Fact-checked by Grok 2 weeks ago

CiteScore

CiteScore is a free bibliometric indicator developed and released annually by Elsevier since 2016 to assess the citation impact of peer-reviewed serial titles, including journals, book series, and conference proceedings, using data from the Scopus database. It calculates the average citations per document as the total citations received in a given year by documents published in that year and the prior three years, divided by the total number of such documents (including articles, reviews, conference papers, book chapters, and data papers). Unlike the Journal Impact Factor, which relies on Web of Science data and a two-year citation window limited to citable items, CiteScore employs a four-year window, includes all document types, and covers approximately 28,000 titles as of 2024, providing broader applicability especially for newer or interdisciplinary journals. While it promotes transparency by freely releasing metrics and percentile ranks without embargo, CiteScore has drawn scrutiny for methodological differences that may inflate scores for journals with high non-research content and for Scopus's overrepresentation of Elsevier titles, potentially introducing ownership bias in rankings. These factors have positioned it as a competitive yet contested tool in academic evaluation, often used alongside other metrics to mitigate limitations in assessing journal prestige or research quality.

Overview and Purpose

Definition and Core Metrics

CiteScore is a bibliometric indicator developed by Elsevier to assess the citation impact of serial publications, including journals, book series, and conference proceedings, using data from the Scopus abstract and citation database. It represents the average number of citations received per document published by a serial, calculated over a multi-year window to provide a robust measure of scholarly influence. The primary core metric is the CiteScore value itself, derived annually from Scopus-indexed content and reflecting citations to a broad set of document types such as articles, reviews, conference papers, book chapters, and data papers. This inclusion of diverse formats—unlike narrower metrics limited to certain peer-reviewed items—aims to capture a fuller spectrum of academic output. Accompanying core metrics include CiteScore percentiles, which position a serial relative to others in its subject category (ranging from 0 for the lowest to 100 for the highest), and CiteScore ranks, enabling normalized evaluations across varying field sizes and citation practices. Recent iterations, such as CiteScore 2024, incorporate a longer citation accumulation window of up to four years for enhanced comprehensiveness and stability in impact assessment, while maintaining transparency through publicly detailed methodology and free access via Scopus sources.

Scope of Coverage in Scopus

Scopus, an abstract and citation database owned by Elsevier, forms the basis for CiteScore by indexing peer-reviewed serial publications such as journals, book series, and conference proceedings series, enabling comprehensive citation tracking across disciplines. As of 2024, it covers more than 28,000 active titles eligible for CiteScore metrics, selected through a rigorous process by the independent Scopus Content Selection and Advisory Board (CSAB) based on criteria including editorial quality, peer review standards, and citation potential. The database spans four main domains—Physical Sciences, Health Sciences, Life Sciences, and Social Sciences & Humanities—encompassing 27 subject areas with broad multidisciplinary representation, though coverage is deeper in sciences and social sciences than in arts and humanities, where approximately 3,500 titles are included. Content originates from over 5,000 publishers worldwide, prioritizing serials with ISSNs while incorporating non-serial conference papers published in dedicated proceedings series. For CiteScore purposes, citable documents in the denominator consist solely of peer-reviewed items—articles, reviews, conference papers, data papers, and book chapters—published within the four-year window, excluding non-research outputs like editorials or letters despite their indexing in Scopus. Citations in the numerator are drawn from all document types indexed in Scopus, providing a fuller picture of impact compared to metrics restricting sources. This scope ensures CiteScore reflects global scholarly output but may underrepresent fields with limited Scopus indexing, such as certain humanities subdisciplines.

Historical Development

Origins and Launch in 2016

Elsevier developed CiteScore as a response to demands from the academic community for more transparent and broadly applicable journal metrics, leveraging the Scopus database's extensive coverage of over 22,000 peer-reviewed serials spanning multiple disciplines. The initiative stemmed from critiques of proprietary metrics like the Journal Impact Factor (JIF), which were seen as opaque, delayed in release, and limited to fewer journals indexed in Web of Science, prompting Elsevier to create a freely accessible alternative after several years of internal development focused on enhancing comprehensiveness and timeliness. On December 8, 2016, Elsevier officially launched CiteScore alongside a suite of supporting indicators, including percentiles and subject averages, calculated using Scopus data to provide average citations per document over a four-year window. The initial release covered citations received in 2015 by documents published from 2012 to 2015, divided by the number of such documents, enabling comparisons across a wider array of publication types than the JIF, which primarily emphasized citable items like articles and reviews. This broader inclusion aimed to reflect diverse scholarly output more accurately, with metrics updated annually and made available without subscription fees via Scopus previews. The launch positioned CiteScore as a direct competitor to Clarivate's JIF, sparking debate over potential biases favoring Elsevier-published journals due to Scopus's indexing practices, though proponents highlighted its transparency in methodology and data sourcing as advantages over the JIF's black-box calculations. Early adoption was facilitated by integration with Scopus tools, allowing users to track journal performance without relying on delayed or paywalled alternatives.

Key Updates and Formula Evolutions

In 2016, CiteScore was initially calculated using a three-year window, where the numerator represented citations received in the calculation year to documents published in the three preceding years, divided by the total number of documents (including non-peer-reviewed items such as editorials and letters) published in those three years. This approach mirrored aspects of the Journal Impact Factor but drew from Scopus data, encompassing a broader range of document types without initial restrictions to peer-reviewed content. A significant revision occurred in June 2020 with the release of 2019 CiteScore values, prompted by expert and user feedback to enhance stability and fairness across disciplines. The updated methodology expanded the window to four years, shifted the numerator to cumulative citations received from the publication year through the end of the four-year period (rather than solely the final year's citations), and restricted the denominator to peer-reviewed document types—articles, reviews, conference papers, and book chapters—excluding non-peer-reviewed items. These changes were applied retrospectively to CiteScore values from 2011 to 2018 for consistency. The 2020 updates aimed to reduce volatility from single-year citation fluctuations and better accommodate fields with slower citation accrual, while maintaining transparency through Scopus's open data access. No substantive formula alterations have been announced since, though annual releases incorporate refined percentile calculations for comparative ranking. Empirical analyses post-update indicate improved correlation with prior metrics in many categories, though effects vary by discipline, with top-ranked journals showing minimal shifts relative to Journal Impact Factor rankings.

