Impact factor, quartiles and h-index, explained

The metrics at a glance

Before the detail, the one idea that clears up most confusion: every journal metric belongs to a database. The Journal Impact Factor only exists for journals in Clarivate’s Web of Science Core Collection. CiteScore, SJR and SNIP only exist for journals in Scopus. If a journal isn’t in the database, it cannot have that metric, whatever its website says.

Journal Impact Factor (Clarivate)

The Journal Impact Factor is published once a year, in June, in Clarivate’s Journal Citation Reports (JCR). It is calculated as:

Citations received in year Y to items the journal published in years Y−1 and Y−2, divided by the number of citable items (articles and reviews) it published in Y−1 and Y−2.

So a JIF of 3.0 means that, on average, the journal’s recent articles and reviews were each cited about three times in the year measured. It is an average, and citation distributions are highly skewed: a handful of heavily cited papers can carry a journal’s JIF while most of its papers are cited far less.

Changes worth knowing about

• Since the 2023 JCR release, Clarivate gives a JIF to all journals in the Web of Science Core Collection that meet its criteria, including the Arts & Humanities Citation Index and the Emerging Sources Citation Index (ESCI). Before that, ESCI journals had no JIF. Some older university rules still equate “has an impact factor” with “is in SCIE or SSCI”; that is no longer the same thing.
• Since the same release, JIF is shown to one decimal place rather than three, to discourage over-precise comparison.
• Clarivate also publishes the Journal Citation Indicator, a field-normalised measure, which is useful when comparing journals across disciplines.

JCR quartiles

Within each Web of Science subject category, journals are ranked by JIF and split into four equal parts: Q1 is the top 25%. A journal in two categories can be Q1 in one and Q2 in the other.

CiteScore (Scopus)

CiteScore is Elsevier’s journal metric, calculated from Scopus data. Since its 2020 methodology revision it uses a four-year window: for example, CiteScore 2024 counts citations received in 2021–2024 to documents published in 2021–2024, divided by the number of those documents.

It counts five document types (articles, reviews, conference papers, book chapters and data papers) in both the numerator and the denominator, so the calculation is consistent.

• Annual value: released once a year and then fixed.
• CiteScore Tracker: a monthly-updated running estimate for the current year. Useful for spotting a trend, but it is not the official figure.
• Percentile and rank: Scopus Sources shows where the journal ranks in each of its subject areas. When someone says “Scopus Q1”, they usually mean the top quarter by CiteScore percentile.

CiteScore is free to view on Scopus Sources. See our Scopus guide for how to find a journal there by ISSN.

SJR, SNIP and why “Q1” means different things

SJR (SCImago Journal Rank)

SJR is calculated by SCImago from Scopus data. Instead of counting every citation equally, it weights citations by the prestige of the citing journal, in a way similar to how PageRank weights links. It looks at citations in one year to documents published in the previous three years. It is free on the SCImago Journal & Country Rank website, which is where most Indian scholars meet the familiar coloured Q1–Q4 quartile badges.

SNIP (Source Normalised Impact per Paper)

SNIP is produced by the Centre for Science and Technology Studies (CWTS) at Leiden University, also from Scopus data. It corrects for the fact that some fields cite far more than others, so a citation in a field where citations are scarce counts for more. That makes SNIP more useful than raw citation averages for comparing, say, a mathematics journal with a biomedical one.

Why “Q1” is ambiguous

Because these rankings use different databases, different formulas and different category schemes, the same journal can legitimately be Q1 in SCImago and Q2 in JCR, or vice versa. If your university or funder specifies a quartile, ask which source it means and which year, and keep a dated screenshot of the evidence when you submit.

How to check a journal’s quartile in SJR and JCR, step by step

In SCImago (free)

1. 1

   Search the journal on scimagojr.com

   Type the ISSN into the home page search box. A title search works too, but the ISSN avoids look-alikes.

2. 2

   Confirm it is the right journal

   Check the ISSN, publisher and coverage years on the journal page before reading any numbers.

3. 3

   Read the subject categories

   The Subject Area and Category box lists every category SCImago ranks the journal in. Quartiles are calculated separately for each one.

4. 4

   Find the Quartiles chart

   Further down, the Quartiles chart shows each category year by year, coloured green for Q1, yellow for Q2, orange for Q3 and red for Q4. Read along your category to the year you need.

