CiteScore vs Impact Factor vs SJR: Which Metric Should You Actually Use?

Three numbers appear on almost every journal comparison: CiteScore, Journal Impact Factor (JIF) and SJR. They are often treated as interchangeable, and they are not. They use different numerators, different windows and different databases, which is why the same journal can look strong on one and mediocre on another. The useful question is not "which metric is best?" but "which metric answers the decision I am making?"

What each metric measures

The JIF divides by a curated count of research and review items. CiteScore historically counted all document types, which includes editorials and letters; Scopus now publishes additional filtered variants. SJR adds a PageRank-like weight, so a citation from a high-prestige journal counts more than one from a low-prestige journal.

Why the numbers differ

Four reasons explain most discrepancies:

1. Window length. A four-year CiteScore captures slower-citing fields better than a two-year JIF.
2. Document types. Counting editorials inflates the denominator and can lower the score; excluding them raises it.
3. Database coverage. Scopus and Web of Science index overlapping but different sets of journals and citations.
4. Field citation density. Molecular biology cites quickly and heavily; mathematics and humanities cite slowly and sparsely. Raw metrics are not comparable across fields.

Which metric to use for which decision

• Comparing journals within one field: any of the three works, but use the same metric for every journal and prefer the one your field reads. Medicine often uses the JIF; broader science uses CiteScore or SJR.
• Comparing across fields: use the quartile, not the raw number. Quartiles are computed within subject categories, so a Q1 mathematics journal is not weaker than a Q1 oncology journal just because its raw score is lower.
• Checking prestige weighting: SJR is useful when you want citation quality rather than raw volume.
• Funder or promotion paperwork: use the metric the form names. If it asks for the Journal Impact Factor, CiteScore will not be accepted, and vice versa.

Quartiles are category-relative

A journal can be Q1 in one subject category and Q3 in another because quartile assignment happens inside each category. That is not a bug; it reflects the journal's position among its closest peers. When a journal claims "Q1", check which category and which database. Our Q1-Q4 explanation covers the mechanics.

The mistakes to avoid

• Using one year's value as a stable fact. Metrics move; a single-year spike from one highly cited paper can mislead.
• Comparing Q1 in Scopus with Q1 in Web of Science as if they were the same. The category sets differ.
• Treating a high metric as scope fit. A Q1 journal with the wrong scope is still a desk rejection.
• Ignoring the trend. Three years of stable or rising values is more useful than one number.

Where to check each metric

The Journal Impact Factor comes from Clarivate's Journal Citation Reports, usually via your library. CiteScore and SJR are open on Scopus and SCImago. Our category pages display the metrics we hold for each journal, labelled honestly: for API-discovered journals we show "2-year mean citedness" rather than calling it an Impact Factor, because a citedness average and a JCR Impact Factor are not the same measure.

A practical rule

Pick the metric your field and your paperwork recognise, read it as a range rather than a verdict, and always check the quartile within the relevant category. Then spend your remaining time on scope fit, because that is what decides the outcome.

If you want to compare journals across both databases with the metrics side by side, start with the journal finder and filter by quartile. For the full definition of the JIF, see Impact Factor explained; for the database difference, see Scopus vs Web of Science.