Re-Ranking Journals with New Bibliometric Parameters

When a library committee sits down to decide which journals to keep, renew, or drop, the first number most people reach for is the citation count or the Impact Factor. The journal with the most citations wins the shelf space. But what if that simple ranking is misleading? What if a journal looks important only because it publishes a huge volume of articles, while a leaner, more focused journal quietly delivers far more value per page? This is exactly the problem the Indian bibliometrician I. N. Sengupta set out to solve in 1986, when he proposed three new bibliometric parameters to re-rank scientific periodicals more accurately. These parameters-commonly written as D/A, CIA, and D/C-still offer a sharper, fairer way to judge journals than raw citation counts alone.

Why traditional citation counts fall short 🔗

Citation counting has been the backbone of journal evaluation for decades. The logic is straightforward: a journal that gets cited often must be publishing influential work. The most famous version of this idea is the Journal Impact Factor (JIF), which divides the citations a journal received in a year by the number of articles it published in the previous two years. This ratio was designed partly to stop large, frequently published journals from dominating rankings simply because of their size.

Despite this, ranking lists built on citations carry several hidden flaws. A journal that publishes review articles, which tend to attract heavy citation, can sit artificially high. Coverage is uneven across disciplines, so clinical or applied journals often score lower than research-heavy ones even when they serve readers well. Citation practices also differ sharply between fields, meaning a mathematics journal and a molecular biology journal cannot be compared on the same citation scale.

There is a deeper concern too. A growing body of research shows that citation counts and Impact Factors are weak and inconsistent predictors of actual research quality, and are sometimes even negatively related to it. A systematic survey of the metric concluded that the JIF is frequently used well beyond what it was originally meant to measure. Sengupta’s insight was that a single ratio cannot capture everything that makes a journal valuable. We need more than one lens.

Introducing Sengupta’s three new parameters 🔗

Working at the Indian Institute of Chemical Biology in Calcutta, Sengupta argued that traditional ranking lists should not be discarded but refined. He proposed three additional parameters that, when applied on top of an existing citation-based list, reveal the true position of each journal in order of usefulness. His paper described them as measures of (1) scientific interest in relation to the total number of articles published, (2) the compactness of information content, and (3) scientific value in relation to that compactness. In the notation used in most Indian library science syllabi, these become D/A, CIA, and D/C.

Before defining them, it helps to fix the symbols. D represents the demand or scientific interest a journal generates, expressed through the citations or use it attracts. A is the number of articles the journal publishes. C stands for compactness, a measure of how densely information is packed into each article. The three parameters simply combine these quantities in different ways.

The D/A ratio – scientific interest per article 🔗

The D/A ratio divides the demand a journal attracts by the number of articles it publishes. In plain terms, it asks how much scientific interest each article generates on average. A journal that publishes 500 papers and attracts a large number of citations may look impressive in absolute terms, but if most of those papers go unnoticed, its D/A ratio will be modest. By contrast, a journal that publishes 80 carefully selected papers, each drawing strong attention, will record a high D/A value.

This parameter directly tackles the volume problem. It prevents a journal from climbing the ranks merely by flooding the literature with articles. The focus shifts from how much a journal publishes to how much each contribution actually matters to the research community.

CIA – compactness of information content 🔗

The second parameter, CIA, captures the compactness of information per article. Not all articles deliver the same amount of usable knowledge. Two papers of equal length can differ enormously in how much information they actually communicate. One may be padded with repetition and filler, while the other presents dense, well-organised findings, tables, and data in the same space.

CIA rewards journals whose articles are information-rich rather than merely long. This idea connects to wider work on quantifying the information content of scientific articles, which found enormous variation in how much real content individual papers carry. For a reader or a librarian, a journal with high information compactness offers more value for the time and money invested, because each article does more work.

The D/C ratio – scientific value against compactness 🔗

The third parameter, the D/C ratio, links demand back to compactness. It measures the scientific value of a journal’s papers in relation to how compactly information is presented. A high D/C ratio indicates that a journal’s compact, information-dense articles are also attracting genuine attention and citation. A low D/C ratio suggests the opposite: either the journal publishes a great deal that earns little notice, or its citations are spread thinly across loosely packed content.