5. 5

   Take a dated screenshot

   Capture the chart with the category names visible, and note the date. Figures change when SCImago releases a new year’s data.

In Journal Citation Reports (through your library)

JCR needs a subscription, so go in through your university library. Search the journal by title or ISSN and open its profile. The profile shows each Web of Science category the journal belongs to, with its rank by Journal Impact Factor and the quartile in that category, and you can look back at earlier years. Clarivate reorganises these pages from time to time, so labels may not match a senior’s old screenshot. If your library has no JCR access, ask your supervisor or the research cell how the university wants the impact factor evidenced.

Which quartile do you report?

A journal in three categories can have three quartiles, and each can change every year. Many scholars quote the best one without saying so.

• If your university rule names a source, use that source and nothing else. VTU’s thesis checklist, for example, asks recent batches for a Q1 to Q3 Scopus journal but doesn’t say whether it means the SJR quartile or the CiteScore percentile that Scopus itself displays. In a case like that, ask the research section which one it accepts (our guide to PhD publication requirements has the detail).
• If it names a year, use that year. If it doesn’t, the latest year available when you submit is the usual choice, but ask the research cell rather than guess.
• If the journal has different quartiles in different categories, report the category closest to your paper’s subject, and name it.

Write it out in full on your CV or in your pre-submission file: “Q2, Economics and Econometrics, SJR 2024”. That line survives a sceptical DC member. “Q1 journal” on its own does not, and keep the dated screenshot either way.

The h-index: a researcher metric, not a journal metric

The h-index, proposed by the physicist Jorge Hirsch in 2005, describes an individual researcher. You have an h-index of h if h of your papers have each been cited at least h times. An h-index of 8 means eight papers with at least eight citations each.

Why your h-index differs between services

• Google Scholar counts citations from everything it can find on the web: journals, but also theses, preprints, conference papers, reports and sometimes duplicates. It is usually the highest of the three. You maintain your own profile, and Scholar shows both an all-time and a recent (last five years) version, plus the i10-index (papers with at least ten citations).
• Scopus counts only citations from documents indexed in Scopus, and builds author profiles automatically. Profiles are sometimes split or merged incorrectly, which can distort the number until corrected.
• Web of Science counts only citations within the Web of Science Core Collection, and is often the most conservative figure.

None of these is “the real” h-index. When you report one, say where it comes from and the date. For early-career researchers the h-index is also of limited value: it grows with career length and varies enormously by field.

Unofficial metrics to set aside

Any number calling itself an impact factor that is not Clarivate’s JIF is not an official metric and can be set aside. Common examples on journal websites include:

• “Global Impact Factor”, “Universal Impact Factor”, “International Impact Factor”
• “SJIF” or “Scientific Journal Impact Factor”, and “Cosmos Impact Factor”
• “Index Copernicus Value (ICV)”: a score from a commercial database, not an impact factor, and not recognised by Clarivate or Scopus
• “ISI indexed”, where ISI stands for an unrelated organisation borrowing the old name of Clarivate’s predecessor

The test is simple: a genuine JIF can be found in JCR, and a genuine CiteScore, SJR or SNIP can be found on Scopus Sources or SCImago. If it isn’t there, it isn’t an official metric. Our predatory journal guide covers the other signs worth checking alongside these numbers.

Using metrics sensibly when choosing a journal

• Compare within a field, never across fields. A JIF of 2 may be excellent in mathematics and modest in cell biology.
• Look at the trend, not one year. A journal whose metrics jumped sharply in a single year, alongside a sudden rise in the number of papers, deserves a closer look.
• Metrics describe journals, not your paper. Many papers in high-JIF journals are rarely cited; many important papers appear in modest journals. The San Francisco Declaration on Research Assessment (DORA) specifically warns against using journal-based metrics to judge individual articles or researchers.
• Fit beats rank. A paper aimed at a journal slightly above its level is likely to be desk-rejected and lose weeks. A well-matched Q2 journal that publishes your kind of work is often the faster, surer route.
• Check what your institution actually requires. UGC’s 2025 suggestive parameters for judging journals mention impact factor and CiteScore among many criteria, but it is your university that decides what counts. See our UGC-CARE guide.