Calculation and Methodology

Current CiteScore Formula

The current CiteScore for year y is defined as the total citations received in year y by documents published in the preceding three years (y-1, y-2, y-3), divided by the total number of such documents published in those three years. This yields an average citation rate per document over a four-year evaluation window, where citations are counted only from the final year to ensure recency while allowing time for accrual. Documents eligible for the denominator encompass peer-reviewed content types indexed in Scopus, including articles, reviews, conference papers, book chapters, and data papers, but excluding non-citable items like editorials or letters. Citations in the numerator are drawn exclusively from Scopus-covered sources, reflecting the database's broad multidisciplinary scope of over 28,000 active titles as of the 2024 release. This formula, unchanged since CiteScore's inception in 2016, prioritizes transparency by relying on publicly verifiable Scopus data, enabling independent replication without proprietary algorithms. For example, the 2017 CiteScore for a journal might compute as \frac{114639}{7860} = 14.59, representing citations in 2017 to its 2014–2016 publications divided by those publications' count.

Inclusion of Document Types and Citation Window

CiteScore's calculation incorporates only peer-reviewed document types in both the numerator and denominator, specifically including articles, reviews, conference papers, book chapters, and data papers. This restriction, implemented in the 2024 methodology update, excludes non-peer-reviewed items such as editorials, letters, and notes that were previously counted, aiming to focus more accurately on research-based contributions rather than ancillary publications. The citation window spans four years, capturing cumulative citations received by these peer-reviewed documents over that period and dividing by the total number of such documents published in the same four years. For instance, CiteScore 2024 aggregates citations from 2021 to 2024 to documents published in 2021–2024. This extended window, lengthened from the prior one-year citation capture to prior three-year publications, enhances stability by accounting for delayed citations common in certain fields, though it may inflate scores for journals with rapid publication cycles. Prior to 2024, the metric used a narrower window—citations in a single year to documents from the preceding three years—divided by those three years' documents, while still broadening document inclusion beyond traditional citable items. This evolution addresses criticisms of undercounting citations in slower-citing disciplines but introduces potential biases toward journals with more immediate impact.

Supporting Metrics like Percentiles

CiteScore Percentile ranks a journal's performance relative to others in the same subject category, providing a normalized measure from 0 to 100 where higher values indicate superior standing. A value of 96, for instance, means the journal outperforms 96% of peers in its category by CiteScore. This metric addresses disparities in citation norms across fields by enabling direct comparisons within discipline-specific groupings defined by Elsevier's All Science Journal Classification (ASJC) system. The percentile is derived by ordering journals in a category by their CiteScore values and calculating the relative position, excluding the possibility of a perfect 100th percentile to reflect competitive rankings. It complements the raw CiteScore by offering context for absolute scores, which can vary widely; for example, a CiteScore of 5.0 might place a journal in the 90th percentile in a low-citation field like humanities but only the 50th in high-citation fields like physics. Related supporting metrics include CiteScore Quartiles, which segment journals into four equal groups based on percentiles: Q1 (top 25%, 75-100 percentile), Q2 (50-75), Q3 (25-50), and Q4 (0-25). These quartiles simplify prestige assessment, with Q1 journals deemed highest impact within categories. CiteScore Rank provides the absolute ordinal position among category peers, further aiding precise evaluations. Together, these metrics support informed decisions in journal selection, funding allocation, and performance tracking by emphasizing relative rather than absolute impact, though they inherit Scopus's coverage limitations in underrepresented regions or languages. Empirical analyses confirm percentiles enhance comparability, reducing misinterpretations from unnormalized scores.

Comparisons with Alternative Metrics

Differences from Journal Impact Factor

CiteScore employs a four-year citation window, aggregating citations received by a journal's documents published from year y-3 to y and dividing by the number of those documents, whereas the Journal Impact Factor (JIF) uses a two-year window, calculating the average citations in year y to citable items (research articles and reviews) published in y-1 and y-2, divided by the number of such items. This longer window in CiteScore aims to capture more stable citation patterns, particularly beneficial for fields with delayed citation accrual, though it may inflate values for journals with rapid publication growth. In terms of document inclusion, CiteScore encompasses all content types indexed in Scopus, such as articles, reviews, conference papers, editorials, and letters, providing a broader representation of a journal's output. JIF, however, restricts the denominator to "citable" items—typically peer-reviewed articles and reviews—excluding other formats, which can lead to discrepancies for journals heavy in non-article content like proceedings. Scopus's wider database coverage, indexing over 25,000 journals as of 2023 compared to the roughly 12,000 in Journal Citation Reports, enables CiteScore to evaluate thousands more titles without an Impact Factor, including those in emerging or non-English-language fields. Transparency differs markedly: CiteScore's methodology, raw citation counts, and self-citation rates are publicly accessible via Scopus sources without subscription barriers, updated annually in June with monthly previews. JIF data, historically less open due to reliance on proprietary Web of Science processing, only became partially transparent in 2023 with the release of detailed citation distributions, though full reproducibility remains limited. Both metrics include self-citations without exclusion, but CiteScore reports them explicitly for user adjustment, potentially mitigating manipulation risks more directly than JIF's separate disclosure. Empirical correlations between the two metrics are high (often r > 0.9 across disciplines), yet rank-order differences persist, with CiteScore tending higher for Elsevier titles due to Scopus's indexing preferences.

