Impact Factor

The impact factor is a journal level metric that measures the average number of citations received per article published in a journal over a specific period. Calculated annually by Clarivate (called Clarivate Analytics until 2020) as part of the Journal Citation Reports, it divides the number of citations received in a given year by articles published in the two preceding years by the total number of citable items published in those same two years.

Despite decades of criticism, the impact factor remains one of the most widely recognized and influential metrics in academic publishing. It shapes where researchers submit their manuscripts, how institutions evaluate faculty for hiring and promotion, and how funding bodies assess the reach of supported research. High impact factor journals attract more submissions, more attention, and more prestige, creating a self reinforcing cycle. At the same time, a growing movement led by initiatives such as the San Francisco Declaration on Research Assessment (DORA) urges the research community to evaluate research on its own merits rather than on the basis of the journal in which it appears. Understanding the impact factor, including its limitations, is essential for navigating the academic publishing landscape.

The calculation is simple in principle. For a given year (say 2025), the impact factor equals the number of citations received in 2025 by articles published in 2023 and 2024, divided by the number of citable articles published in 2023 and 2024. For example, if a journal published 300 citable articles across 2023 and 2024, and those articles received 2,100 citations in 2025, the 2025 impact factor would be 7.0. Since the 2023 release, Clarivate has reported the impact factor to one decimal place rather than three, and the 2026 release of the Journal Citation Reports, published on 17 June 2026, carries the 2025 impact factors. Since the 2025 release, citations to and from retracted items are left out of the numerator, while retracted items still count in the denominator.

Several features of this calculation have important implications. The two year citation window favors fields where citation accrues rapidly (such as biomedical science) over fields with longer citation cycles (such as mathematics or humanities). Only "citable items" (research articles and reviews) count in the denominator, so editorials, letters, and news pieces are left out of it, but citations to those items still count in the numerator, which can inflate the ratio. The distribution of citations across articles is typically highly skewed: a small number of highly cited papers can substantially raise the impact factor while most articles in the journal receive far fewer citations than the average suggests.

Alternative metrics have emerged to address these limitations. The five year impact factor extends the citation window. The Eigenfactor score weights citations by the prestige of the citing journal. Author level metrics such as the h index summarize a researcher's publications and citations, while article level metrics such as Altmetric scores and download counts provide information about individual papers rather than the journal as a whole. The CiteScore from Scopus uses a four year window and, since 2020, counts the same five peer reviewed document types (articles, reviews, conference papers, book chapters and data papers) in both numerator and denominator.

The impact factor can serve as a rough guide when researchers want a quick comparison of journals within the same discipline. It is one factor, among many, to consider when choosing where to submit a manuscript. However, it should never be used as a proxy for the quality of an individual paper, the productivity of an individual researcher, or the importance of a research program. Cross discipline comparisons based on impact factor are particularly misleading because citation norms differ dramatically between fields.

The most fundamental mistake is equating a journal's impact factor with the quality of any individual article published in it. A paper in a high impact journal may receive zero citations, while a paper in a lower ranked journal may become highly influential. Another error is comparing impact factors across disciplines without normalization; an impact factor of 3.0 is excellent in mathematics but below average in molecular biology. Researchers also sometimes chase high impact factor journals at the expense of reaching the most appropriate audience for their work, leading to mismatches between the study's focus and the journal's readership. Finally, gaming strategies such as excessive self citation and citation stacking have been documented and can result in journal suppression from the Journal Citation Reports.

A biomedical researcher has completed a study on a novel biomarker for early detection of pancreatic cancer. She considers submitting to Journal A (impact factor 12.5, broad readership, 18 month review time) and Journal B (impact factor 4.2, specialized in gastrointestinal oncology, 8 week review time). Despite Journal A's higher impact factor, she chooses Journal B because its audience of GI oncologists is more likely to apply her findings clinically, the review process involves domain experts who can provide more relevant feedback, and the faster turnaround allows her results to reach the field sooner.