Impact factor

Clarivate
Company type	Public company
Industry	Information services, scientometrics
Founded	2018
Headquarters	Philadelphia, United States
Products	Web of Science, Journal Citation Reports (JCR)
Website	clarivate.com

The impact factor (IF), also known as the journal impact factor (JIF), serves as a quantitative measure for ranking academic journals. Journals boasting higher impact factors are generally regarded as more authoritative or significant within their respective disciplines.

This metric represents the mean number of citations that articles published in the preceding two years receive. It is extensively utilized by academic institutions and funding agencies when making decisions about promotions and research grants, though it has attracted criticism for possibly influencing scientific behavior in unintended ways.

The Impact Factor is a scientometric indicator computed by Clarivate's Web of Science platform.

History

The concept of the impact factor was developed by Eugene Garfield, who founded the Institute for Scientific Information (ISI) based in Philadelphia. Since 1975, annual impact factors have been calculated for journals indexed in the Journal Citation Reports (JCR). In 1992, ISI was acquired by Thomson Scientific & Healthcare, becoming Thomson ISI. Later, in 2018, Thomson-Reuters divested ISI, which was purchased by Onex Corporation and Baring Private Equity Asia, leading to the formation of Clarivate, now responsible for publishing the JCR.

Calculation

The two-year journal impact factor quantifies citations made in a specific year to articles published during the two preceding years, divided by the total number of "citable items" released in those same years:

<math display=block>\text{IF}_y = \frac{\text{Citations}_y}{\text{Publications}_{y-1} + \text{Publications}_{y-2}}.</math>

To illustrate, Nature had an impact factor of 41.577 in 2017:

This means that papers published in 2015–2016 were cited on average around 42 times each during 2017. Impact factors are typically published the year after citations are fully tallied.

The definition of "citations" and "publications" (or "citable items") affects the impact factor calculation. ISI categorizes "publications" as articles, reviews, or conference papers indexed within Web of Science (WoS), excluding materials such as editorials or corrections. While WoS users can verify publication counts, citation data derives from the proprietary Journal Citation Reports (JCR), which is not accessible to external parties. Consequently, the JCR impact factor remains a proprietary metric defined by ISI.

Newly indexed journals receive an impact factor only after two full years of indexing; if indexing begins post volume one, three years are required. Occasionally, partial impact factors may be assigned to newer journals based on incomplete datasets. The calculation always involves two complete years of publication data; for recent titles, one year might have zero publications. Journals with irregular publishing schedules may have years without citable items, influencing these counts.

Impact factors can also be calculated over different time spans; for example, JCR offers a five-year impact factor, which divides citations received in a given year by articles published over the prior five years.

Use

Initially designed to assist university librarians in selecting journals, the impact factor rapidly evolved into a pivotal metric for assessing academic performance. Hoeffel (1998) summarized its role as follows:

While not an ideal instrument for measuring article quality, Impact Factor—being well-established—functions as a valuable scientific evaluation method. The top journals within each specialty—those most challenging to publish in—generally possess high impact factors. These esteemed journals existed before the impact factor’s creation, and their elevated impact factors reflect their excellence rather than causing it.

As a metric at the journal level, applying the impact factor to evaluate individual articles or researchers remains contentious. Eugene Garfield himself agreed with Hoeffel’s assessment but warned against its inappropriate use in appraising individuals due to citation variability within journals. Nevertheless, the JIF has become central to decisions regarding researchers’ employment and funding opportunities. In 2007, The Journal of Cell Biology observed:

Impact factor data significantly shape choices about where authors submit work, who gets promoted or hired, success with grant applications, and even salary incentives.

Studies confirm that JIF is deeply embedded in both formal and informal evaluation frameworks. A 2019 review revealed that 40% of universities in the United States and Canada explicitly reference the JIF within policies related to review, promotion, and tenure. Furthermore, a 2017 survey among life sciences researchers found that pressures to publish in high-impact journals heavily influence decision-making processes. This dependence on impact factors not only affects assessment but also steers research directions, often discouraging innovative or unconventional projects.