Relations to SJR and SNIP

CiteScore, SJR (SCImago Journal Rank), and SNIP (Source Normalized Impact per Paper) constitute a trio of journal-level metrics computed from Scopus citation data, offering distinct yet interrelated perspectives on scholarly influence. All three are updated annually and accessible via the Scopus Sources portal, facilitating their joint use in evaluating journals across disciplines. CiteScore serves as a baseline by averaging citations per document over a four-year window (citations in year y to documents published in y-3 through y), without normalization or weighting, thus reflecting raw citation volume relative to output. In relation to SJR, CiteScore lacks the prestige-based adjustments that SJR applies through an iterative process weighting citations by the scientific standing of the citing journal, analogous to PageRank, which prioritizes influential sources and mitigates self-citation effects more effectively. SNIP builds on a similar raw citation ratio but divides it by the aggregate citation rate within the journal's subject category, correcting for field-specific citation norms—such as higher rates in biomedicine versus mathematics—to enable cross-disciplinary equity absent in CiteScore. These refinements in SJR and SNIP often yield rankings that align closely with CiteScore within homogeneous fields but diverge in heterogeneous ones, where unadjusted CiteScore may overstate impact in low-citation domains. Empirical analyses reveal high correlations among the metrics, underscoring their shared reliance on Scopus data; for example, Spearman rank correlations between CiteScore and SJR frequently surpass 0.90 across journal samples, though SNIP's normalization can temper extremes in citation-heavy fields. Such concordance supports using them complementarily: CiteScore for straightforward productivity gauging, SJR for prestige hierarchies, and SNIP for balanced interfield assessments, as recommended in Scopus documentation. Divergences, however, highlight CiteScore's sensitivity to document volume and field citation practices without the causal adjustments of SJR or SNIP.

Strengths and Empirical Advantages

Transparency and Data Accessibility

CiteScore's methodology emphasizes transparency through the public disclosure of its formula and calculation parameters by Elsevier. The metric divides the total citations received by a journal's documents in a given year (plus the three preceding years) by the total number of documents published in those four years, with all inputs drawn exclusively from Scopus-indexed content. This explicit approach contrasts with the Journal Impact Factor's relative opacity, where Clarivate does not release granular data on document categorization or citation counting, limiting external reproducibility. Underlying Scopus data supports this transparency, as the database provides detailed, queryable records of publications and citations, enabling users with access to reconstruct CiteScore values for verification. While full Scopus access requires institutional subscription, Elsevier publishes aggregated inputs—such as annual citation and publication counts—for each journal during CiteScore releases, facilitating partial independent checks without proprietary barriers. Annual updates, like the CiteScore 2024 release on June 9, 2025, include methodological notes on coverage and exclusions, further enhancing reproducibility. Data accessibility is bolstered by free public availability of CiteScore metrics, including core values, percentiles, and subject-area rankings, via the Scopus Sources platform and CiteScore Hub. No subscription is needed to view these, allowing global researchers to compare journals without cost, unlike some proprietary metrics requiring paid reports. This open dissemination, initiated with CiteScore's 2016 launch, has enabled widespread adoption in evaluations while mitigating gatekeeping concerns inherent in less accessible alternatives.

Broader Applicability and Inclusivity

CiteScore's applicability extends beyond the selective indexing of the Journal Impact Factor (JIF), as it leverages Scopus data covering over 37,000 active journals, book series, and conference proceedings, compared to the approximately 28,000 titles in Web of Science's Science Citation Index Expanded and Social Sciences Citation Index combined. This broader database inclusion facilitates evaluation of a wider array of serials, including those in niche or emerging disciplines where Web of Science coverage is limited. For instance, Scopus indexes a higher proportion of journals from non-Western countries and in non-English languages, enhancing geographic and linguistic inclusivity relative to JIF's more Anglo-centric focus. In terms of document types, CiteScore incorporates all Scopus-indexed publications—such as articles, reviews, conference papers, and editorials—in both numerator and denominator, avoiding the JIF's restriction to citable items like research articles and reviews only. This approach promotes inclusivity by reflecting the full spectrum of scholarly output, particularly benefiting fields like engineering and computer science where conference proceedings constitute significant contributions. Empirical analyses across six subject fields indicate that CiteScore effectively captures relative citation impact in diverse domains, supporting its use in cross-disciplinary comparisons. Additionally, CiteScore's free public availability and annual updates enable broader accessibility for researchers and institutions without subscription barriers, unlike the proprietary JIF. It also accommodates open access journals, with studies showing a positive correlation between open access status and CiteScore values, though effects vary by discipline, thus incentivizing inclusive publishing models without inherent bias against them. Overall, these features position CiteScore as a more inclusive metric for global academic evaluation, though its reliance on Scopus selection criteria still excludes some low-impact or regionally focused outlets.

Criticisms, Limitations, and Controversies

Methodological Flaws and Biases

One primary methodological flaw in CiteScore arises from its inclusion of all document types—such as editorials, letters, news items, and other front matter—in the denominator of the formula, without restricting the numerator to citations primarily accruing to research articles. This approach dilutes the metric for journals that publish substantial non-research content, as these items typically receive fewer citations, thereby lowering the overall average. For instance, high-impact journals like those from Nature experience a 25-40% reduction in their CiteScore relative to what a more selective denominator might yield, contrasting with the Journal Impact Factor (JIF), which excludes non-citable items from the denominator while still counting citations to them in the numerator. Another issue stems from inconsistencies in Scopus's subject area classifications, which underpin CiteScore's categorical rankings and percentiles. Journals may be assigned to mismatched categories, leading to anomalous placements; for example, the Annual Review of Plant Biology has been ranked highly in "General Medicine" despite its specialization. Such errors introduce bias in comparative assessments within fields, as CiteScore's broader journal coverage (over 22,000 titles) amplifies discrepancies compared to JIF's narrower scope of around 11,000 journals with more rigid categorization. Additionally, CiteScore lacks field normalization, rendering it unsuitable for cross-disciplinary comparisons, as citation practices vary widely by subject—e.g., mathematics articles accrue citations slowly over decades, while biomedicine sees rapid bursts. This non-normalized structure perpetuates biases favoring high-citation fields, similar to JIF, without adjustments like those in metrics such as SNIP. Empirical analyses further reveal a systemic advantage for Elsevier-owned journals, with average CiteScores 25% higher than expected relative to JIF, raising concerns over inherent publisher bias in Scopus indexing and metric computation.

