Impact factor (2-year) · explained
Why our impact factor differs from the one journals report
Journal websites usually show Clarivate’s Journal Impact Factor™. Our Impact factor (2-year) follows the same basic idea but is calculated from open data, with stricter counting, so the two numbers can differ. This page explains why, and when to use which.
In short: both numbers divide recent citations by recent research papers. The Journal Impact Factor also counts citations to editorials, news and letters on top. We don’t, and we use different citation data. Use ours to compare journals on this site, not to check a journal’s JIF.
What journals usually quote
When a journal advertises its “impact factor”, it almost always means Clarivate’s Journal Impact Factor™ (JIF), published in Journal Citation Reports™ (JCR). Clarivate releases it each June for the previous year: the edition released in June 2026 gives JIFs for 2025.
The two-year JIF for year Y divides the citations made in Y to anything the journal published in Y−1 and Y−2 by the number of “citable items” (articles and reviews) it published in those two years. Its numerator counts citations to the journal as a whole, including ones that cannot be linked to a specific article. Since the JCR 2025 release (2024 data), citations to and from retracted items are left out of the numerator; retracted articles still count in the denominator. Clarivate also publishes a 5-year JIF, which uses the previous five years instead of two.
The JIF is built from Web of Science, a subscription citation index, and JCR is a subscription product. We cannot show JIF values on this site.
How ours is calculated
Our Impact factor (2-year) for year Y is:
citations made in Y to the journal’s articles, reviews and conference papers from Y−1 and Y−2number of those articles, reviews and conference papers
- The data come from OpenAlex, an open index of scholarly works.
- Self-citations count, as they do in the JIF.
- Citations from works hosted on repositories, such as preprints, are left out.
- The same papers are on the top and the bottom of the fraction. Nothing is counted on top that is not also counted below.
- It is not adjusted for field. To compare journals across fields, use our main score, OJR.
The four differences
What counts on top of the fraction
The JIF’s numerator counts citations to everything the journal published: research papers, but also editorials, news, letters and commentary, and citations it cannot link to a specific item. Its denominator counts only articles and reviews. Citations to front matter are added on top, with nothing added below. We count citations only to the papers we divide by.
JIF-style counting
5,000 ÷ 1,000 = 5.0
Our counting
4,000 ÷ 1,000 = 4.0
Effect on our value, other things equal: lower than a JIF-style count whenever a journal’s front matter is cited. How much depends on the journal and has not been measured across this release.
How document types are decided
Clarivate classifies items by hand, so editorials and news are kept out of the denominator. OpenAlex assigns document types automatically. Some editorials and news items end up typed as articles, so we divide by them although they are rarely cited.
4,100 ÷ 1,100 = 3.7 instead of 4.0
Effect on our value, other things equal: lower when the wrongly typed items are cited less than research papers. Typing can also differ in other ways, so the net effect for a given journal is not known.
Which citations are found
The JIF uses Web of Science; we use OpenAlex. The two indexes cover different sets of citing journals, books and conference proceedings, and link references with different methods. A citation only counts when the citing work’s reference list has been matched to the cited paper.
Effect on our value: either way. It depends on which citing works each index covers and links; the size and direction have not been established across this release.
Timing
OpenAlex may date a paper by the year it first appeared online, while the JIF usually follows the year of the print issue. A paper that appears online in December and in an issue the next January can fall into different two-year windows. The two data sets are also taken at different times, and late-arriving citations change the counts.
For citations made in 2025, both windows cover papers from 2023 and 2024 (shaded). Dated by its online year, the paper falls outside the window; dated by its issue year, it falls inside.
Effect on our value: either way. The size of the timing effect has not been measured.
Which way, and by how much
The values can differ because of citation coverage, document classification and publication dates. The size and direction of the difference have not been established across this release.
| Difference | Pushes our value |
|---|---|
| Citations to front matter counted on top (JIF only) | lower than a JIF-style count |
| Rarely cited items typed as research papers | lower |
| Different citation index and reference matching | either way |
| Online year versus issue year, and data extraction dates | either way |
Because the effects can add up or cancel out, the difference for any one journal can be large, small or in either direction. A difference from a journal’s advertised JIF does not by itself establish an error. Check the metric, indicator year and counting rules.
When to use which
Use our Impact factor (2-year)
- To compare journals in the same field on this site
- To follow a journal’s trend over the years with one consistent method
- When you need a free, open number you can check
Use the Journal Impact Factor™
- When a funder, employer or form asks for the JIF by name
- To check a figure a journal advertises
- Through Journal Citation Reports™, which needs a subscription
To compare journals across fields, use neither. Both are raw citation averages, and citation habits differ between fields. Use OJR, our field-normalized score, instead. To check whether a journal is indexed by Clarivate at all, search the free Master Journal List.
Other numbers journals quote
- CiteScore
- Elsevier’s metric from Scopus: citations over four years to articles, reviews, conference papers, data papers and book chapters published in those four years, divided by the number of those five document types.
- SJR (SCImago Journal Rank)
- Citations from Scopus weighted by the prestige of the citing journal, over a three-year window.
- SNIP (Source Normalized Impact per Paper)
- Citations per paper from Scopus, corrected for how often papers in the journal’s field cite others.
- 5-year Journal Impact Factor™
- Clarivate’s JIF with a five-year publication window instead of two.
Questions
Is your number wrong?
A difference from a journal’s advertised JIF does not by itself establish an error. Check the metric, indicator year and counting rules: ours is a different calculation from different data, applied the same way to every journal on this site, and it does not reproduce the JIF. If a value still looks wrong after that, tell us and we will check it.
Can you show the JIF next to it?
No. The Journal Impact Factor is licensed data owned by Clarivate, and we cannot republish it. Journals often show their own JIF on their websites, and Journal Citation Reports™ lists all of them for subscribers.
Why not use Clarivate’s counting rules?
Clarivate’s numerator relies on its own hand classification and on Web of Science citation links, which are not open. We count only what we can check in open data, and we count the same papers on both sides of the fraction.
Why call it an impact factor at all?
“Impact factor” is the plain name most people search for when they want citations per paper. We label ours “Impact factor (2-year)” and say that it is calculated from OpenAlex, so it is clear that it is our own calculation and not Clarivate’s Journal Impact Factor™.
Journal Impact Factor™ and Journal Citation Reports™ are trademarks of Clarivate. OpenJournalRank is not affiliated with or endorsed by Clarivate.