Criticism

The reliance on impact factors has been widely challenged due to concerns about its statistical soundness and its role in assessing scientific work. One major issue is that impact factors represent the arithmetic mean of data that do not follow a normal distribution, leading some experts to argue that the median would provide a more accurate measure. There is ongoing debate about whether impact factors truly reflect a journal's significance, with attention drawn to editorial strategies designed to elevate these metrics, sometimes to the detriment of both readers and authors.

Opponents highlight how this metric shapes researcher behavior, promoting publication in high-impact journals as a priority over the intrinsic quality or relevance of research. This dynamic fosters a "privilege paradox," wherein early-career researchers—particularly those from less-represented regions—find it challenging to gain acknowledgment outside these numerical indicators. Such emphasis risks limiting the variety of research topics and methodologies pursued.

Additional critiques connect the focus on impact factors to neoliberal trends influencing academia, advocating not only for improved evaluative tools but also for wider conversations concerning the societal value of research assessment and the increasing instability faced by scientific professionals in higher education.

Limitations of impact factor for individual assessments and cross-disciplinary comparisons

Impact factors and citation metrics are heavily affected by discipline-specific variables, making it unreliable to compare journals across different fields or even within sub-disciplines. The percentage of citations accrued within two years after publication varies greatly—from as low as 1–3% in mathematical and physical sciences to 5–8% in biological sciences—thereby restricting the applicability of impact factors for comparing journals across disciplines.

The inappropriate use of impact factors to judge individual articles can unfairly disadvantage certain subjects. In 2004, the House of Commons Science and Technology Select Committee urged the Higher Education Funding Council for England to remind panels evaluating research that article quality should be prioritized over journal prestige. Numerous studies reinforce that impact factor is an indicator at the journal level and is unsuitable for assessing individual researchers or institutions.

Editorial practices influencing impact factor reliability

Given the widespread acceptance of impact factor as an indicator of research quality, some journals implement editorial policies—ranging from acceptable to questionable—to enhance their scores. For instance, increasing the proportion of review articles, which typically attract more citations than original studies, is a common approach. A 2020 study on dentistry journals revealed that systematic reviews have a notable effect on boosting impact factors, whereas clinical trials do not; moreover, publishing a larger number of articles correlates with higher impact factors.

Editors may also reduce the count of "citable items" (the denominator in impact factor calculations) by rejecting submissions unlikely to be cited or modifying article formats—such as removing abstracts or reference lists—to prevent classification as citable content. These tactics have caused fluctuations in impact factors exceeding 300%. Despite being labeled uncitable, such items can still contribute citations to the numerator, complicating precise measurement.

Another strategy involves releasing highly cited papers early in the year to maximize citation accumulation time. Journals might also engage in extensive self-citation practices to artificially inflate their impact factors.

Some publications have resorted to blatant manipulation. For example, in 2007 Folia Phoniatrica et Logopaedica published an editorial citing all its recent articles, temporarily doubling its impact factor but leading to its removal from subsequent Journal Citation Reports.

Coercive citation—where editors compel authors to add irrelevant citations to increase a journal’s impact factor—is reported by approximately one in five researchers in fields such as economics, sociology, psychology, and business. This practice is more prevalent in business journals and those with lower impact factors. While prominent editors within leading business journals have publicly condemned coercive citation, isolated instances continue across various disciplines.

Presumed link between impact factor and quality

Originally developed by Eugene Garfield to assist librarians in selecting journals for indexing based on citation frequency, the journal impact factor (JIF) has since become a prevalent indicator of journal "quality" and is frequently employed in evaluating researchers at institutional levels. This transformation has considerably shaped research methodologies and academic conduct.

