Impact factor vs cites per paper
Impact factor vs cites per paper
The Journal Impact Factor (JIF) is a proprietary metric published by Clarivate in the Journal Citation Reports. We neither compute nor license it, so no page on this site states one. What our journal profiles show instead is an open two-year cites-per-paper rate from OpenAlex, together with the Scimago quartile and SJR. The two rates answer the same question — how often is a recent paper in this journal cited — but they are computed on different data and are not interchangeable.
What the Journal Impact Factor is
The Journal Impact Factor is published once a year by Clarivate in the Journal Citation Reports (JCR). For a given year it divides the citations received that year by the journal's papers from the previous two years by the number of 'citable items' — articles and reviews — the journal published in those two years. It exists only for journals indexed in Web of Science, and both the citation counts and the decision about what counts as citable come from Clarivate's own database.
It is a licensed product. Researchers usually see it through a library that subscribes to the JCR, or on a journal's own site when the publisher chooses to print it. JCR quartiles are cut from it within each Web of Science category.
What 'cites per paper (2y)' is
The cites-per-paper figure on our journal profiles is OpenAlex's two-year mean citedness: citations received in a year by the works the journal published in the previous two years, divided by the number of those works, computed on OpenAlex's open citation graph. It has the same shape as the JIF and asks the same question, but it is not the JIF.
The two differ because they count different things. OpenAlex and Web of Science index different sets of journals and documents, so the citing side is not the same; they classify document types differently, so the denominator is not the same; and OpenAlex counts citations from sources Web of Science does not cover, and vice versa. The result is that a journal's OpenAlex rate and its stated impact factor are usually in the same neighbourhood but rarely equal, and can differ substantially for journals with many editorials, letters or conference papers.
Why we do not show the impact factor
Three reasons, all of them about honesty rather than preference. We do not license the Journal Citation Reports, so we hold no impact-factor values to show. We could not compute one ourselves, because the definition depends on Clarivate's index and its citable-item classification, not on a formula anyone can run on open data. And a number printed under the heading 'impact factor' that was not the JIF would be a false claim on exactly the page an answer engine is most likely to quote.
So every journal profile says what it shows and where it comes from: an OpenAlex two-year rate, a Scimago quartile and SJR, and an OpenAlex h-index, with the date the venue index was built and a glossary entry for each figure. The profile's own 'impact factor' question answers with those open equivalents and says so.
How to compare journals honestly
Whichever figure you use, four rules keep the comparison fair:
- Compare like with like — an OpenAlex rate against OpenAlex rates, a JIF against JIFs. Never set a journal's OpenAlex rate beside another journal's stated impact factor.
- Compare within a field. Citation rates differ by an order of magnitude between disciplines: a rate that is unremarkable in cell biology is exceptional in mathematics or history. For a cross-field comparison use the Scimago quartile, which ranks each journal within its own category.
- Remember the window. Two-year rates reward fast-moving fields and penalise fields where citations accrue over a decade; SJR's three-year window and the cumulative h-index tell a different part of the story.
- Treat any journal-level rate as a property of the journal, not of a paper in it. Most papers in a journal are cited far less than its average, and a few far more.