CiteScore (Elsevier/Scopus) and Impact Factor (Clarivate/Web of Science) are the two most-cited journal metrics — and they answer slightly different questions. This 2026 guide breaks down the methodology, shows worked examples for five real journals, covers the newer Journal Citation Indicator (JCI), and gives a decision flowchart for which metric matters in your specific situation.
The Big Picture: Two Different Citation Worlds
The split between CiteScore and Impact Factor reflects a deeper split in academic publishing: Scopus (Elsevier) and Web of Science (Clarivate) are two competing citation databases with overlapping but non-identical coverage. Each maintains its own headline metric.
- Scopus covers ~28,000+ journals (2026). Free to view metric values.
- Web of Science covers ~22,000 journals across SCIE, SSCI, AHCI, and ESCI. Full JCR access requires subscription.
Both citation universes are reasonably comprehensive but neither is the union of the other. A journal can be in Scopus but not WoS, or vice versa — meaning it has a CiteScore but no IF, or an IF but no CiteScore.
Impact Factor: Methodology
The Journal Impact Factor (JIF) is calculated annually by Clarivate and published in the JCR each June. The formula uses a 2-year citation window:
IF = Citations in year X to articles published in X-1 and X-2 ÷ Citable items published in X-1 and X-2
Key properties:
- Covers ~22,000 journals across SCIE / SSCI / AHCI / ESCI (as of 2026, post-expansion).
- 2-year window — advantages fast-citation cultures (biomedicine) over slow-citation fields (mathematics, humanities).
- Counts only "citable items" (research articles + reviews) in the denominator, but all citations in the numerator. This asymmetry can inflate the IF for review-heavy journals.
- Requires institutional JCR subscription to access the official value.
- Updated annually; new IFs in June.
CiteScore: Methodology
CiteScore is calculated by Elsevier using Scopus data and published openly. The formula uses a 4-year citation window:
CiteScore = Citations in year X to documents published in X-1, X-2, X-3, X-4 ÷ All documents published in X-1, X-2, X-3, X-4
Key properties:
- Covers ~28,000+ Scopus-indexed journals — broader than JCR.
- 4-year window — more stable, less skewed by a single exceptional year.
- Counts all document types in both numerator and denominator (including editorials, letters, conference papers) — generally produces lower values than IF for the same journal.
- Free to access at scopus.com/sources.
- Updated annually; new CiteScores typically in May/June.
Direct Comparison Table
| Feature | Impact Factor (JIF) | CiteScore |
|---|---|---|
| Publisher | Clarivate (JCR) | Elsevier (Scopus) |
| Citation window | 2 years | 4 years |
| Journal coverage | ~22,000 | ~28,000+ |
| Denominator | Citable items only (articles + reviews) | All documents |
| Open access | No (subscription required) | Yes (free) |
| Self-citation protection | Limited; periodic suppression | Stronger; explicit exclusion thresholds |
| Typical value | Higher (narrower denominator) | Lower (broader denominator) |
| Update cycle | Annual (June) | Annual (May–June) |
| Used by | Most tenure committees globally | Many committees; default for non-JCR journals |
Worked Examples: 5 Real Journals Across Both Metrics
Approximate 2025 metric values (from JCR June 2025 and Scopus May 2025). The numbers shift annually — use these as illustrative not authoritative.
| Journal | IF | CiteScore | SJR | CiteScore / IF ratio |
|---|---|---|---|---|
| Nature | ~50 | ~91 | ~17.0 | ~1.8 |
| The Lancet | ~99 | ~105 | ~22.0 | ~1.06 |
| Cell | ~45 | ~75 | ~21.0 | ~1.67 |
| IEEE TPAMI | ~24 | ~46 | ~6.6 | ~1.92 |
| PLOS ONE | ~3.4 | ~5.5 | ~0.86 | ~1.62 |
Notice the CiteScore/IF ratio is not constant — it varies between roughly 1.0 and 2.0 depending on journal mix and citation behaviour. Older guides claim "CiteScore ≈ 0.85 × IF". That heuristic is outdated for elite multi-disciplinary journals; CiteScore is often higher than IF for them because the 4-year window captures more accumulated citations.
The Other Important Metrics
Journal Citation Indicator (JCI)
JCI is Clarivate's newer field-normalized metric, introduced in 2021 and now reported alongside the traditional IF in JCR. A JCI of 1.0 means the journal's citation impact matches the global average in that subject category. JCI > 1.0 is above average; JCI < 1.0 is below. JCI is much fairer for cross-field comparison than raw IF.
SJR (SCImago Journal Rank)
SJR is the "PageRank for journals" — a citation from a high-prestige journal counts more than one from a low-prestige journal. Published free at scimagojr.com. Better measure of prestige-weighted influence than raw citation volume. Also drives the popular SJR quartile (Q1–Q4) system widely used in tenure cases. See our Q1/Q2 finding guide.
SNIP (Source Normalised Impact per Paper)
SNIP adjusts for differences in citation behaviour between fields. A SNIP of 1.0 means the journal's citation impact matches the Scopus-wide field average. SNIP is widely considered the fairest metric for cross-disciplinary journal comparisons. Published free by CWTS Leiden.
Eigenfactor Score
Eigenfactor (eigenfactor.org) estimates a journal's overall influence by analysing the full citation network, weighting citations by the citing journal's prestige (similar to SJR but using Web of Science data). Removes self-citations. Free.
h5-index (Google Scholar)
The h-index applied to a journal's articles published in the last 5 years. Calculated by Google Scholar Metrics. Free. Useful for newer journals not yet in JCR or Scopus, and for fields where Scholar coverage is broader (CS conferences, humanities).
