Journal and researcher metrics drive too many decisions in higher education — promotion, hiring, ranking submissions, and funding allocation — to be treated casually. The institutions that use metrics well are not those that quote them most often, but those who know exactly what each one measures, where it misleads, and how it sits within a wider evidence base. This guide is a practical reference for research offices, VPs of research, and CIOs who need a clear, defensible position on the metrics their institution relies on.
It covers the three layers that matter: journal-level metrics (Journal Impact Factor, CiteScore, SJR, SNIP), researcher-level metrics (h-index, FWCI, and beyond), and the identifier infrastructure (Scopus Author ID, ORCID) without which none of the others can be trusted.
Key takeaways
- Journal-level metrics describe journals, not researchers. Use JIF, CiteScore, SJR, and SNIP for journal context; do not use them as proxies for individual quality.
- Researcher-level metrics depend on clean identifiers. Without ORCID and Scopus Author ID reconciliation, every h-index and FWCI is a guess.
- SNIP normalises for field; SJR weights by citing-journal prestige; CiteScore averages over a longer window than JIF. Each answers a different question.
- DORA and CoARA reframe metric use. Both signatories evaluate research on its own merits, not the journal it appeared in.
- A RIMS keeps all of this defensible. Reconciled identifiers and a complete output record make every reported metric stand up to scrutiny.
Why metrics quality starts with data quality
Every metric is only as credible as the publication record it is computed from. Missing outputs depress an h-index; misattributed authorship inflates a researcher count; duplicate records double-count citations. Before debating which metric to favour, an institution should be confident its underlying dataset is reconciled — the practical foundation we describe in Building a Single Source of Truth for Research Data. Without that, metric choice is decorative.
Journal-level metrics
Journal Impact Factor (JIF). The most cited and most contested metric in scholarly publishing. JIF is a two-year average of citations per article in a given journal. Useful as a coarse signal of a journal's citation environment; unreliable as a researcher-level proxy. Read more in Journal Impact Factor Explained: What It Measures, What It Misses.
CiteScore, SJR, and SNIP. Three Scopus-derived alternatives that adjust the JIF idea in different ways: CiteScore widens the citation window; SJR weights citations by the prestige of the citing journal; SNIP normalises for citation behaviour by field. Each is fit-for-purpose for different questions — see CiteScore, SJR and SNIP Compared: Choosing the Right Journal Metric.
Researcher-level metrics
Researcher-level metrics are even more dependent on clean data, because they aggregate across a single person's record. The widely used ones are the h-index and Field-Weighted Citation Impact (FWCI), each balancing different things. Productivity-only counts mislead; pure impact ignores volume; only a combination, on reconciled data, supports defensible decisions. We unpack this in Researcher-Level Metrics: h-index, FWCI and What They Really Tell You.
Identifier infrastructure: the prerequisite
The reason researcher-level metrics so often disappoint is name ambiguity. The same researcher may appear under several name variants; common names produce hundreds of false matches. Persistent identifiers — Scopus Author ID, ORCID — make metrics meaningful by anchoring outputs to the right person. Our companion article Scopus Author ID and ORCID Explained for Research Offices describes what each identifier is, how researchers and admins find them, and how a RIMS reconciles them automatically.
Aligning with DORA and CoARA
Increasingly, institutions are signing the San Francisco Declaration on Research Assessment (DORA) and joining the Coalition for Advancing Research Assessment (CoARA). Both commit signatories to using journal-based metrics responsibly — not as a stand-in for evaluating the research itself. A RIMS supports this shift by presenting the full evidence base (citations, collaboration, societal impact, output type diversity) rather than reducing a researcher to a single number.
What this looks like in production
Universitas Hasanuddin uses Discover RIMS in production across 2,500+ researchers, 15,300+ publications, and 18 faculties and research units. Author identity is reconciled across Scopus, OpenAlex, ORCID, Crossref, and Scimago, so the journal- and researcher-level metrics the institution reports are computed on a defensible record — not assembled from a one-off extract.
Frequently asked questions
Which metric should we use? No single metric. Use journal-level metrics for journal context, researcher-level metrics carefully and always alongside collaboration and impact evidence, and never let any one number drive a hiring or promotion decision.
Are we DORA-compliant if we still use JIF? DORA does not ban metrics; it asks institutions to evaluate research on its own merits and not use journal-based metrics as a proxy for individual quality.
Does a RIMS calculate these metrics? A RIMS surfaces the underlying data — outputs, citations, identifiers, journal context — that every metric is computed from, and presents the metrics on reconciled data so they reflect reality.
Where to start
Begin with identifiers and reconciliation. Discover RIMS unifies five global sources — Scopus, OpenAlex, ORCID, Crossref, and Scimago — and resolves author identity automatically, so every journal- and researcher-level metric your institution uses rests on data you can defend.