Taken together, the three parameters form a layered picture. D/A checks whether interest is concentrated or diluted across articles. CIA checks how much knowledge each article carries. D/C checks whether that dense knowledge is genuinely valued by the field. No single number could do all three jobs.

How these parameters improve journal rankings 🔗

The real power of Sengupta’s approach lies in re-ranking. He did not throw away the original citation-based list. Instead, he applied the three new parameters to it and watched the order change. As a test case, he took the top ten core journals of biochemistry that had already been identified through citation counting, recalculated their standing using D/A, CIA, and D/C, and produced a revised ranking.

The outcome was telling. Journals that had ranked high purely on citation volume slipped when judged on interest-per-article and information density. Others that had looked ordinary rose, because their compact articles delivered strong value. This kind of movement is what bibliometricians call re-ranking, and it exposes distortions that a single metric hides. The principle echoes a broader finding in scientometrics that significant ranking parameters cause strong shifts in journal order, and that each parameter carries information the others cannot replace.

It is worth placing this within the wider family of quantitative methods. Bibliometrics, scientometrics, and informetrics all apply statistical reasoning to scholarly communication, and Sengupta’s parameters sit comfortably within this tradition of measuring science more carefully. Rather than competing with the Impact Factor, they complement it, adding dimensions of efficiency and quality that citation counting overlooks.

Implications for library management 🔗

For libraries, these parameters are not just academic curiosities. They speak directly to two of the hardest decisions any library faces: what to acquire and what to keep paying for.

Smarter acquisition and subscription decisions 🔗

Journal subscriptions consume a large share of any academic library budget, and prices keep rising while budgets rarely do. When a committee must decide between two journals in the same field, raw citation counts can mislead. A journal with a strong D/A ratio offers more interest per article, which often means better value for the subscription cost. A high CIA tells the librarian that each issue delivers dense, usable knowledge rather than thin, scattered content.

This matters because general indicators like the Impact Factor were never designed for local collection decisions. As one analysis of library journal collections noted, the referencing patterns of a whole discipline may not reflect the needs of a particular institution. Sengupta’s parameters give librarians a more nuanced toolkit to match journals to the genuine usefulness of their content, not just their headline popularity.

Building a balanced, cost-effective collection 🔗

The same logic applies in reverse during cancellation reviews. Libraries routinely build models that weigh subscription cost against actual use to decide which titles to drop, and citation statistics are a recognised basis for such deselection decisions. A journal that ranks high on citations but poorly on D/C may be a candidate for cancellation, because its content is not delivering value in proportion to what is paid for it.

Used together, the three parameters help a library construct a collection that is both economical and genuinely useful. They guide the librarian toward journals that concentrate scientific interest, pack information efficiently, and earn real attention for that content. In a context where every rupee of the acquisitions budget must be justified, that is a practical advantage, not a theoretical one.

What do you think? If you were managing a college library with a fixed budget, would you trust a single Impact Factor figure, or would you re-rank your journals using parameters like D/A, CIA, and D/C before renewing subscriptions? And do you believe information density should weigh as heavily as citation count when we judge a journal’s true worth?

References

1. https://link.springer.com/article/10.1007/BF02016772
2. https://clarivate.com/academia-government/essays/impact-factor/
3. https://osu.libguides.com/c.php?g=110226&p=714742
4. https://royalsocietypublishing.org/rsos/article/9/8/220334/96861/Citation-counts-and-journal-impact-factors-do-not
5. https://pmc.ncbi.nlm.nih.gov/articles/PMC7458102/
6. https://link.springer.com/article/10.1007/BF02016906
7. https://link.springer.com/article/10.1007/BF02019163
8. https://www.frontiersin.org/journals/research-metrics-and-analytics/articles/10.3389/frma.2021.742311/full
9. https://pmc.ncbi.nlm.nih.gov/articles/PMC4957939/
10. https://www.sciencedirect.com/science/article/abs/pii/S0740818806000260