By 2010, numerous funding agencies advised against relying on quantitative metrics such as the JIF to assess quality. Investigations revealed that the JIF is highly susceptible to manipulation, and its persistent application is largely due to its straightforwardness rather than an accurate representation of research excellence.

The inappropriate application of the JIF and comparable indicators adversely affects scholarly communication by creating mismatches between journal prominence and the merit of individual articles. It also contributes to the marginalization of social sciences, humanities, and research outputs from Latin America, Africa, and Southeast Asia. Additionally, it diminishes visibility for research published in local languages and topics of regional significance while promoting unethical citation and authorship behaviors. More broadly, an overreliance on the JIF cultivates a reputation-driven system that favors journal prestige over methodological rigor, reproducibility, and societal relevance, ultimately undermining the integrity of research.

Various regional and global initiatives advocate for alternative evaluation frameworks, including the Leiden Manifesto and the San Francisco Declaration on Research Assessment (DORA). Plan S supports adopting these approaches alongside comprehensive reforms in scholarly communication. The redefinition of research excellence now prioritizes transparent processes and open access to results.

Despite these initiatives, the JIF continues to be extensively utilized worldwide, creating challenges due to its lack of transparency and influence from publisher negotiations affecting its values.

Negotiated impact factor values

The calculation of impact factors can differ greatly depending on which items are classified as "citable" and included in the denominator. A notable example occurred in 1988 when meeting abstracts were removed from FASEB Journal’s count, causing its impact factor to surge dramatically from 0.24 to 18.3 within a year. Publishers frequently engage in discussions with Clarivate to improve their journals’ impact factors, resulting in "negotiated values" that can cause significant fluctuations in scores, occasionally triggered by events such as acquisition by large publishing houses.

Skewed citation distributions

Citation data typically exhibit pronounced skewness, rendering average citation counts potentially deceptive when evaluating the typical influence of individual articles as opposed to overall journal impact. For example, approximately 90% of Nature's 2004 impact factor derived from only about 25% of its published papers. Therefore, most articles receive fewer citations than suggested by the journal’s mean. Furthermore, since digital access has expanded, the correlation between journal-level impact factors and citations received by individual articles has weakened.

Exceptional cases can heavily distort impact factors; for instance, the paper "A short history of SHELX" amassed over 6,600 citations, causing Acta Crystallographica Section A's impact factor to leap from 2.051 in 2008 to 49.926 in 2009—exceeding those of Nature and Science. The second most cited article in that issue had merely 28 citations.

Critics highlight that using an arithmetic mean for impact factor calculations is problematic due to citation distribution skewness, prompting suggestions for alternative metrics that better represent citation patterns. Nonetheless, some research argues that statistical critiques against applying the impact factor at the article level are unpersuasive and that it may actually serve as a more reliable proxy for article value than individual citation counts alone.

Transparency and reproducibility

Although the formula underpinning the JIF is publicly available, access to the underlying dataset used for its computation remains restricted, leading to concerns regarding transparency. A study conducted in 2019 demonstrated that with appropriate access and meticulous data processing, reproducing the JIF is feasible but demands considerable effort. Further work published in 2020 showed that approximate impact factors can be generated using open or partially open databases such as Google Scholar, ResearchGate, and Scopus without requiring purchase of Web of Science or Journal Citation Reports data.

Wider detrimental effects on scientific research

Beyond direct criticisms, reliance on impact factors has been connected with broader negative consequences within scientific inquiry. Evidence suggests that dependence on bibliometric indicators like JIF can degrade peer review standards, discourage data sharing practices, reduce article quality overall, and narrow what kinds of research are deemed publishable. Many researchers tend to pursue projects expected to score well on these metrics exclusively. Additionally, publication timelines lengthen as authors target journals with higher impact factors first, resulting in editors and reviewers managing submissions poorly aligned with their outlets’ scopes.

Institutional reactions addressing impact factor criticism

In light of concerns surrounding widespread use of the journal impact factor (JIF) for research evaluation, several organizations have taken steps to curtail its influence. In 2007, the European Association of Science Editors (EASE) advised exercising caution when using JIFs solely for comparing entire journals rather than individual articles or researchers.

