«PrevIP Strategy Creation and Its Importance to IP Operations NextChina’s Market for Orthopedic Devices to See Explosive Growth»

Confronting an ever-expanding population of journals, and given the perennial stresses on their time and resources, how can librarians hope to identify the most significant, useful and trustworthy titles for their own communities of students and researchers?

Four decades ago, Thomson Reuters introduced the Journal Citation Reports® (JCR) and its Impact Factor metric, providing a consistent, uniform measure of journal influence as reflected by citations – that is, as judged by scientists themselves, in acknowledging the most significant research to underpin and advance their own work. Since then, for the 12,000-plus journals indexed in the Web of Science™, the Impact Factor has continued to be a widely anticipated, scrutinized and quoted figure each year upon the annual release of the JCR update.

Nowadays, however, the Impact Factor is only one component of the journal-metrics tools found in the JCR. Newer measurements afford a more extensive, detailed and nuanced view of journal impact. Chief among these indicators is the Eigenfactor® Score.

While the Impact Factor conveys the ratio of citations in the latest JCR year to all citable items published by a journal in the previous two years, the Eigenfactor Score adds another dimension – taking into account the journals in which the citing articles are published. The Eigenfactor’s premise, developed by Carl T. Bergstrom and Jevin D. West of the University of Washington and their colleagues, is that not all citations are necessarily equal.

Consulting citation statistics from the JCR, Bergstrom and colleagues envisioned scholarly communication in journals as a network, in which some journals emerge as larger and more-central nodes of activity, as their publications prove to be consistently valuable to researchers.

Following Citations Ad Infinitum

To illustrate the concept, Bergstrom, West and colleagues invoke a hypothetical (and apparently immortal) researcher spending “all eternity” in the library, randomly following citations in scientific periodicals, selecting a citation in a given journal, consulting the cited journal, selecting another citation, and so on. Ultimately, as the process is repeated “ad infinitum,” the researcher’s pursuit will lead her more frequently to, as Bergstrom and colleagues observe, “journals that are highly cited by journals that are also highly cited.”

The Eigenfactor Score represents the percentage of time the researcher would be directed to a given journal, thus providing a measure of the journal’s influence. An Eigenfactor score of 2 for Nature, for example, indicates that the researcher would have spent 2 percent of her time tracking citations to that journal, whereas other journals might merit only (and literally) a fraction of that time.

Eigenfactor scores are scaled so that, together, the scores of all journals covered in the JCR will add up to 100. In the current JCR, the journal PLOS One has the highest Eigenfactor Score, at 1.53341, indicating roughly 1.5 percent of the total influence of all indexed journals.

In sum, the Eigenfactor Score embodies the theory that a single citation from an influential, highly cited journal may hold more value than multiple citations from a peripheral journal.

The Eigenfactor Score is calculated based on five years of publication, in contrast to the original Impact Factor, which calculates impact based on two years (although, since 2007, the JCR has also featured Five-Year Impact Factor scores). Moreover, by discounting citations from one journal to papers published in the same journal, the Eigenfactor Score eliminates the effect of journal self-citation.


Other Eigenfactor metrics found in the JCR provide additional dimensions to journal evaluation. The Normalized Eigenfactor Score starts with a calculation of 1 for the average journal and then measures journals relative to that benchmark. A Normalized Eigenfactor Score of 2, for example, indicates that a journal wields twice the influence of an average JCR-listed journal.

The Eigenfactor’s creators also addressed the matter of measuring the average influence, per article, of the papers in a given journal. The result is the Article Influence Score, a metric somewhat analogous to the Impact Factor, calculated as a journal’s Eigenfactor Score divided by the number of articles published in that journal over the previous five years. This score is normalized so that the average article in journals covered in the JCR is assigned the value of 1.00. Therefore, if a journal has an Article Influence Score of 2.0, its articles, on average, are twice as influential as the average journal covered in the JCR.

Used in combination, Eigenfactor metrics and the other tools available in the JCR provide a detailed view of journal impact in all its aspects.