Which Metric Should You Use? Decision Table
| Your situation | Best metric(s) |
|---|---|
| Tenure/promotion case at most universities | Impact Factor (JIF) + quartile |
| Journal not in JCR (Scopus only) | CiteScore + SJR |
| Comparing journals across different fields | SNIP or JCI (field-normalized) |
| Prestige / influence rather than raw citation volume | SJR or Eigenfactor |
| Newer journal with no JCR record yet | h5-index + CiteScore |
| Computer science (where conferences dominate) | CORE conference rank + h5-index + CiteScore |
| Business / management research | ABDC list + FT-50 + JCR Quartile |
| Quick free check before submission | CiteScore + SJR (both free) |
| Humanities / arts | SJR + ERIH PLUS + peer reputation (JIF coverage thin) |
Self-Citation and Gaming: How the Metrics Differ
All citation-based metrics can be gamed. The protections differ.
- Impact Factor — Clarivate periodically suppresses ~40–60 journals per year from JCR for citation manipulation (self-citation, citation stacking, citation cartels). Suppression is reactive, not preventive.
- CiteScore — Elsevier applies explicit self-citation thresholds before publication. Journals exceeding the threshold are flagged but not automatically removed.
- SJR — Prestige-weighted by design, so self-citations from the same journal contribute less. Harder to game with mass self-citation.
- SNIP — Field-normalized, so gaming requires manipulating both citing and cited papers within a field — significantly harder.
- JCI — Field-normalized like SNIP, with similar gaming resistance.
For a deeper look at journal-quality red flags including IF gaming, see our predatory journals guide.
What Tenure Committees Actually Use in 2026
- North America (US, Canada): JIF + Q1/Q2 quartile is dominant in STEM. Business schools weight ABDC + FT-50 + UTD-24. Humanities lean on peer reputation.
- UK / Europe: JIF + REF unit-of-assessment ratings. Some Scandinavian universities now formally adopt DORA-aligned policies de-emphasizing JIF.
- China: JIF threshold + journal whitelists at institutional level. Recent policy shifts (2020) urge moving away from JIF-only evaluation.
- India: UGC CARE list compliance + Scopus indexing + JIF threshold. ABDC mandatory for many business schools.
- Australia / NZ: ERA rankings + ABDC + Scopus quartile widely used.
- Middle East / SE Asia / Latin America: Strict Q1/Q2 Scopus requirements for promotion; JIF often supplementary.
If you do not know your institution's exact rule, ask. Generic "high IF" advice is often wrong locally.
Frequently Asked Questions
Why is a journal's CiteScore much lower than its Impact Factor?
Primarily because CiteScore uses a 4-year window and counts all document types (editorials, letters, conference papers) in the denominator. These attract fewer citations and drag the average down. For most journals, CiteScore ≈ 0.75–0.85× the Impact Factor, though the ratio varies.
Can a journal have CiteScore but no Impact Factor?
Yes. Many Scopus-indexed journals not in Web of Science have a CiteScore but no JCR Impact Factor. This is common for newer journals, regional publications, and humanities titles where WoS coverage is thinner than Scopus.
Which is better for cross-field comparison: Impact Factor, CiteScore, SJR, or SNIP?
SNIP is best for cross-field comparison because it normalizes for citation density differences between disciplines. A SNIP of 1.0 means citation impact matches the average across all Scopus journals in that field. Raw IF and CiteScore are not field-normalized.
Is SJR more reliable than Impact Factor?
Reliable in a different sense. SJR weights citations by the prestige of the citing journal (like PageRank), making it harder to game via low-quality self-citation. IF treats all citations equally. For prestige signaling, SJR is arguably more robust; for raw citation volume, IF is the more direct measure.
What is the Journal Citation Indicator (JCI)?
JCI is Clarivate's newer field-normalized metric, released alongside the traditional Impact Factor. JCI of 1.0 means the journal's citation impact matches the global average in that subject category. Use JCI for cross-field comparison; use IF for within-field comparison.
Is CiteScore free to access?
Yes. CiteScore is published openly on the Scopus Sources page at scopus.com/sources without a subscription. You can also access SJR and SNIP free at scimagojr.com. Only the full JCR (and the official Impact Factor) requires institutional access.
Which metric do most tenure committees use?
Most tenure committees in North America, Europe, and East Asia primarily use Journal Impact Factor (JCR) supplemented by quartile (Q1/Q2). Business schools often add ABDC or FT-50 lists. Always check your institution's specific guidance — the right metric depends on the committee, not the field.
Can CiteScore and Impact Factor disagree about journal quality?
Yes. A journal can have a Q1 SJR and Q2 JCR, or high CiteScore and modest IF, because the windows, document types, and coverage differ. Treat them as complementary signals. Major divergence often points to journal-mix differences (review-heavy vs original research) rather than quality.
Should I use h5-index instead of Impact Factor?
Use h5-index as a complement, not a replacement. It measures recent impact (last 5 years) and is free via Google Scholar Metrics, making it useful for newer journals not yet in JCR. But it covers fewer journals systematically than IF and is harder to verify formally for tenure cases.
Which metric should I use when none of my target journals are in JCR?
Use CiteScore + SJR + SNIP together. CiteScore gives raw citation volume, SJR gives prestige-weighted influence, SNIP gives field-normalized comparison. The JournalsHub Impact Factor Checker shows all three side by side for any ISSN.
Conclusion
CiteScore and Impact Factor are not rivals — they answer different questions. For formal evaluation at most institutions, the Impact Factor remains the dominant signal. For broader journal research, especially journals not in JCR, use CiteScore + SJR + SNIP in combination. For cross-field comparison, prefer field-normalized metrics like JCI or SNIP over raw IF. The JournalsHub Impact Factor Checker displays all five metrics side by side for any ISSN, so you can see the whole picture in one lookup.