[Paper Review] Growth rates of modern science: A latent piecewise growth curve approach to model publication numbers from established and new literature databases

Lutz Bornmann, Robin Haunschild+1 more|arXiv (Cornell University)|Dec 14, 2020

scientometrics and bibliometrics research4 citations

TL;DR

This study models the growth of modern science from the 17th century to present using publication data from four bibliographic databases—Web of Science, Scopus, Dimensions, and Microsoft Academic—via a latent piecewise growth curve approach. It reveals an overall scientific publication growth rate of 4.10% per year (doubling every 17.3 years), with five distinct growth segments best explained by major historical, economic, and political events such as industrialization and World War II.

ABSTRACT

Growth of science is a prevalent issue in science of science studies. In recent years, two new bibliographic databases have been introduced which can be used to study growth processes in science from centuries back: Dimensions from Digital Science and Microsoft Academic. In this study, we used publication data from these new databases and added publication data from two established databases (Web of Science from Clarivate Analytics and Scopus from Elsevier) to investigate scientific growth processes from the beginning of the modern science system until today. We estimated regression models that included simultaneously the publication counts from the four databases. The results of the unrestricted growth of science calculations show that the overall growth rate amounts to 4.10% with a doubling time of 17.3 years. As the comparison of various segmented regression models in the current study revealed, the model with five segments fits the publication data best. We demonstrated that these segments with different growth rates can be interpreted very well, since they are related to either phases of economic (e.g., industrialization) and / or political developments (e.g., Second World War). In this study, we additionally analyzed scientific growth in two broad fields (Physical and Technical Sciences as well as Life Sciences) and the relationship of scientific and economic growth in UK. The comparison between the two fields revealed only slight differences. The comparison of the British economic and scientific growth rates showed that the economic growth rate is slightly lower than the scientific growth rate.

Motivation & Objective

• To analyze long-term growth trends in scientific publication output from the dawn of modern science to the present.
• To assess the validity and utility of newly available bibliographic databases—Dimensions and Microsoft Academic—for historical science growth analysis.
• To identify distinct phases of scientific growth and relate them to major socio-historical developments.
• To compare scientific growth rates across broad disciplines (Physical & Technical Sciences vs. Life Sciences) and between science and economic growth in the UK.
• To develop and apply a latent piecewise growth curve model that captures non-linear growth patterns more accurately than standard exponential models.

Proposed method

• Collected publication counts from four databases: Web of Science, Scopus, Dimensions, and Microsoft Academic, covering the period 1600–2020.
• Applied a latent piecewise growth curve model to estimate segmented growth rates, allowing for distinct growth phases with different slopes.
• Used regression analysis to simultaneously model publication counts from all four databases, improving robustness and reducing bias.
• Evaluated multiple segmented models (2 to 7 segments) using fit statistics to determine the optimal number of growth phases.
• Incorporated historical markers (e.g., World War II, industrialization) to interpret the timing and causes of growth rate shifts.
• Conducted field-specific and country-level comparisons (UK) to assess differences in growth dynamics across domains and between science and economy.

Experimental results

Research questions

• RQ1 What is the overall growth rate of scientific publications from the 17th century to 2020, and what is the corresponding doubling time?
• RQ2 How many distinct growth phases exist in the history of modern science, and what explains their timing and differing rates?
• RQ3 To what extent can the observed growth segments be linked to major economic or political events such as industrialization or World War II?
• RQ4 How do growth rates in Physical and Technical Sciences compare to those in Life Sciences over the same period?
• RQ5 How does the rate of scientific publication growth in the UK compare to the concurrent rate of economic growth?

Key findings

• The overall growth rate of scientific publications from 1600 to 2020 is 4.10% per year, with a doubling time of 17.3 years.
• The five-segment piecewise growth model provided the best fit to the data, outperforming models with fewer or more segments.
• The five growth phases correspond closely to major historical developments: pre-industrialization, early industrialization, World War II, post-war expansion, and late 20th-century acceleration.
• Growth rates in Physical and Technical Sciences and Life Sciences show only slight differences, indicating broadly similar long-term publication trends across broad scientific fields.
• In the UK, the scientific publication growth rate (4.10%) slightly exceeds the concurrent economic growth rate, suggesting science is expanding faster than the economy.
• The integration of new databases (Dimensions and Microsoft Academic) with established ones (Web of Science and Scopus) significantly improves the accuracy and temporal coverage of growth modeling in science studies.