations in the current year to items published in the previous 2 years, and the denominator, which is the number of substantive articles and reviews published in the same 2 years” (2). That calculation has inherent limitations even when used to measure a journal’s scientific impact. Determining impact by tallying up citations garnered over a two-year period can inflate a journal’s overall impact if only a small number of articles, especially in prolific but rapidly changing areas of research, account for the vast majority of citations. (Consider the recent explosion of interest in areas such as stem cell biology and genome editing, for example.) Reflecting that limitation, a 2005 editorial in Nature reported that 89% of the journal’s impact factor of 32.2 at the time could be attributed to 25% of the papers published during the relevant period (3). Moreover, as Garfield himself observed, only 0.5% of the 38 million items cited from 1900 to 2005 were cited more than 200 times; half of the articles received no citations, and a quarter were not substantive reports of original research (2). Furthermore, the common practice of ranking journals based on vanishingly small differences in impact factor—sometimes down to three decimal points—lends the metric a semblance of precision and discriminatory power. Admittedly, it is unrealistic to think that the choice of journal in which to publish is unimportant; exciting and important papers are often published in journals with high impact factors. But the above facts must give the scientific community pause regarding the usefulness of the impact factor as an indicator of the quality of individual articles published