Susceptibility to Manipulation and Self-Citation

CiteScore, like other average citation-based metrics, is vulnerable to manipulation through strategic increases in self-citations, where authors or editors encourage references to prior publications within the same journal to inflate the numerator of citations received. This practice exploits the metric's formula, which divides total citations (including journal self-citations) by the number of documents published in a four-year window, without routinely excluding self-cites in standard calculations. Excessive self-citation can artificially elevate scores, particularly in smaller or specialized journals where citation pools are limited, distorting perceived journal impact and incentivizing non-merit-based citation behavior. In response to such vulnerabilities, Scopus, the database underlying CiteScore, monitors journal self-citation rates and suppresses metric publication for outlets exhibiting anomalous patterns. For instance, on June 29, 2020, Scopus withheld CiteScores for 48 journals after identifying self-citations comprising at least 25% of their total references, with one journal reaching 71% self-citation; this action aimed to deter manipulation while preserving metric integrity for compliant journals. Despite these safeguards, critics argue that thresholds for suppression are reactive rather than preventive, allowing moderate self-citation inflation to persist and bias rankings, as self-cites remain fully weighted in unsuppressed CiteScores. Broader manipulation tactics, such as citation rings or coerced self-referencing by reviewers, further expose CiteScore to gaming, though Scopus's policies mitigate extreme cases by excluding suspect citations from counts when detected. Proposals for reform include modified CiteScore variants that discount self-citations proportionally—e.g., weighting them at 50% or less—to reduce incentives for such practices while maintaining transparency. Empirical analyses indicate that unadjusted self-cites can inflate journal metrics by 10-20% in affected fields, underscoring the need for ongoing vigilance against these distortions.

Conflicts of Interest with Elsevier

Elsevier, the publisher of Scopus—the bibliographic database underlying CiteScore—also operates a large portfolio of academic journals, creating a structural conflict of interest in the metric's development and application. This vertical integration allows Elsevier to influence both the data inputs for CiteScore calculations and the journals that benefit from high scores, potentially incentivizing prioritization of its own titles in indexing, promotion, and metric design. Critics have highlighted that Scopus's proprietary nature restricts external audits of data selection and citation counting, unlike more transparent alternatives, exacerbating concerns over impartiality. Empirical analyses reveal that CiteScore tends to favor Elsevier journals relative to competitors. An examination by Eigenfactor researchers found that, excluding Nature journals, Elsevier titles receive approximately a 15% higher score under CiteScore compared to the Journal Impact Factor (JIF), a disparity not observed to the same degree for non-Elsevier publishers. This boost arises partly from methodological differences, such as CiteScore's four-year citation window and inclusion of non-article content, but the pattern aligns with Elsevier's incentives to calibrate the metric in ways that enhance its own portfolio's visibility. While no evidence of deliberate manipulation has been substantiated, the lack of independent oversight—stemming from Scopus's closed ecosystem—raises questions about whether such advantages are coincidental or systemic. Further scrutiny points to CiteScore's role in Elsevier's marketing strategies, where the metric is prominently featured on Scopus previews and journal pages to attract submissions, disproportionately benefiting Elsevier's hybrid and subscription models. Institutions evaluating journals via CiteScore may thus indirectly subsidize Elsevier through higher subscription revenues tied to inflated rankings, a dynamic less prevalent with JIF due to Clarivate's more limited publishing footprint. Proponents argue the metric's transparency in formula mitigates bias, yet the proprietary database remains a barrier to verifying claims of neutrality, underscoring the need for third-party replication studies.

Adoption, Usage, and Broader Impact

Integration in Academic Evaluations

CiteScore serves as a journal-level metric in academic promotion and tenure processes, where faculty committees evaluate publication venues by referencing a journal's CiteScore to assess the relative impact and quality of candidates' outputs, often alongside or as an alternative to the Journal Impact Factor. This integration allows authors to highlight CiteScore values in dossiers submitted to department chairs and review panels, providing a transparent indicator of citation performance derived from Scopus data covering over 25,000 serial titles as of 2023. In fields such as computer science, CiteScore extends to conference rankings, informing decisions on venue selection that influence hiring, tenure eligibility, and resource allocation by quantifying average citations per document over a four-year window. National research evaluation frameworks incorporate CiteScore through Scopus-derived percentiles to grade institutional and individual research quality. In Italy's Valutazione della Qualità della Ricerca (VQR) 2020–2024 exercise, administered by ANVUR, journal percentiles based on CiteScore are applied across evaluation groups (GEVs) to classify publications, contributing to funding distributions and performance rankings for universities. Similarly, funding bodies utilize CiteScore within Elsevier's SciVal platform to compute prominence scores, aiding grant assessments by weighting journals' citation shares in specific topics or fields. Despite its adoption, reliance on CiteScore in evaluations raises concerns over field-specific normalization, as higher values in biomedicine (often exceeding 10) contrast with lower averages in social sciences (typically under 3), potentially disadvantaging interdisciplinary or less citation-heavy disciplines without adjusted benchmarks. Institutions integrating the metric emphasize its free accessibility and broader journal coverage compared to proprietary alternatives, enabling more inclusive assessments beyond Web of Science-indexed titles.

