Eigenfactor and Article Influence Score: Advanced Journal Ranking Methods

The Eigenfactor Score and Article Influence Score are two related but distinct journal-level metrics developed at the University of Washington, designed to address structural limitations in the traditional Impact Factor. Both draw on network analysis rather than simple citation counts, which changes what they measure — and what they miss. Understanding how these metrics work helps researchers, librarians, and institutions make more informed decisions about where to publish and which journals to trust.

Definition and scope

The Eigenfactor Score, introduced by Jevin West and Carl Bergstrom at the University of Washington around 2007, treats the scholarly literature as a network of journals connected by citations. A journal's Eigenfactor Score reflects how often a hypothetical researcher following citation links at random would land in that journal. Citations from highly influential journals carry more weight than citations from peripheral ones — a structural choice that separates Eigenfactor from the raw arithmetic of the Impact Factor and other journal metrics.

The Article Influence Score divides a journal's Eigenfactor Score by the fraction of all articles that journal contributes to the literature over a five-year window. The result is a per-article measure that allows direct comparison across journals of very different sizes. A journal publishing 5,000 articles per year and one publishing 50 articles per year are evaluated on comparable footing.

Both metrics are calculated using data from Journal Citation Reports (JCR) and are freely available at Eigenfactor.org, a project hosted under the University of Washington's research infrastructure.

How it works

The Eigenfactor calculation is adapted from Google's PageRank algorithm. Citations from high-Eigenfactor journals count for more than citations from low-Eigenfactor journals — a recursive weighting that PageRank applies to web pages and Eigenfactor applies to academic journals.

The calculation follows these steps:

  1. Construct a citation matrix using five years of citation data from journals indexed in the Web of Science, stripping out self-citations (citations from a journal to itself, which are excluded entirely from Eigenfactor).
  2. Normalize the matrix so that each column (representing a journal's outgoing citations) sums to one, creating a probability distribution of citation destinations.
  3. Apply iterative multiplication — the equivalent of simulating millions of random walks through the citation network — until the distribution converges on a stable set of scores.
  4. Scale the result so that all Eigenfactor scores across the entire journal set sum to 100.
  5. Derive Article Influence by dividing the Eigenfactor Score by the journal's proportional share of all articles published in the dataset.

An Article Influence Score of 1.00 represents the average influence per article across the literature. Journals above 1.00 are above average; journals below 1.00 are below. Nature and Science consistently score well above 10.00 on this measure. A specialized but rigorous journal in a narrower discipline might score between 0.5 and 1.5 — respectable, but not because the field lacks rigor, only because its citation network is smaller.

Common scenarios

The practical difference between these metrics and simpler citation counts shows up clearly in two common situations.

Large high-volume journals versus small specialty journals. A journal publishing 3,000 articles per year can accumulate a large raw Eigenfactor Score simply because it contributes more to the literature. The Article Influence Score normalizes this, making it the more appropriate tool when comparing a broad-scope journal to a focused one — much like the discussion in SciMago Journal Rank explained, which also attempts network-weighted normalization.

Identifying citation laundering. Because Eigenfactor removes self-citations from the calculation entirely, journals that inflate their Impact Factor through self-citation receive no corresponding Eigenfactor benefit. This structural choice has made Eigenfactor a useful cross-check when evaluating journals that appear strong by one metric but weaker by another. It's also worth consulting predatory journals: how to identify them when metric discrepancies appear extreme and unexplained.

The broader index of journal ranking topics on this site provides context for how these metrics fit alongside h-index measures and database-level indicators.

Decision boundaries

Eigenfactor and Article Influence Score are most useful when applied with their constraints clearly in mind.

Where they work well: Comparing journals within a shared citation ecosystem — biomedicine, chemistry, physics — where the underlying citation network is dense and the JCR coverage is comprehensive. The recursive weighting rewards journals that are cited by other important journals, which captures scientific influence more faithfully than counting citations equally.

Where they fall short: Humanities, social sciences, and interdisciplinary fields where book citations, non-English literature, and conference proceedings dominate. JCR coverage is uneven in these areas, and citation networks are sparser. The H-index and citation metrics page addresses related coverage problems at the author level.

What they cannot tell you: Neither metric measures article-level quality, peer review rigor, editorial standards, or the peer review process itself. A journal's aggregate influence score says nothing about any individual paper within it. Researchers choosing between submission targets should combine Eigenfactor data with discipline-specific norms, scope alignment, and the journal's editorial practices — all of which are covered in how to choose the right journal.

The Eigenfactor and Article Influence systems represent a genuine methodological advance over single-window citation ratios. They are also, like every bibliometric tool, a model — and models simplify the things they measure.

References

Read Next

Impact Factor and Journal Metrics: What They Mean and How They Are Calculated Journal metrics — led by the Impact Factor but extending into a constellation of related scores — function as the credit... SCImago Journal Rank (SJR): How It Works and Why It Matters Developed by the SCImago Lab research group using data from Elsevier's Scopus database, SJR applies a logic borrowed from... Predatory Journals: How to Identify and Avoid Them The problem is neither fringe nor new — it has grown alongside the open-access publishing movement, exploiting the fee-based...