The BiPublishers ranking: Main results and methodological problems when constructing rankings of academic publishers

We present the results of the Bibliometric Indicators for Publishers project (also known as BiPublishers). This project represents the first attempt to systematically develop bibliometric publisher rankings. The data for this project was derived from the Book Citation Index, and the study time period was 2009-2013. We have developed 42 rankings: 4 for by fields and 38 by disciplines. We display six indicators by publisher divided into three types: output, impact and publisher’s profile. The aim is to capture different characteristics of the research performance of publishers. 254 publishers were processed and classified according to publisher type: commercial publishers and university presses. We present the main publishers by fields. Then, we discuss the main challenges presented when developing this type of tools. The BiPublishers ranking is an on-going project which aims to develop and explore new data sources and indicators to better capture and define the research impact of publishers.

💡 Research Summary

The paper introduces the Bibliometric Indicators for Publishers (BiPublishers) project, which represents the first systematic attempt to construct bibliometric rankings of academic book publishers. Using data extracted from the Book Citation Index (BCI) for the period 2009‑2013, the authors processed 254 publishers and produced 42 distinct rankings: four broad field‑level rankings and thirty‑eight discipline‑specific rankings. Each publisher is evaluated on six indicators grouped into three conceptual dimensions—output, impact, and publisher profile. Output is measured by total titles published and average annual titles; impact is captured by total citations received and citations per title; profile is expressed through disciplinary diversity (using Shannon entropy) and the proportion of commercial versus university presses.

The results show that a handful of large commercial houses (e.g., Springer, Elsevier, Wiley) dominate the output and citation metrics across most fields, while university presses such as Oxford University Press and Cambridge University Press score highly on disciplinary breadth and citation efficiency. Field‑by‑field analyses reveal that in the humanities and social sciences, commercial publishers lead in volume but university presses exhibit higher impact per title. In the sciences and engineering, commercial publishers again dominate both output and impact, whereas in medicine and health sciences a mix of commercial and university presses appears near the top.

Beyond presenting these descriptive findings, the authors devote substantial discussion to methodological challenges inherent in building such rankings. First, the reliance on BCI introduces a language and coverage bias: the index heavily favors English‑language books, causing non‑English publishers to be under‑represented. Second, book citation behavior differs markedly from journal citation behavior; books have longer citation windows, multiple editions, and may be cited by non‑scholarly works, which can distort raw citation counts. Third, the binary classification of publishers into “commercial” and “university” presses oversimplifies the publishing landscape, ignoring hybrid models such as commercial firms that operate university imprints or non‑profit research institutes that run small presses. Fourth, the six indicators are not independent—output and impact measures are highly correlated, raising concerns about redundancy and the need for multicollinearity checks. Fifth, the weighting scheme (equal weights across dimensions) is arbitrary; different disciplines may value impact over output or vice versa, suggesting that field‑specific weighting or optimization techniques would improve validity.

The paper also highlights classification issues: the BCI’s subject categories are applied directly, yet many monographs are interdisciplinary, leading to ambiguous field assignments. The authors propose incorporating text‑mining and machine‑learning classification to achieve finer granularity. They recommend augmenting BCI data with additional sources such as Scopus Books, Google Scholar, library circulation statistics, and publisher‑provided metadata to mitigate coverage gaps and to capture alternative impact signals (e.g., downloads, sales, altmetrics).

In conclusion, the BiPublishers project provides a valuable proof‑of‑concept for publisher‑level bibliometrics, offering a transparent set of indicators and a publicly available ranking framework. However, the study underscores that robust publisher rankings require careful handling of data quality, citation dynamics, classification schemes, and indicator design. Future work should pursue multi‑source data integration, refined disciplinary mapping, and mixed‑methods validation (including expert surveys) to develop more reliable and nuanced assessments of academic publishers’ research contributions.