The International Council for Science proposed in 2008 measures including capping annual publication counts per scientist and even penalizing excessive output. In 2010, Germany’s Deutsche Forschungsgemeinschaft introduced guidelines limiting publications considered within funding applications while highlighting problematic emphasis on quantity over quality alongside growing dependence on metrics such as the h-index and JIF. The UK's 2014 Research Assessment Exercise prohibited use of JIFs; however enforcement was inconsistent.

The American Society for Cell Biology together with publishers initiated the San Francisco Declaration on Research Assessment (DORA) in 2013 aiming to discourage inappropriate use of JIFs when evaluating research contributions and career progressions. DORA has since gained endorsement from thousands of individuals and institutions including the League of European Research Universities.

Publishers have acknowledged inherent flaws in JIFs; Nature underscored their statistical limitations and detrimental effects on scientific integrity by warning that pressure to publish exclusively in high-impact journals may foster questionable research practices. Some publishers now incorporate multiple metrics; for example, PLOS journals refrain from displaying JIFs altogether. Microsoft Academic excludes both JIF and h-index metrics citing their limited reliability as indicators of true scholarly influence.

In 2021, Utrecht University declared it would cease all use of quantitative bibliometrics including JIFs, labeling them a "very sick model" irrelevant for advancing science. This followed a 2018 decision by the Dutch Research Council (NWO) removing references to JIFs and h-indices from funding calls and applications. Utrecht’s stance faced opposition from over 150 Dutch academics who argued that despite imperfections these metrics remain useful indicators and that abandoning them risks arbitrariness undermining scientific standards.

Associated metrics

Other related indicators published by Clarivate include:

Cited half-life: representing the median age of articles cited each year within Journal Citation Reports. For instance, a half-life value of five years in 2005 indicates half of citations originate from articles published between 2001–2005.
Aggregate impact factor: computed by aggregating citations and publications across all journals within a subject category.
Immediacy index: calculated as citations received during a given year divided by articles published within that same year.
Journal citation indicator (JCI): a field-normalized metric akin to Source Normalized Impact per Paper based on Scopus data covering all journals indexed in the Web of Science Core Collection (WoS CC), including AHCI, ESCI, BCI, CPCI segments. While traditionally limited to SCIE and SSCI journals, starting June 2023 JIF will extend coverage across all WoS CC titles.
JIF quartile ranking: classification placing journals into four quartiles within each subject category.
JIF percentile ranking: percentile position within relevant subject categories.

Journals may receive differing quartile or percentile rankings depending on their classification categories.

Clarivate excludes certain content types such as news items, correspondence letters, or errata from denominator calculations to refine metric accuracy.

Alternative scientific impact measures

Additional journal-level metrics are provided by other entities like Elsevier’s introduction of CiteScore in December 2016 covering serial titles indexed in Scopus. While these focus on journals broadly, author-level metrics such as the h-index evaluate individual researcher performance; article-level metrics assess influence at the single publication scale.

Alternative metrics ("altmetrics") capture immediate social engagement through article views, downloads, social media mentions emphasizing societal impact beyond traditional citation-based measures.

Fraudulent impact factors

Certain organizations, such as Global Impact Factor, Citefactor, and Universal Impact Factor, generate counterfeit or misleading impact factor metrics. Jeffrey Beall compiled a list identifying these deceptive measurement systems. Some of these dubious approaches involve presenting "alternative impact factors" derived from citation databases like Google Scholar or Microsoft Academic, instead of relying on the official Journal Citation Reports (JCR).

These spurious metrics are frequently exploited by predatory publishers. Confirming a journal's legitimate indexing status can be done by consulting the master journal list provided by Journal Citation Reports. The existence of fabricated impact factors is widely acknowledged as a critical warning sign when assessing the credibility and quality of academic journals.