Effects on Journal Rankings and Publishing Incentives

CiteScore's integration into journal evaluation frameworks has reshaped relative rankings compared to the Journal Impact Factor (JIF), with significant shifts favoring certain publishers. Independent analysis indicates that Elsevier journals, which dominate Scopus indexing, achieve an average rank improvement of 127 positions under CiteScore versus JIF, reflecting a performance boost of 12-25% relative to competitors due to differences in citation windows and document inclusion. Conversely, journals from publishers like Nature experience rank declines of around 173 positions on average, as CiteScore penalizes outlets with high volumes of non-citable items such as editorials and news pieces by incorporating them fully into the denominator. These discrepancies arise from CiteScore's four-year averaging of all Scopus documents against JIF's narrower two-year focus on citable items, leading institutions to recalibrate prestige assessments and potentially overlook journals disadvantaged by the metric's broader scope. Such ranking alterations influence author publishing incentives by directing submissions toward Scopus-indexed journals exhibiting strong CiteScore trajectories, as the metric's free public availability enhances its role in career advancement metrics like tenure reviews and grant applications. Authors increasingly weigh CiteScore alongside factors like review quality and audience reach, prioritizing venues with demonstrated citation momentum to bolster personal h-indexes and institutional benchmarks. This shift amplifies competitive pressures in fields with Scopus-heavy coverage, where high-CiteScore status correlates with greater visibility and funding appeal, though the metric's Elsevier ownership introduces credibility concerns, as evidenced by the systematic uplift for its portfolio. For publishers, CiteScore creates incentives to optimize content strategies by minimizing low-citation document types in the denominator, countering JIF's encouragement of expanded front matter and potentially fostering more focused research outputs. However, the metric retains broader pressures akin to JIF, such as pursuing special issues or editorial policies that elevate citation rates, though direct evidence of CiteScore-specific gaming remains sparse compared to well-documented Impact Factor manipulations. Overall, while promoting transparency through annual releases, CiteScore reinforces a publication ecosystem oriented toward quantifiable impact over intrinsic scientific merit, with evaluations often conflating correlation with causation in journal quality.

Empirical Studies on Correlation and Predictive Value

Empirical studies across disciplines have consistently reported strong positive correlations between CiteScore and the Journal Impact Factor (JIF), typically ranging from 0.78 to over 0.90, reflecting their shared reliance on citation-to-publication ratios despite methodological differences. In urological journals, a Pearson correlation of 0.787 was observed, highlighting substantial alignment in ranking high-impact outlets. Similarly, in radiology, nuclear medicine, and medical imaging journals analyzed using 2017 data, CiteScore exhibited a Spearman rank correlation of 0.911 (p < 0.001) with the Scimago Journal Rank (SJR), another Scopus-derived metric, underscoring internal consistency within the Scopus database while indirect links to JIF via overlapping high correlations (e.g., JIF-SJR rho ≈ 0.90) suggest comparable ordering. These correlations imply that CiteScore often reinforces JIF-based assessments but can diverge due to its inclusion of all document types in the denominator and a four-year (versus two-year) window, leading to higher average values and occasional quartile shifts—e.g., journals like Pharmacy Practice rank in the top quartile by CiteScore but lower by JIF. Fewer studies address CiteScore's standalone predictive value for future journal performance or article-level impact, though its extended window is posited to offer greater stability over JIF's shorter horizon, potentially reducing volatility from anomalous years. Machine learning approaches have successfully forecasted CiteScore one year ahead using historical citation and publication trends, achieving viable accuracy in pilot models that leverage lagged data, which supports its utility in anticipating medium-term impact trajectories. However, direct comparisons of predictive validity remain sparse; one analysis in library and information science journals found CiteScore's broader coverage (22,800+ journals versus JIF's 11,000+) enhances inclusivity but does not unequivocally outperform JIF in prognostic power for citation growth, as both metrics exhibit similar sensitivity to field-specific citation norms. Overall, while correlations affirm CiteScore's alignment with established benchmarks, its predictive edge— if any—appears tied to averaging effects rather than superior causal insight into research quality, warranting caution in high-stakes evaluations.

Recent Developments and Trends

Annual Releases from 2023 Onward

Elsevier has maintained an annual release cadence for CiteScore metrics since 2023, typically announcing updates in early June each year via the Scopus platform, approximately one month ahead of comparable journal impact factor releases. These releases incorporate the latest Scopus-indexed citation data, covering journals, book series, and conference proceedings, with metrics calculated over a four-year window to reflect serial citation impact. The 2023 release, on June 8, updated CiteScore values for 2022, drawing from citations received in 2022 to documents published during 2019–2022. This edition reported an average CiteScore increase of 8.6% across covered sources relative to the prior year's metrics, attributed to expanded Scopus coverage and rising citation volumes in certain fields. It expanded eligibility to additional journals, with over 31 new serials receiving their first CiteScore values, bringing the total to more than 216 for one publisher alone. In 2024, CiteScore 2023 was previewed on June 5 and fully released in early June, emphasizing enhanced transparency through detailed percentile ranks and subject category benchmarks. The metrics utilized citations from 2023 to publications in 2020–2023, enabling comparisons across over 28,000 serials and supporting evaluations in academic and institutional assessments. Notable outcomes included gains for specialized journals, such as a rise from 1.1 to 1.3 for certain biodiversity-focused outlets. The 2025 release of CiteScore 2024 occurred around June 3–9, continuing the focus on current, verifiable impact data amid growing Scopus content volume. Covering citations in 2024 to documents from 2021–2024, it documented incremental improvements for many titles, including a value of 1.4 for select open-access journals, reflecting sustained citation trends post-pandemic. As of October 2025, CiteScore 2025 previews via the monthly-updated CiteScore Tracker provide interim estimates based on partial 2025 data, with final computations deferred to mid-2026 after full-year citation accrual. These trackers forecast potential values by dividing accumulated citations by prior three-year publications, aiding early trend analysis without guaranteeing annual outcomes.

