Impact Factor, h-index and journal rankings explained

By the time you pick a Master’s topic or start a doctorate, you run into numbers like JIF 4.2, h-index 34 or Q1 journal. They show up in peer-review feedback, in departmental guidelines, in conversations with your supervisor — as if everyone knows what is meant. This article explains the metrics that matter most — Impact Factor, h-index, SJR, CiteScore — and, more importantly, how to use them in your thesis without being dazzled.

Why metrics matter in a thesis at all

Bibliometric indicators try to compress scholarly quality into a number. That only works up to a point, but they give you three genuinely useful things:

• A rough quality signal for journals you don’t know.
• Comparability inside a single field (never across fields).
• Discoverability: highly cited work leads you to the classics of your topic.

What they don’t give you: a verdict on whether a specific article is relevant to your question, whether the methodology is sound, or whether the claim you cite matches your interpretation. That, you have to read for yourself.

The Journal Impact Factor (JIF)

Clarivate’s JIF is the oldest and best-known indicator. It measures how often articles in a journal were cited on average over the last two years.

How it is calculated:

• Numerator: citations in the reporting year to articles the journal published in the two prior years.
• Denominator: number of “citable items” (original articles + reviews) published in those same two years.
• Example: JIF 2025 = citations received in 2025 to articles from 2023/24, divided by the number of those articles.

What the JIF does well:

• Compare journals inside a single field (cardiology vs. cardiology).
• Signal whether a journal is at least not entirely marginal.

What it can’t do:

• Compare fields. In molecular biology top journals sit above JIF 40; in mathematics anything above 5 is exceptional. A biology-vs-maths comparison via JIF is nonsense.
• Judge individual articles. The citation distribution inside a journal is extremely skewed — a few highly cited papers pull the mean up while many articles are never cited at all.
• Separate reviews from originals. Review-heavy journals structurally get higher JIFs.

The JIF is published each June in the Journal Citation Reports (JCR), accessed via Web of Science with a university licence. It always refers to the previous year.

SJR, CiteScore and SNIP — the alternatives

Because the JIF drew a competitor from Elsevier, several journal metrics now live side by side:

SJR (SCImago Journal Rank) weights citations by how prestigious the citing journals themselves are — a citation from Nature counts more than one from a niche journal. Free at scimagojr.com, which makes it the first choice for a quick check without a university VPN.

SNIP (Source Normalized Impact per Paper) corrects for a field’s citation culture. That makes cross-field comparisons at least approximately meaningful — a SNIP of 1.0 means average citation rate for that field, whatever the field.

CiteScore is Elsevier’s answer to the JIF, with a wider window (4 years) and without the contested “citable items” trick. Freely visible at scopus.com for every indexed journal.

For most bachelor’s or master’s theses this is enough: if your university has Web of Science, look at the JIF and the quartile rank (Q1–Q4). If not, SJR on scimagojr.com is the best free alternative.

The h-index — authors instead of journals

The h-index (Hirsch, 2005) measures researchers, not journals. It is the largest number h for which the researcher has h publications each cited at least h times.

Example: An author with 20 publications, 8 of them cited at least 8 times but none cited 9 or more, has h = 8.

Why it makes sense:

• Raw publication count rewards prolific but low-impact authors.
• Raw citation count rewards a single viral paper.
• The h-index requires both — volume and impact.

Why it stays problematic:

• Career length dominates. Anyone who has published for 30 years mechanically has more chances at a high h.
• Fields aren’t comparable. In biomedicine, h-indexes of 40+ are normal; in mathematics, h = 15 is remarkable; in the humanities a highly influential career might sit at h = 8.
• Self-citation distorts if authors cite themselves aggressively.
• Name collisions — there are hundreds of “M. Miller”s. Without an ORCID link the h-index becomes useless.

Google Scholar, Scopus and Web of Science each report different h-indexes for the same person because they index different sets of publications and citations. Google Scholar’s number is usually the highest because it includes everything — preprints, book chapters, conference volumes.

Other author-level metrics

• i10-index (Google Scholar): number of publications with at least 10 citations. A quick supplementary figure.
• m-quotient: h-index divided by career length in years. Fairer to early-career researchers.
• g-index: gives highly cited articles more weight than the h-index does.
• Altmetrics: social signals (tweets, news, policy citations). A complement, not a replacement, for citation-based measures.

What the metrics do NOT tell you

Before you mention any of these numbers in your thesis, remind yourself what they don’t say:

• Not the quality of an individual paper. A highly cited paper can be famous because it is controversial or wrong (Andrew Wakefield’s MMR paper: heavily cited, retracted).
• Not the fit to your topic. The New England Journal of Medicine has a JIF near 158, but for a medieval-art-history thesis it is irrelevant.
• Not methodological soundness. Metrics are blind to p-hacking, small samples or sloppy statistics.
• Not currency. A 1975 classic with 5,000 citations may be superseded today.

How to use these numbers in your thesis

For most bachelor’s and master’s theses a very defensive use is right:

• Journal selection before you search: confirm that your core sources come from Q1 or Q2 journals (Web of Science or SJR). That doesn’t replace substantive assessment, but it filters out predatory journals.
• Literature review: if you argue why a particular study is influential, you can cite the citation count (Scholar, Scopus, Web of Science — name the source) as one argument alongside substantive ones. “This meta-analysis (n = 1,842 citations per Web of Science, August 2026) has shaped the field since …”
• Methodological reflection: in a systematic review you justify your inclusion criteria. “Peer-reviewed, indexed in Web of Science or Scopus” is a legitimate methodological rule.
• Discussion: be careful with “the author holds an h-index of …”. That reads quickly as an argument from authority rather than an argument. In most fields it isn’t customary.

Where to look the numbers up

• Journal Citation Reports (Clarivate): JIF, quartile, field comparison. University licence needed.
• Scopus (Elsevier): CiteScore, SJR, SNIP, author h-index. University licence needed.
• SCImago Journal Rank (scimagojr.com): SJR, quartile, country ranking. Free.
• Google Scholar Profile: h-index, i10-index for individual authors. Free, but only if the person maintains a profile.
• Journal homepage: the current JIF is usually on the About page as a marketing figure. Check the year — old numbers often stay up.

Source quality is only half the job

A high Impact Factor tells you a journal is worth taking seriously. It does not tell you whether the citations you pulled from an article in that journal actually say what you claim. This is exactly where the most common thesis mistakes happen: cleanly selected journal sources, but the concrete claim is missing from the original — a wrong page reference, a misread passage, an AI-hallucinated citation.

That is what Acurio checks: you upload your thesis and the PDFs of your sources, and Acurio verifies claim by claim whether the statement actually appears in the source — with a page reference. The JCR tells you which journals to trust. Acurio tells you whether your text represents those journals correctly.

Short checklist before your next literature session

• Do I know the Q1–Q4 quartile ranks in my target discipline?
• Am I using field-normalised metrics (SNIP) whenever I compare across fields?
• When I quote h-index numbers, have I noted which database (Scholar / Scopus / WoS) and the date?
• Am I treating metrics as a signal, never as proof of relevance?
• Do my core sources sit in Q1/Q2, or have I justified why a Q3 hit is nonetheless on-topic?

Handle metrics that way and you approach them as a tool — one that answers a specific question and leaves a dozen others open — instead of a verdict.