Revisiting Lotka’s law: patterns of author productivity in sustainability science

Poster (konferens), 2026

On the centenary of Alfred J. Lotka's seminal paper on the frequency distribution of scientific productivity, we revisit the classical inverse-square law using a ten-year (2016 – 2025) Scopus dataset of sustainability science journals. The dataset comprises 143,183 publications, yielding 707,094 authorship records from 398,085 unique authors, forming a large-scale, co-authored corpus well suited to examining productivity patterns through the lens of Lotka's law. We analyze author productivity distributions for heavy-tailed data, including maximum-likelihood estimation, goodness-of-fit testing, and model comparison. The estimated scaling exponent is substantially higher than Lotka’s classical value, and statistical tests reject a pure power-law specification in favor of a lognormal model. The findings reveal strong concentration of productivity among a small group of authors while showing departure from the classical inverse-square exponent in the upper tail. A better-fitting lognormal model does not invalidate Lotka’s law; rather, it refines its application to modern, multi-author research corpora.