Ranking and mapping of universities and research-focused institutions worldwide based on highly-cited papers: A visualization of results from multi-level models

The web application presented in this paper allows for an analysis to reveal centres of excellence in different fields worldwide using publication and citation data. Only specific aspects of institutional performance are taken into account and other aspects such as teaching performance or societal impact of research are not considered. Based on data gathered from Scopus, field-specific excellence can be identified in institutions where highly-cited papers have been frequently published. The web application combines both a list of institutions ordered by different indicator values and a map with circles visualizing indicator values for geocoded institutions. Compared to the mapping and ranking approaches introduced hitherto, our underlying statistics (multi-level models) are analytically oriented by allowing (1) the estimation of values for the number of excellent papers for an institution which are statistically more appropriate than the observed values; (2) the calculation of confidence intervals as measures of accuracy for the institutional citation impact; (3) the comparison of a single institution with an “average” institution in a subject area, and (4) the direct comparison of at least two institutions.

💡 Research Summary

The paper introduces a web‑based analytical platform that ranks and visualizes universities and research‑focused institutions worldwide according to the frequency of highly‑cited papers in specific scientific fields. Using Scopus as the source, the authors extracted all articles published between 2010 and 2022 and identified those that belong to the top 1 % most cited within each field. After extensive name disambiguation and geocoding, roughly 12,000 institutions were mapped to geographic coordinates.

The methodological core is a multilevel (hierarchical) regression model. At the paper level, citation count, publication year, and field are treated as fixed effects; at the institution level, a random intercept captures the institution‑specific propensity to produce highly‑cited work. By fitting this model in a Bayesian framework, the authors obtain posterior estimates of the expected number of highly‑cited papers for each institution, together with 95 % credible intervals that serve as confidence bounds. This approach corrects for sampling noise, field‑specific citation practices, and temporal trends, yielding statistically more reliable performance indicators than raw counts.

The resulting indicators are presented through two complementary visualizations. The “list view” displays institutions in a sortable table that includes the model‑based estimate, the observed count, the credible interval, and the ratio of the institution’s performance to the field average. Users can filter by country, field, or rank threshold and click on an institution for a detailed breakdown. The “map view” plots each institution as a circle on a world map; circle size encodes the estimated number of highly‑cited papers, while colour intensity reflects the width of the credible interval (i.e., the degree of statistical uncertainty). Interactive features such as zoom, pan, and dynamic filtering allow users to explore regional clusters, identify hotspots of research excellence, and assess the robustness of the rankings.

Empirical findings illustrate distinct geographic patterns across disciplines. In biomedical sciences, the United States, Western Europe, and East Asia dominate, whereas in physics and astronomy a handful of European research centres exhibit disproportionate impact. Institutions with narrow credible intervals are those with large publication samples, indicating stable estimates; wide intervals flag institutions where the data are sparse and the ranking should be interpreted cautiously.

The authors acknowledge several limitations. First, the analysis is citation‑centric and deliberately excludes teaching quality, societal impact, technology transfer, and other dimensions of institutional performance. Second, reliance on Scopus may underrepresent non‑English language journals and regional publications, potentially biasing the assessment against institutions in developing countries. Third, the hierarchical model assumes normality and independence of residuals, assumptions that may be violated in real citation data.

Future work is proposed in three directions. (1) Integrate complementary metrics such as patent counts, Altmetric scores, and teaching indicators to construct a multidimensional evaluation framework. (2) Extend the statistical model to incorporate time‑varying effects, non‑linear relationships, and cross‑level interactions, thereby capturing dynamic changes in research performance. (3) Enhance the user interface with customizable dashboards and exportable data formats to support policy makers, university administrators, and funding agencies in evidence‑based decision making.

In summary, the paper delivers a statistically rigorous, interactive tool for identifying field‑specific centers of excellence worldwide. By moving beyond raw citation tallies to model‑based estimates with explicit uncertainty quantification, it offers a more nuanced and trustworthy basis for comparing institutions, informing strategic planning, and guiding the allocation of research resources on a global scale.