Observed Citation Trends and Metric Growth

CiteScore values for journals have exhibited steady growth since the metric's inception in 2016, reflecting broader increases in scientific output and citation volumes within the Scopus database. For instance, among the top 400 Scopus-indexed journals, the average CiteScore rose from 13.877 in 2015 to 16.536 in 2019, indicating a consistent upward trajectory driven by accumulating citations to prior-year publications. This pattern aligns with the metric's four-year citation window, where later releases incorporate additional citation data, often resulting in higher scores for established journals as references build over time. Recent annual releases underscore dynamic growth in specific segments of the literature. The 2023 CiteScore update revealed that 668 journal titles achieved an increase of 100% or more compared to 2022, highlighting rapid impact gains in emerging or high-output fields. Similarly, the 2024 release reported 649 titles with at least a doubling from 2023 levels, demonstrating sustained momentum amid expanding global research activity. These surges are partly attributable to the proliferation of publications, with Scopus-indexed scientific output reaching 3.3 million articles by 2022, alongside rising citation counts that amplify average per-document metrics. Observed citation trends in Scopus further contextualize CiteScore expansion, as total citations have grown in tandem with publication volumes, though at varying rates across disciplines. Studies of journal performance indicate that citations peak in later years post-publication, contributing to elevated CiteScores in subsequent calculations; for example, articles often accrue more annual citations between years 6 and 10 than in their initial five years. However, this growth is not uniform, with some fields experiencing steeper rises due to interdisciplinary referencing and open-access dissemination, while overall metric inflation risks arise from denominator effects like increased non-citable outputs in the publication base.

