Extended definition
SJR (SCImago Journal Rank) is a journal prestige indicator proposed by González-Pereira, Guerrero-Bote, and Moya-Anegón (2010, Journal of Informetrics), the SCImago research group in partnership with Elsevier (Scopus). The central methodological innovation is applying a PageRank-derived algorithm to academic citation networks: each citation does not count as a homogeneous “one citation” — its value depends on the prestige of the citing journal, calculated iteratively until convergence by eigenvector centrality. Citations from highly cited journals weigh more than citations from rarely cited journals. Journal self-citation is excluded from the calculation. SJR is released annually in January through the SCImago Journal & Country Rank platform, free of charge, based on a 3-year window. Falagas et al. (2008) offered the first systematic comparison between SJR and JIF, showing strong correlation but with significant ranking changes in specific cases.
When it applies
SJR is appropriate when the goal is to evaluate prestige (rather than gross citation volume) among Scopus journals. It is especially useful in fields where editorial manipulation of JIF via self-citation or citation rings is documented — SJR is structurally more resistant to these schemes through self-citation exclusion and prestige weighting. Useful in comparisons among journals in related fields when the researcher wants a more qualitative signal than JIF. SCImago is also the canonical source of Q1-Q4 in evaluation systems that adopt SJR-based stratification (ANVUR, some Latin American systems).
When it does not apply
It does not apply in cross-field comparisons — same limit as JIF and CiteScore. It does not apply as a metric of individual article quality. It does not replace direct reading. There is documented methodological critique (Mañana-Rodríguez, 2015): lack of precise theoretical definition of the “prestige” construct, low transparency of the PageRank damping parameter, and longitudinal ranking instability. In small fields or niches, the PageRank algorithm can produce distortions because the citation network is too sparse for stable convergence.
Applications by field
— Bibliometrics and scientometrics: SJR is often preferred over JIF in studies of field structure and prestige. — Countries with SCImago-aligned evaluation systems: SJR substitutes for or complements JIF in institutional decisions. — Surgery and medicine: systematic review (Arshad et al., 2021) recommends SJR as more robust than JIF. — Journals without JIF: SJR extends comparison to the Scopus universe, similar to CiteScore but with prestige weighting.
Common pitfalls
The first pitfall is treating SJR as an absolute quality measure — it is a proxy of relative prestige within a citation network, subject to structural biases. The second is ignoring Mañana-Rodríguez’s (2015) critique on algorithm transparency: some parameters of the PageRank applied by SCImago are not public, hampering full reproducibility. The third is confusing SJR with Eigenfactor Score — both derive from PageRank but use different corpora (Scopus vs. WoS) and weight citations differently. The fourth is assuming high correlation with JIF in all contexts: on average correlation is 0.79–0.83, but in specific quartiles disagreement can be dramatic. The fifth is interpreting annual SJR changes as real quality evolution of the journal — variations reflect in part changes in the citation network as a whole, not only in the journal evaluated.