Inequality in Societies, Academic Institutions and Science Journals: Gini and k-indices

Social inequality is traditionally measured by the Gini-index ($g$). The $g$-index takes values from $0$ to $1$ where $g=0$ represents complete equality and $g=1$ represents complete inequality. Most of the estimates of the income or wealth data indicate the $g$ value to be widely dispersed across the countries of the world: \textit{g} values typically range from $0.30$ to $0.65$ at a particular time (year). We estimated similarly the Gini-index for the citations earned by the yearly publications of various academic institutions and the science journals. The ISI web of science data suggests remarkably strong inequality and universality ($g=0.70\pm0.07$) across all the universities and institutions of the world, while for the journals we find $g=0.65\pm0.15$ for any typical year. We define a new inequality measure, namely the $k$-index, saying that the cumulative income or citations of ($1-k$) fraction of people or papers exceed those earned by the fraction ($k$) of the people or publications respectively. We find, while the $k$-index value for income ranges from $0.60$ to $0.75$ for income distributions across the world, it has a value around $0.75\pm0.05$ for different universities and institutions across the world and around $0.77\pm0.10$ for the science journals. Apart from above indices, we also analyze the same institution and journal citation data by measuring Pietra index and median index.

💡 Research Summary

The paper investigates inequality in three distinct domains—household income, academic institutions, and scientific journals—by applying the classic Gini coefficient and introducing a novel metric called the k‑index. The authors begin by reviewing the Gini index, which quantifies inequality as twice the area between the Lorenz curve and the line of perfect equality, taking values from 0 (complete equality) to 1 (complete inequality). They note that worldwide income Gini values typically lie between 0.30 and 0.75, with corresponding k‑index values ranging from 0.60 to 0.75.

To extend this analysis to the scientific arena, the authors extracted citation data from the ISI Web of Science for the years 1980‑2013. They compiled yearly publication and citation counts for a selection of prominent universities and research institutes (e.g., Melbourne, Tokyo, Harvard, MIT, Cambridge, Oxford) as well as for leading journals (Nature, Science, PNAS, Cell, Physical Review series, etc.). For each year, papers were sorted by citation count, and cumulative fractions of papers (n) and citations (w) were plotted to construct Lorenz curves. The Gini coefficient for each institution or journal was then computed from the area under these curves.

The empirical results reveal markedly higher inequality in scholarly citations than in household income. Across all examined universities and institutes, the average Gini is about 0.70 ± 0.07, while for journals it is roughly 0.65 ± 0.15. These values indicate that a small fraction of papers garners a disproportionate share of citations, mirroring the “80‑20” rule observed in wealth distributions.

The k‑index is defined as the abscissa (k) at which the Lorenz curve intersects the line perpendicular to the equality line (i.e., the line n + w = 1). Conceptually, it states that the top (1 − k) fraction of papers receives more citations than the bottom k fraction. For income data, k varies between 0.60 and 0.75; for academic institutions it clusters around 0.75 ± 0.05, and for journals around 0.77 ± 0.10. The authors note an approximate relationship g ≈ 2k − 1, which holds reasonably well for moderate inequality but deviates for larger Gini values.

Beyond the k‑index, the authors examine the tail behavior of citation distributions. By plotting 1 − w versus 1 − n on log‑log scales, they find that for n ≥ k the data follow a Pareto‑type power law: 1 − w ∼ (1 − n)^α with α ≈ 0.5 ± 0.1. This suggests that the region beyond the k‑index marks a transition to a regime where a few highly cited papers dominate the citation pool, analogous to wealth concentration in economies.

To provide a broader perspective on inequality measures, the paper also computes the Pietra index (p‑index), defined as the maximum vertical distance between the Lorenz curve and the equality line, and the median index (m‑index), derived from the point where the Lorenz curve reaches w = 0.5 (expressed as 2m − 1). Both indices range between 0.4 and 0.9 across the datasets and show consistent patterns with the Gini and k‑indices, reinforcing the robustness of the findings.

The discussion emphasizes that the k‑index offers an intuitive, h‑index‑like interpretation for institutions and journals: it quantifies the fraction of output that accounts for the majority of impact. This makes it a potentially valuable tool for research evaluation, funding allocation, and policy design aimed at mitigating excessive concentration of scientific influence. Moreover, identifying the Pareto tail beyond k provides a quantitative basis for modeling citation dynamics and for understanding how scientific prestige propagates.

In conclusion, while the Gini coefficient remains a fundamental measure of inequality, the k‑index complements it by delivering a more direct, interpretable metric for scholarly output. The authors suggest future work should track temporal evolution of k, explore disciplinary differences, and investigate policy implications of high citation concentration in academia.