References

  1. [1]
    Scopus CiteScore - Elsevier
    Calculated using data from Scopus, CiteScore metrics help you evaluate journals, book series and conference proceedings to empower well-informed decisions.Using Citescore Metrics · Gather More Insights By... · Citescore Metrics Benefits
  2. [2]
    Track journal performance and make decisions with CiteScore
    Dec 7, 2016 · Today Elsevier is launching CiteScore metrics: a new standard that gives a more comprehensive, transparent and current view of a journal's ...Transparent · Current · Contributors
  3. [3]
    What is CiteScore and why should you care about it? - Elsevier
    Jun 3, 2021 · CiteScore is a metric that measures the impact of scholarly journals by calculating the average number of citations per article over a ...
  4. [4]
    Introducing CiteScore 2024: A Comprehensive and Transparent ...
    Jun 9, 2025 · CiteScore is a metric calculated for active titles rigorously screened and selected by independent Content Selection Advisory Boards (CSAB) for ...
  5. [5]
    Differences and similarities between Journal Impact Factor ... - NIH
    7. CiteScore: a new metric to help you track journal performance and make decisions. [(accessed Jun 1 2018)]; https://www.elsevier.com/editors-update/story/ ...<|separator|>
  6. [6]
    CiteScore-Flawed But Still A Game Changer - The Scholarly Kitchen
    Dec 12, 2016 · The CiteScore metric is controversial because of its overt biases against journals that publish a lot of front-matter.
  7. [7]
    Analysis of the CiteScore metric - Eigenfactor
    Elsevier not only owns the division (Scopus) that produces the CiteScore rankings, but also publishes a large fraction of the journals ranked therein. We ...
  8. [8]
    Controversial impact factor gets a heavyweight rival - Nature
    Dec 8, 2016 · “As there is intense competition among top-tier journals, adoption of CiteScore will push editors to stop publishing non-research documents, or ...
  9. [9]
    CiteScore: A Non-Rival for the Journal Impact Factor - Clarivate
    Dec 14, 2016 · The Eigenfactor.org study notes that Elsevier journals perform 15% higher than their competitors with CiteScore than they do with JIF.Missing: controversies criticisms
  10. [10]
    Scopus | Abstract and citation database - Elsevier
    A comprehensive abstract and citation database for impact makers. Find relevant research, identify top experts and gain strategic insights.Scopus content · Scopus data · Scopus AI · Scopus Author Profiles
  11. [11]
    Scopus LibGuide: Metrics
    Derived from the Scopus database, CiteScore metrics offer a more transparent, current, comprehensive and accurate indication of a serial's impact.Missing: formula | Show results with:formula<|control11|><|separator|>
  12. [12]
    CiteScore 2023: A comprehensive, clear and current metric for ...
    Jun 5, 2024 · CiteScore is a metric calculated for active titles within Scopus, with publications rigorously screened and selected by our independent Content Selection and ...
  13. [13]
    (PDF) Scopus Content Coverage Guide - ResearchGate
    Jan 5, 2019 · Scopus coverage has now grown to almost 3,500 humanities titles and includes all serial publication types, such as journals,. book series and ...
  14. [14]
    [PDF] Content Coverage Guide - Scopus
    CiteScore. The journal has a substantially lower. CiteScore, when compared to peer journals in its subject field. If a journal does not meet all of the three ...
  15. [15]
    Citation Research and Impact Metrics - Scopus
    Oct 1, 2025 · CiteScore only includes peer-reviewed research: articles, reviews, conference papers, data papers and book chapters, covering 4 years of ...
  16. [16]
    CiteScore metrics: Creating journal metrics from the Scopus citation ...
    Jun 19, 2019 · The initial launch of CiteScore metrics on 8 December 2016 was accompanied by significant promotional activity and attracted a great deal of ...<|control11|><|separator|>
  17. [17]
    (PDF) CiteScore metrics: Creating journal metrics from the Scopus ...
    In December 2016, after several years of development, Elsevier launched a set of transparent, comprehensive, current, and freely-available journal citation ...<|separator|>
  18. [18]
    Exploring CiteScore, Elsevier's new journal impact metrics
    Last week, the publisher launched CiteScore, a set of metrics that measure a scholarly journal's impact by looking at the average number of ...Missing: origins | Show results with:origins
  19. [19]
    2019 CiteScore™ Values Released, and Improved Methodology ...
    Jun 23, 2020 · Citations will now be counted cumulatively, from the year of publication until the end of the calculation window, which is up to four years.
  20. [20]
    Elsevier releases 2019 CiteScore - Library Learning Space
    Jun 24, 2020 · The new methodology will also be used to retrospectively calculated CiteScore values for all previous years (2011 – 2018) to allow for clear and ...
  21. [21]
    The new Scopus CiteScore formula and the Journal Impact Factor
    Jul 28, 2020 · In June 2020 Elsevier announced that the CiteScore metric of journals underwent a change. This work examines the effect of these changes for ...
  22. [22]
    [PDF] CiteScore™ insights 2024
    An annual value, released each June, that measures the citation impact of a title (journal, book series, conference proceeding and trade journals, including ...
  23. [23]
    Journal Citation Reports | Clarivate
    The Journal Impact Factor is a single journal-level measurement that should not be considered in isolation. We continue to advise that the JIF is a useful ...Journal Impact Factor (JIF) · Academia and Government blog · Search · Publishers
  24. [24]
    Measuring a journal's impact - Elsevier
    For more details, see this FAQ . CiteScore is calculated for the current year on a monthly basis until it is fixed as a permanent value in May the following ...<|separator|>
  25. [25]
    https://www.scopus.com/sources
  26. [26]
    Scopus Metrics (CiteScore, SNIP & SJR, h-index)
    Sep 30, 2025 · CiteScore is the number of citations received by a journal in one year to documents published in the three previous years, divided by the ...
  27. [27]
    CiteScore: Calculation Features and Application in Scopus
    How is CiteScore calculated? Calculation = the ratio of the average number of citations to the published number of articles over 4 years. The number of ...
  28. [28]
    Scholarly Metrics: SJR and SNIP - Research Guides
    Sep 19, 2025 · Source Normalized Impact per Paper (SNIP) measures contextual citation impact by weighting citations based on the total number of citations in a subject field.
  29. [29]
    Scopus: Journal Level Metrics - Library Guides
    Aug 2, 2025 · "SNIP measures a source's contextual citation impact by weighting citations based on the total number of citations in a subject field. It helps ...
  30. [30]
    [PDF] COMPARISON OF TOP 10 JOURNAL JIF, Citescore, SNIP, SIR ...
    SNIP and SJR use the same dataset as Citescore. SNIP uses a similar ratio but normalizes based on subject. SJR uses the same size independent approach and also ...
  31. [31]
    Impact factor correlations with Scimago Journal Rank ... - PubMed
    Feb 6, 2019 · Results: The stronger bivariate correlations were represented by the ES↔Total Cites RS = 0.968, p < 0.001, R2 = 0.937; and the CiteScore↔SJR RS ...
  32. [32]
    CiteScore, SNIP and SJR - Research Impact Metrics
    Oct 14, 2025 · SJR, SNIP and CiteScore are the journal level metrics used by Scopus. Great question! Here's a clear comparison of CiteScore, SNIP, and SJR ...
  33. [33]
    Differentiate JIF, JCI, Citescore, SJR, SNIP - WoS Journals
    SNIP (Source Normalized Impact per Paper) measures contextual citation impact by weighting citations based on the total number of citations in a subject field.
  34. [34]
    CiteScore Hub - Elsevier
    CiteScore is a journal impact metric by Elsevier, calculated by averaging citations over four years, including a broader range of publications.Citescore Tracker: A... · Citescore Rank & Trend · Interpreting Citescore Rank...
  35. [35]
    [PDF] Web of Science versus Scopus: Journal Coverage Overlap Analysis1
    28,560 journal titles are indexed in Web of Science and 37,535 journals are indexed in Scopus ... the number of years with overlap between WoS and Scopus ...
  36. [36]
    [PDF] The Journal Coverage of Web of Science, Scopus and Dimensions
    Web of Science is most selective, Dimensions most exhaustive. Scopus has wider coverage than Web of Science. Dimensions has 82.22% more journals than Web of ...
  37. [37]
    The Journal Coverage of Web of Science and Scopus - ResearchGate
    Aug 6, 2025 · This research compares the journal coverage of both databases in terms of fields, countries and languages, using Ulrich's extensive periodical directory as a ...
  38. [38]
    Web of Science (WoS) and Scopus: The Titans of Bibliographic ...
    For example, Scopus was shown to cover even 99% of nursing journals that were covered by WoS [85]. Meanwhile, in another study focused on computer sciences, ...
  39. [39]
    CiteScore metrics are suitable to address different situations
    Aug 10, 2025 · ... A case study with journals from six different subject fields suggested that CiteScore may effectively measure relative citation impact ...Missing: broader applicability inclusivity
  40. [40]
    Will open access increase journal CiteScores? An empirical ...
    Aug 30, 2018 · This paper empirically studies the effect of Open Access on journal CiteScores. We have found that the general effect is positive but not uniform across ...
  41. [41]
    Web of Science and Scopus are not global databases of knowledge
    Both Web of Science and Scopus are critical components of the current research ecosystem, providing the basis for university and global rankings as well as ...
  42. [42]
    The impact of size-independence, citing-journal weighting, and ...
    Non-normalized metrics do not account for field- or subfield-specific differences in citation impact. They are therefore inappropriate for the comparison of ...
  43. [43]
    Modified CiteScore metric for reducing the effect of self-citations
    Aug 10, 2025 · The ultimate aim is to reduce the numerical effect of self-citations (SC) in academic publishing. The reduction is designed to discourage SC but ...Missing: susceptibility | Show results with:susceptibility
  44. [44]
    Major indexing service sounds alarm on self-citations by nearly 50 ...
    Jun 29, 2020 · At least 25% of the references in all of the suppressed journals were self-citations, with 71% of the citations in one journal, the Journal of ...Missing: CiteScore susceptibility
  45. [45]
    Citation Performance Indicators - A Very Short Introduction
    May 15, 2017 · ... self-citation and gaming, especially in Google Scholar. PRODUCER ... CiteScore–Flawed But Still A Game Changer · Will Citation ...
  46. [46]
    Citation manipulation through citation mills and pre-print servers
    Feb 14, 2025 · Citations are widely considered in scientists' evaluation. As such, scientists may be incentivized to inflate their citation counts.Missing: CiteScore | Show results with:CiteScore
  47. [47]
    Billions at Stake: How Self-Citation Adjusted Metrics Can Transform ...
    May 11, 2025 · Our findings demonstrate that traditional citation metrics can be significantly inflated by self-citation, with an average h-index inflation of ...
  48. [48]
    News Keeping Score of CiteScore - Galter Health Sciences Library
    Sep 25, 2023 · CiteScore: a new metric to help you track journal performance and make decisions. 2016. Available from https://www.elsevier.com/editors-update/ ...
  49. [49]
    A Focus on Faculty: Promotion, Tenure and Impact Factors
    Jun 22, 2022 · Promotion and tenure (P&T) are milestones in a professor or ... Its “About” page may include the Journal Impact Factor or CiteScore if it has one, ...
  50. [50]
    Introducing CiteScore, Our Journal's Preferred Citation Index
    May 10, 2020 · Authors who hold academic positions will be able to easily share this information with department chairs and Promotion and Tenure committees ...
  51. [51]
    Using Scopus's CiteScore for assessing the quality of computer ...
    This study reviews relevant literature and tests Scopus's CiteScore database and method for evaluating the quality of 395 conferences in the field.
  52. [52]
    [XLS] Journal Percentiles - ANVUR
    This file is prepared for ANVUR for use in the Valutazione della Qualità della Ricerca (VQR - Research Quality Assessment) 2020-2024. ... CiteScore ...
  53. [53]
    Tools and resources for research quality assessment | IMT
    ... CITESCORE, SJR values and ANVUR classification of journals ... assessment criteria adopted by ANVUR as part of the national research evaluation procedures:.
  54. [54]
    Scopus CiteScore: an overview - IAES Journals
    May 21, 2024 · Funding and grants: funding agencies can consider a journal's CiteScore when assessing the quality of research output for grant applications.
  55. [55]
    Predicting the citation count and CiteScore of journals one year in ...
    Aug 10, 2025 · What is required is the knowledge of producing the forecast in the easiest of ways possible after establishing that such forecasting is possible ...
  56. [56]
    Elsevier Releases 2022 CiteScore Annual Update | NISO website
    Jun 8, 2023 · Elsevier Releases 2022 CiteScore Annual Update. June 2023. NISO Member News. Elsevier has released the 2022 #CiteScore annual update, its ...
  57. [57]
    https://www.worldscientific.com/do/10.1142/news20230612.463937/full
  58. [58]
    2022 CiteScores – Released - MDPI
    8 June 2023. 2022 CiteScores – Released · 31 journals received their first CiteScore, taking the total number of journals with a CiteScore to 216 ...
  59. [59]
    Announcement of the Release of Scopus CiteScore 2023 in 2024
    Jun 10, 2024 · We are proud to announce that the Scopus CiteScore 2023 has officially been released in 2024! We are delighted to share that, overall, our journal's CiteScore ...<|separator|>
  60. [60]
    New Scopus CiteScore for Check List - Pensoft Publishers
    Jun 26, 2024 · The latest Scopus CiteScore for Check List, released in early June 2024, stands at 1.3. The score marks an increase from last year's 1.1.
  61. [61]
    New Scopus CiteScore for Check List - Pensoft Publishers
    Jun 9, 2025 · A Scopus CiteScore is estimated for each journal indexed in Scopus, the largest abstract and citation database of peer-reviewed literature. The ...
  62. [62]
    Scopus CiteScoreTracker for ETASR (monthly announcement
    Jun 6, 2025 · Scopus CiteScoreTracker 2025 is updated on a monthly basis by Scopus, until the final annual CiteScore calculation which is usually ...
  63. [63]
    2024 CiteScore Released: Open Exploration Journals' Performance
    Elsevier officially released the 2024 CiteScore in Scopus, the most comprehensive multidisciplinary literature database in the world.
  64. [64]
    Ten Years of Evidence for the Top 400 Scopus-indexed Journals of ...
    Sep 4, 2023 · In mid-2020, the equation used to calculate CiteScore changed to reflect a four-year data window from the previous three-year data set. By ...
  65. [65]
    Publications Output: U.S. Trends and International Comparisons | NSF
    Dec 11, 2023 · Global publication output reached 3.3 million articles in 2022, based on data from the Scopus database of S&E publications.
  66. [66]
    Factors Affecting Publication Impact and Citation Trends Over Time
    Jun 15, 2023 · Publications received a greater number of citations annually in their 6th to 10th years, compared to the first 5. The number of references ...<|control11|><|separator|>
  67. [67]
    (PDF) Scopus 1900–2020: Growth in articles, abstracts, countries ...
    Jan 13, 2022 · (2022) Scopus 1900-2020: Growth in articles, abstracts, countries, fields, and journals. 4.0) license. (25%). citation indexes, Scopus.